Knock Detection Method Research Articles

The sensitivity of pressure-based knock threshold values to alternative fuels: A comparison of methanol vs. gasoline

The heavy-duty industry typically uses a high cetane fuel to power their vehicles, currently typically fossil diesel. Switching to renewable fuels such as methanol requires a major changeover in engine technology. Due to the high octane number of methanol, the diesel engines will have to be replaced by more suitable spark-ignited engines. An accurate knock control system will then be needed to accommodate the high power density requirements of this sector. However, current pressure-based knock detection methods were primarily developed for gasoline fuel operation. Blindly copying these methods for a fuel which has vastly different knock properties can lead to suboptimal engine operation with an efficiency and power loss as result. In this work, experiments were conducted on a single cylinder port fuel injected spark-ignited CFR engine. The applicability of five different pressure-based knock metrics was reassessed for methanol fuel and their performance was compared to each other. Knock threshold values were established for both gasoline and methanol for the MAPO (maximum amplitude of pressure oscillations), AE (average energy) and MVTD (minimum value of third derivative) methods. The results show that the performance of the single peak metrics MAPO and IMPO (integral of modulus of pressure oscillations), and the time-averaged metrics AE and HRR (heat release rate) remains identical between methanol and gasoline operation. For the MVTD method however, the suggested low sampling rate of 1 sample/°ca is insufficient for methanol combustion since the fuel by nature already shows a narrow pressure peak. Overall, the HRR method outperforms the others through the addition of a noise constraint. The calculated knock threshold values, however, are drastically different between both fuels. Compared to gasoline, the thresholds for methanol were 1.25, 2.40 and 5.45 times higher for the MAPO, AE and MVTD methods respectively. This highlights the need for dedicated, fuel-specific, knock threshold values.

Knock recognition of knock sensor signal based on wavelet transform and variational mode decomposition algorithm

Natural gas has become the most widely used alternative energy for engines because of its low emission pollution and sufficient energy reserve. In order to improve the thermal efficiency of the natural gas engine, it is necessary to increase the compression ratio, but knock restricts the increase of the engine compression ratio. The main knock detection method of engine is to use knock sensor. However, the body vibration detected by the knock sensor contains a lot of interference noise, which brings bad effect on the accuracy of knock detection. Thus, in this paper, a knock recognition method based on wavelet transform (WT) and variational mode decomposition (VMD) algorithm is proposed to process the knock sensor signal and extract the knock feature. Based on the test bench of natural gas engine, experimental investigation of knock combustion characteristics was conducted by adjusting the spark timing, and simultaneously collects cylinder pressure signal and knock sensor signal. Burg algorithm based on Auto-regressive (AR) model was used to extract the knock characteristic frequency of the research engine. Then, wavelet transform and variational mode decomposition algorithm were employed to process the knock sensor signal to remove the interference noise and extract the knock feature. Maximum amplitude of pressure oscillation (MAPO) was selected as the knock intensity evaluation index of cylinder pressure signal. Then, taking the evaluation result of the knock intensity of cylinder pressure signal as reference, the root-mean-square of the knock characteristic signal was used as knock intensity evaluation index of knock sensor signal to determine the knock intensity classification of the knock sensor signal. The results show that the proposed method can effectively extract the knock feature from the knock sensor signal and realize the accurate engine knock recognition. The recognition accuracy is higher than the most existing knock recognition methods.

An unsupervised machine learning technique to identify knock from a knock signal time-frequency analysis

Knock is a characteristic phenomenon of spark ignition engines that can result in lower fuel efficiency and higher combustion temperatures. When it is not adequately identified and controlled, knock might lead to intensive engine damage. Knock detection methods are based on recognising the large in-cylinder pressure oscillations that resonate in the combustion chamber as a consequence of the rapid auto-ignition of the end-gas. However, in-cylinder pressure sensors may be invasive and expensive. In this work, a method for knock classification is proposed based on the knock sensor measurement, which is a low-cost and easy-to-implement vibration sensor. Unfortunately, the vibration of the engine block contains information from several phenomena, and only intensive knock can be identified by classical methods. To counteract the amount of noise, machine learning (ML) techniques are applied to the knock sensor signal. A training database composed of several operating conditions is used to extract the main characteristics of knock. The signal is analysed in the time-frequency domain by using the short-time Fourier transform (STFT), the features of the signal are obtained by singular value decomposition (SVD), and finally, support vector machine (SVM) techniques are used to identify outliers. Experimental data from a spark-ignited engine at 2000 rpm and 3000 rpm with spark advance steps was used to train and validate the method. Results are compared to state-of-the-art knock recognition methods based on pressure and knock sensors. The technique was able to detect 73% of the knock cycles, which was similar to the results achieved using the pressure sensor data (77%). Compared to another method using the knock sensor signal as well, the percentage of false negative cases was reduced by 20%.

