Knock Events Research Articles

Biofuel Concentration in Low-Speed Pre-Ignition in Gasoline Engines

Low-Speed Pre-Ignition (LSPI) is a destructive combustion event associated primarily with new, ultra-efficient, downsized gasoline engines, which provide efficiency benefits in general operation. Biofuels, specifically bio-gasoline, are an alternative fuel that attempts to reduce the harmful emissions output by modern Internal Combustion Engines (ICEs). This study attempts to understand the effect of biofuel use on LSPI, through the use of a numerical simulation tool developed in Ricardo Wave. Development of the tool includes the integration of RFlame, an extension capable of modeling autoignition within a 1D domain. Use of the tool highlights the impact of five ethanol blends, E10, E20, E30, E50 and E85, with clear impacts on both the severity and frequency of LSPI events correlated with chemical properties, such as the enthalpy of vaporization (HoV) and octane number. E30 is highlighted as the critical blend for LSPI severity, with both increased severity and intensity seen with a 30% concentration, and a greater sensitivity to effects such as the start of ignition (SOI). Higher-concentration biofuels, such as E50 and E85 bio-gasoline, show much more favorable behaviors, such as a vast reduction in end-gas knock events, but are limited in their use due to their deployment being both cost-prohibitive and potentially damaging in current hardware. Future work on this topic will surround the further development of the simulation tool to integrate 3D solving elements, understand the role of fluid interactions in LSPI, and study optimal fuel characteristics for future use in ICEs.

Experimental study of spatial distribution of knock events in a turbocharged spark-ignition engine

An experimental investigation on the spatial distribution of knock events in a turbocharged spark-ignition engine for hybrid vehicle applications was conducted by using a multichannel fiber optic method. The knock positions were detected under different conditions to investigate the influence of crucial engine design and operating parameters on the knock characteristics including the spatial distribution in the combustion chamber and its relationship to knock intensity. The measured data reveal that the spatial distribution in the engine with a port fuel injection (PFI) system is mainly located on the exhaust side with insignificant influence of engine speed and load, which is attributed to the elevated thermal load around the exhaust valves. However, the knock events under gasoline direct-injection (DI) conditions were found to occur in more scattered locations with more occurring on the engine front-end and rear-end sides. These results indicate that the in-cylinder fuel-air mixing process may have a significant impact on the knock occurrence spots under DI conditions. The knock positions of the engine with different excessive air ratios, injection timings, and intake-valve timings were also detected, indicating that engine operating parameters have complex influences on the knock-region distribution in a DI engine. In addition, experiments were also carried out in two different cylinders to verify the cylinder-to-cylinder variations in knock regions which may be caused by the engine cooling design. Furthermore, no apparent correlations were observed between the knock position and the knock intensity by analysis of the experimental data.

A critical analysis of classical severity-based knock control strategies

A common variant of classical knock control strategies replaces fixed magnitude control moves with control actions that depend on the knock severity of the prior cycle. Though appealing to notions of proportional control, it is unclear whether such strategies offer benefit given the random nature of knock events. In this paper a new generalized event-based framework is developed and used to extend both time domain simulation and stochastic analysis methods for analyzing the closed loop behavior of such systems. These tools are then used to perform a rigorous analysis and comparison between standard advance-retard and severity-based control strategies. The analysis is also repeated using a lower knock threshold definition. The results show that the performance of severity-based knock control differs minimally from a similarly tuned standard advance-retard controller, and there is little benefit to recompense the additional complexity of severity-based control.

Fiber-optical analysis of weak knock and pressure oscillations in a single cylinder research engine

The effect of characteristic cycle-to-cycle variations on knock in gasoline engines is not yet fully understood. The analysis requires the use of experimental techniques to characterize weak knock events on a cycle-individual basis. To this end, a novel processing method is presented that focuses on characterizing weak knock events using data from a fiber optic spark plug linked to a high-resolution (400 kHz) photo-multiplier array. The method was applied to a single cylinder research engine. By combining signal and image processing and reconstruction methods, it is shown that reliable automated detection of spatial knock origin is possible for Knock-Peak-Peak pressure values as low as 0.7 bar. Furthermore, the fiber optic spark plug data are used to accurately localize and characterize the pressure oscillations induced by auto-ignition which are shown to be closely related to the number, intensity, and relative location of individual knock events.