Knock probability determination in a turbocharged gasoline engine through exhaust gas temperature and artificial neural network

In the present study, a new combination of different available engine sensors along with an exhaust gas temperature sensor is used as the input of an artificial neural network to recognize knock incidents probability. In this regard, comprehensive experimental test trips on a turbocharged gasoline-powered vehicle are defined, and about 5 million data sets are extracted. These data are filtered, and then the optimization process is conducted on the network to minimize errors in the presented network. The output of the presented network would be the probability of knock occurrence in different operating conditions which a probability of more than 50% would be recognized as a knock occurrence. The results reveal that the neural network can detect knocking by traditional engine sensors plus exhaust gas temperature sensors with 92% precision. The exhaust gas temperature increases in most knock occurrences due to spark time retarding. The network reveals the little effect of the air temperature sensor on knock incidence, while the strong correlation of intake manifold pressure on the knock occurrence is shown. The spark timing, engine speed, intake temperature, and pressure sensor have the most considerable effect on knock detection, while the camshaft angle sensor has an almost negligible effect on knock detection. The current network promotes knock detection methods by introducing the knock probability, and additional improvement on the presented method can lead to a reliable procedure for knock detection.

Engine Knock Detection Methods for Spark Ignition and Prechamber Combustion Systems in a High-Performance Gasoline Direct Injection Engine

Engine Knock Detection Methods for Spark Ignition and Prechamber Combustion Systems in a High-Performance Gasoline Direct Injection Engine

A Novel Multiple Feature-Based Engine Knock Detection System using Sparse Bayesian Extreme Learning Machine

Automotive engine knock is an abnormal combustion phenomenon that affects engine performance and lifetime expectancy, but it is difficult to detect. Collecting engine vibration signals from an engine cylinder block is an effective way to detect engine knock. This paper proposes an intelligent engine knock detection system based on engine vibration signals. First, filtered signals are obtained by utilizing variational mode decomposition (VMD), which decomposes the original time domain signals into a series of intrinsic mode functions (IMFs). Moreover, the values of the balancing parameter and the number of IMF modes are optimized using genetic algorithm (GA). IMFs with sample entropy higher than the mean are then selected as sensitive subcomponents for signal reconstruction and subsequently removed. A multiple feature learning approach that considers time domain statistical analysis (TDSA), multi-fractal detrended fluctuation analysis (MFDFA) and alpha stable distribution (ASD) simultaneously, is utilized to extract features from the denoised signals. Finally, the extracted features are trained by sparse Bayesian extreme learning machine (SBELM) to overcome the sensitivity of hyperparameters in conventional machine learning algorithms. A test rig is designed to collect the raw engine data. Compared with other technology combinations, the optimal scheme from signal processing to feature classification is obtained, and the classification accuracy of the proposed integrated engine knock detection method can achieve 98.27%. We successfully propose and test a universal intelligence solution for the detection task.

Evaluation of ASTM D6424 standard for knock analysis using unleaded fuel candidates on a six cylinder aircraft engine

The ASTM D6424 standard is used for general aviation piston engine knock detection and is tested for unleaded fuel candidates. Issues are discussed regarding the identification of knocking cycles, filtering frequency bands, and the effects of down-sampling for this knock detection technique. The knock tests were performed on the Continental TSIO-520-VB engine at 12,000 ft for take-off and cruise conditions using three different fuels, the standard leaded 100LL avgas and two unleaded fuel candidates. The ASTM D6424 knock detection method has its own particular disadvantages, which are detailed and compared to other knock detection methods including the third derivative of pressure signal and discrete wavelet transform. Updates to the standard include a minimum sampling rate of 0.2 CAD. Additionally, the current standard does not contain recommendations for filtering the cylinder pressure which results in over detection of knocking cycles with the two new aviation fuel candidates tested. Recommendations are provided regarding the pressure signal processing prior to ASTM D6424 knock-characterization.

Improved knock detection method for lean burn SI engines

The purpose of this study was to improve the accuracy of knock detection for lean burn SI engines. In this study, we conducted an experiment to observe autoignition in an engine cylinder using a borescope and a high-speed camera under an air excess ratio of 2.1. In addition, the resonance frequency of each vibration mode by knocking and the knock intensity were also calculated from the in-cylinder pressure history. Autoignition was optically observed in all cycles with a knock intensity of more than 20 kPa or more, and in some cycles with that of less than 20 kPa. We compared the vibration caused by combustion and it caused by autoignition. As a result, by distinguishing between the vibration caused by combustion and it caused by autoignition using window functions, a knocking cycle that could not be detected by the conventional index was detected.

Study on joint detection of fabricated concrete shear wall based on percussion method

The knock detection method is one of the non-destructive testing methods for assembly prefabricated components. The knocker is used to knock on the concrete prefabricated components to be tested. The internal conditions of the tested components can be judged by analyzing the time-domain and frequency-domain diagrams of the knocker. In this paper, the impact of the change of the knock point on the detection is studied by using the knock detection experiment. It is found that the location of the knock point has a great influence on the detection. With the change of the position of the knock point, the peak frequency in the spectrum obtained by the detection changes correspondingly. According to the frequency information corresponding to the peak value in the spectrum, the depth of the void to the knock surface can be calculated.

Engine knock detection: an eigenpressure approach

Abstract In this work, a knock detection approach based on in-cylinder pressure principal component analysis is proposed. The introduction of a set of basis functions called eigenpressures used to describe the principal components of the pressure traces allows for an easy and effective separation between the typical “bell shape” component of pressure profiles and the knock-induced pressure oscillations, making possible the classification of knocking and not knocking cycles. The proposed approach is compared to a standard knock detection method based on the in-cylinder pressure trace band-pass filtering and to a pure data-driven algorithm. The method shows the best knock classification performances and proves to be advantageous thanks to the low number of easily tunable parameters and their ease of calibration/interpretation.

Improved Knock Detection Method Based on New Time-Frequency Analysis In Spark Ignition Turbocharged Engine

Improved Knock Detection Method Based on New Time-Frequency Analysis In Spark Ignition Turbocharged Engine

Development of a novel knock characteristic detection method for gasoline engines based on wavelet-denoising and EMD decomposition

Abstract The main contribution of this paper is inventing a novel knock characteristic extraction method using vibration signals. Engine knock reduces the thermal efficiency and restricts performance improvement of a gasoline engine. Reliable and rapid detection of engine knock characteristics (including light knock) is key for the engine knock control. Based on the wavelet-denoising and empirical mode decomposition (EMD) methods and the analysis results, this paper discovers that the wavelet-denoising can eliminate the high frequency noise that was not generated by the engine knock. After the wavelet-denoising, EMD decomposition can effectively identify the knock characteristics (include the light knock) from a vibration signal. Compared to existing knock detection methods, the new method in this paper can reduce the computational time while ensuring the reliability of the knock detection.

Normalized Knock Intensity Determination Based on the Knock Sensor Analysis to Have a Fixed Detection Threshold at Different Operating Conditions

Processing the knock sensor's signal is the most common approach for knock detection in series production vehicles. Filtration, rectification, and integration in a defined knock window (KW) are main steps to compute the standard knock intensity (SKI). The SKI strongly depends on the engine operating conditions. In this study, a novel model is proposed based on the knock sensor analysis to determine the normalized knock intensity (NKI) with much less dependency on the operating conditions, cylinder numbers (CNs), and KW. Implementing the proposed normalization model, a fixed detection threshold can be used for knock detection at all operating conditions. To verify the model, an accurate knock detection method based on cylinder pressure analysis is utilized, which comprises intensity calculation and a novel technique for detection threshold determination. Experimental results at all operating conditions show a square of correlation coefficient greater than 0.7 when the knock intensity from the presented model is compared with the reference cylinder pressure based method. In addition, the model detects all heavy knocking cycles and there is no wrongly detected knocking combustion.

Dynamic knock detection and quantification in a spark ignition engine by means of a pressure based method

In spark ignition engines, knock onset limits the maximum spark advance. An inaccurate identification of this limit penalises the fuel conversion efficiency. Thus it is very important to define a knock detection method able to assess the knock intensity of an engine operating point.Usually, in engine development, knock event is evaluated by analysing the in-cylinder pressure trace. Data are filtered and processed in order to obtain some indices correlated to the knock intensity, then the calculated value is compared to a predetermined threshold. The calibration of this threshold is complex and difficult; statistical approach should be used, but often empirical values are considered.In this paper a method that dynamically calculates the knock threshold necessary to determine the knock event is proposed. The purpose is to resolve cycle by cycle the knock intensity related to an individual engine cycle without setting a predetermined threshold.The method has been applied to an extensive set of experimental data relative to a gasoline spark-ignition engine.Results are correlated to those obtained considering a traditional method, where a statistical approach has been used to detect knock.

Robustness investigation of a SVM-based knock detection method

Knock detection on the basis of structure-borne noise signals that are processed through a sequence of filtering, rectifying, and integrating has been performed in series for many years. It is implemented in Bosch engine management systems that are used by automotive manufactures all over the world. The trend towards higher power density, lower fuel consumption and legal limitation of emissions calls for increasingly complex engines, thereby diversifying the requirements for reliable knock detection.

Estimation of knock intensity in Spark-Ignition engines by using wavelet transform

Knock in spark ignition (SI) engines is an undesirable mode of combustion, causing high combustion pressure pulses associated with vibration of the engine block. This phenomenon limits performance, durability and fuel economy of SIengines. In this paper the author propose; a new knock detection method called WBKD (Wavelet Based Knock Detection), which is based on wavelet transform. Real data experiments confirmed the utility of the proposed approach for knock intensity estimating.