The real-time interaction model for transient mode investigations of a dual-piston free-piston engine generator

• A real-time interaction model of a free-piston engine generator is presented. • The study focuses on three critical transient modes. • The model predicts the transient performance at four transient speeds. • Predictive stroke control is incorporated to mitigate piston motion deviations. • Knock event detection provides realistic transient operation simulation. A free-piston engine generator is being developed worldwide as a novel means of electrical power generator for light vehicle application, which can provide a complementary solution towards reducing carbon dioxide emissions of road vehicles in general. However, the absence of a crank-slider mechanism in free-piston engines resulted in poor transient operations. Transient real-time modelling and simulation can provide the interaction between in-cylinder combustion with dynamics of piston motion, which has not been fully explored previously. This paper presents a real-time interaction model between dynamics of the piston motion and thermodynamics of the in-cylinder combustion of a two-stroke spark ignition dual-piston free-piston engine generator for transient operation investigations. The simulation model was developed based on the working prototype and comprised of zero- and one-dimensional sub-models interacted in real-time governed by a single timestep. The study focuses on three critical transient modes: motoring, starting and generating for the transient performance investigation. A series of experimental results during motoring was used for validating the simulation model, which showed good agreement between simulation and experiment results with 2–5% errors. The targeted brake thermal efficiency is around 20–30% at 50–60 Hz engine speed which has been shown achievable. Transient speeds of 10 and 25 Hz produced higher combustion pressure around the top dead centre but suffered pumping loss at the end of the expansion stroke. The lower peak pressures at 50 and 60 Hz have contributed to the lower brake mean effective pressure values. Maximum brake thermal efficiency of 25.8% lies in the mid-range of compression ratios between 10:1 to 15:1 while at the cyclic speeds of 50 Hz to 60 Hz. Maximum brake mean effective pressure contour occurs between 5 and 12.5 compression ratios and corresponding speeds of 10 to 30 Hz. It was observed that the excess energy from combustion results in piston overshoot condition while insufficient energy prevents the piston from achieving the targeted stroke and compression ratio. Knock events were observed for high compression ratio cases regardless of engine speed. Predictive stroke control and knock detection capability are the main contribution of the real-time interaction model presented in this paper for realistic transient operation investigation and performance prediction of the dual-piston free-piston engine generator.

Effects of multiple spark ignition on engine knock under different compression ratio and fuel octane number conditions

Engine knock has long been one of the major obstacles for improving thermal efficiency of spark-ignition (SI) engines. An in-depth understanding of the engine knock mechanism and characteristics is vital for controlling knocking combustion. Experimental investigation of knock events is challenging given their stochastic nature. We employ a single-cylinder research engine equipped with a specialized metal liner with four circumferentially mounted spark plugs to generate multiple ignition sites, and achieve more controlled knock events. Six pressure transducers are mounted to collect the pressure signals from different locations of the cylinder. A series of spark strategies (e.g., spark number, location and timing) are applied to investigate the knock characteristics of different spark ignition strategies. The effects of compression ratio and fuel octane number are also explored. The experimental results show that the knock intensity first increases as the number of active sparks goes from one to three and decreases significantly with four spark ignition, and even below the double spark ignition in some cases. This is due to the trade-off between the mass fraction and temperature of end-gas: nearly 90% of the fuel energy is consumed at knock onset in the four spark ignition cases, and only a small proportion of energy is consumed by auto-ignition, thus limiting the knock intensity. Compared with the single spark case, multiple spark ignition generates higher power output and lower cycle-to-cycle variations. The knock suppressing effect of the four spark ignition strategy is enhanced by higher fuel octane number and lower compression ratio. This study provides a possible way to generate controllable knock and gives insights into the different knock mechanisms under multiple spark ignition conditions.

Analysis of Knocking Characteristic in Dual Fuel Engine - the Effects on Diethyl Ether

In Diesel engines, where fuel is pumped into highly compressed air towards the end of the compression cycle, knocking is more or less unavoidable. By this time there is already a quantity of fuel in the combustion chamber which will first burn in areas of higher oxygen density before the full charge is combusted. The sudden rise in pressure and temperature produces the distinctive 'knock' or 'clatter' diesel, some of which must be allowed in engine design. The aim of knock control strategies is to try to maximize the trade-off between protecting the engine from damaging knock incidents, and optimizing the output torque of the engine. Knock events are a random process and independent. Knock controllers can't be programmed in a deterministic model. Due to the random nature of arriving knock events, a single time history simulation or experiment of knock control methods cannot provide a repeatable measurement of the controller efficiency. The desired trade-off must therefore be achieved in a stochastic context that could provide an appropriate environment for designing and evaluating the output of various knock control strategies with rigorous statistical properties. Clutching characteristics of a dual fuel diesel engine with direct injection of diesel and a liquid petroleum product in dual fuel mode. The engine is tested for knock reduction by adding Diethyl ether in to the diesel along with Liquid petroleum product. Variation of knocking was plotted with respect to different parameters and the result booted as knocking is minimized by the addition of diethyl ether.

Beta distribution-based knock probability map learning and spark timing control for SI engines

In gasoline engines, the spark timing is often advanced to increase fuel economy under certain heavy load engine operating conditions. As a compromise between the risk of knock and the power output, spark timing is regulated at the boundary where a low knock probability is tolerated. Due to the stochasticity of binary knock events, it is necessary to have a large number of engine cycles for probability estimations, which can slow down the response speed of a controller to operating condition changes. To speed up the spark timing regulation and to reduce the spark timing variance, in this article, a knock probability feedforward map learning method and a spark timing control method are proposed under a unified framework. A learning method that applies the beta distribution is the key contribution of this work. The beta distribution in the map learning part is used to describe knock probabilities with uncertainties and to determine the next engine operating condition for sampling and map learning. In the spark timing method, the beta distribution is applied in the conventional control method to adjust the control gains. The proposed methods are experimentally validated on a test bench equipped with a production Toyota 1.8 L, 4-cylinder SI engine.

Advanced Intra-Cycle Detection of Pre-Ignition Events through Phase-Space Transforms of Cylinder Pressure Data

<div class="section abstract"><div class="htmlview paragraph">The widespread adoption of boosted, downsized SI engines has brought pre-ignition phenomena into greater focus, as the knock events resulting from pre-ignitions can cause significant hardware damage. Much attention has been given to understanding the causes of pre-ignition and identify lubricant or fuel properties and engine design and calibration considerations that impact its frequency. This helps to shift the pre-ignition limit to higher specific loads and allow further downsizing but does not fundamentally eliminate the problem. Real-time detection and mitigation of pre-ignition would thus be desirable to allow safe engine operation in pre-ignition-prone conditions. This study focuses on advancing the time of detection of pre-ignition in an engine cycle where it occurs. Phase space transforms through time-delay embedding of cylinder pressure and principal component analysis were applied to same-cycle detection of pre-ignition and shown to enable detection on the order of a crank degree earlier than deviation in cylinder pressure can be identified through direct statistical observation of the pressure data. Additionally, it appears that the deviation of the trajectory in phase space may offer the opportunity to extend this method to further extend the detection window and allow more time for mitigation actions to occur.</div></div>

Beta-Distribution-Based Knock Probability Estimation, Control Scheme, and Experimental Validation for SI Engines

The fuel efficiency and power output of spark ignition (SI) engines are closely related to the spark timing. Advancing the spark timing is usually used as an approach to increase the efficiency. However, under some operating conditions, advanced spark timing can trigger abnormal combustion, which causes knocking. To avoid cylinder damage and to increase the engine efficiency, feedback control, which addresses the knocking phenomenon as a stochastic process, is required. In this brief, a Bayesian estimate of knock probability is used to replace the maximum likelihood estimate in a likelihood-ratio-based knock control strategy. The beta distribution is used to represent the distribution of the knock probability estimate based on the independent and identically distributed property of knock events. The proposed control algorithm is validated on a full-scale test bench with a production SI engine and is compared with the conventional spark advance control approach and the maximum-likelihood-based approach. The results show that the proposed approach is able to control and maintain a knock probability close to the target and introduce a low dispersion of spark timing after convergence.

Stochastic modeling and prediction of IMEP for closed loop knock control performance assessment

Knock control systems aim to maximize power output while also protecting the engine from knock-induced damage. The extent to which this goal is achieved is rarely evaluated, both because the random nature of knock events means that the results of any single experiment are not repeatable, and because control loop performance is generally assessed with respect to the resultant spark angle (the manipulated variable), rather than some measure of output performance. In this work, a new stochastic model for the empirical IMEP distribution is developed and fitted to open-loop experimental data. Stochastic methods are then applied to predict both the steady state and transient closed loop distribution of IMEP based on the open-loop model and the control law that is applied. The results are validated against closed loop data gathered from the engine and show good agreement with the data.

On-board knock probability map learning–based spark advance control for combustion engines

This article presents an on-board map learning–based spark advance control framework for combustion engines. The proposed control framework addresses the knock probabilistic constrained thermal efficiency optimization problem with three layers. First, in the upper layer, maps of knock event distribution and thermal efficiency are learned with manifold pressure and combustion phase as inputs. Second, the middle layer generates the knock probability constrained optimal combustion phase reference that is subsequently tracked by a hypothesis test-based feedback controller. Third, the lower layer employs a partial likelihood-based knock controller that retards the spark advance in case of the frequent knock events. The key contributions of this work are the three-layer control framework and the knock event distribution map learning in the upper layer. The knock event is supposed to obey binomial distribution, and the distribution is modeled by beta distribution and learned in the perspective of Bayesian learning. Moreover, the normalization algorithm is proposed for online feedfoward map update. The proposed map learning–based spark advance control framework is experimentally validated in a test bench equipped with a spark-ignition engine.

Computing the closed-loop characteristics of a generalized multi-threshold knock controller

Most knock controllers respond to knock events which are defined according to some threshold knock intensity. Multi-threshold knock events offer more informative feedback since they encode not just the occurrence of knock events but also some measure of their intensity. While this has the potential for improved control, it is hard to assess the extent to which any benefits are truly realized because (in common with all knock control systems) the results of any single experiment or simulation depends on the random arrival of knock events in that instance. In this article, methods are developed instead to compute the statistical properties of the closed-loop response of a general multi-threshold knock controller, thereby providing a much more complete and rigorous characterization of its performance than has previously been possible. The method is applied to single- and dual-threshold knock controllers and used to provide a rigorous comparison of the transient and steady-state performance of these different control laws. The method can also be used as a calibration aid to assess the effects of different controller gains in reliable, repeatable fashion.

Stochastic Simulation and Performance Analysis of Classical Knock Control Algorithms

The random nature of knock events means that deterministic notions for control system design, analysis, and performance assessment, in particular, cannot be applied to knock-control systems. New methods are, therefore, presented for predicting and quantifying the statistical properties of the closed-loop response of a standard advance–retard knock-control strategy based on a computationally efficient Markov or first-step type analysis. The results provide new insight into the steady-state and transient response of the controller and enable these characteristics to be quantified in a more rigorous and repeatable way than has previously been possible.

Closed loop statistical performance analysis of N-K knock controllers

The closed loop performance of engine knock controllers cannot be rigorously assessed from single experiments or simulations because knock behaves as a random process and therefore the response belongs to a random distribution also. In this work a new method is proposed for computing the distributions and expected values of the closed loop response, both in steady state and in response to disturbances. The method takes as its input the control law, and the knock propensity characteristic of the engine which is mapped from open loop steady state tests. The method is applicable to the ‘n-k’ class of knock controllers in which the control action is a function only of the number of cycles n since the last control move, and the number k of knock events that have occurred in this time. A Cumulative Summation (CumSum) based controller falls within this category, and the method is used to investigate the performance of the controller in a deeper and more rigorous way than has previously been possible. The results are validated using onerous Monte Carlo simulations, which confirm both the validity of the method and its high computational efficiency.

A New Approach for Ignition Timing Correction in Spark Ignition Engines Based on Cylinder Tendency to Surface Ignition

The performance of Spark Ignition (SI) engines in terms of thermal efficiency can be restricted by knock. Although it is common for all SI engines to exhibit knock from compressed end-gas, knocks from surface ignition remains a more serious problem due to its effect on combustion stability and its obscurity to detect. This paper focuses on predicting the occurrence of knocks from surface ignition by monitoring exhaust gas temperature (EGT). EGT measured during an engine cycle without the spark plug firing. Therefore, EGT rises illustrated any combustion made by surface ignition. Modelling and simulation of a one-dimensional engine combustion done by using GT-Power. The new approach reduces the complexity as EGT monitoring does not require high computational demands, and the EGT signals are robust to noise. The method is validated against a variety of fuel properties and across engine conditions. A new approach is proposed as a measure to predict and detect the knock events.

Stochastic Simulation of a CumSum Knock Controller

Single time-history simulations of knock control strategies cannot provide repeatable measures of closed loop performance because such responses are strongly dependent on the random arrival of knock events during the specific simulation or experiment. Stochastic simulation algorithms address this issue by computing the probability distribution of the closed loop response as a function of time from any given initial distribution. In this paper, a new algorithm is presented for stochastic simulation of a Cumulative Summation-based knock control strategy. The stochastic simulator is then used to assess the performance of this strategy in a scenario in which feedback sensitivity is also enhanced through the use of an optimized knock threshold. The results are compared to those for a traditional knock controller operating using a more conventional knock threshold level.

Closed Loop Knock Intensity Characteristics of a Classical Knock Control System

The closed loop behavior of knock control systems is hard to assess in any repeatable way because the response observed in any given simulation or experiment is heavily dependent on the particular random arrival of knock events in that instance. Recent work, however, has begun to address this issue by computing the statistical characteristics and time-varying distribution of the response, thereby encompassing all possible instances of the knock process. This paper briefly reviews this work using an improved notation, and then extends the method in order to compute the closed loop knock intensity distribution which is more fundamentally related to the risk of knock-related engine damage experienced by the engine.

Observations on the Measurement and Performance Impact of Catalyzed vs. Non Catalyzed EGR on a Heavily Downsized DISI Engine

Increasingly stringent regulations and rising fuel costs require that automotive manufacturers reduce their fleet CO2 emissions. Gasoline engine downsizing is one such technology at the forefront of improvements in fuel economy. As engine downsizing becomes more aggressive, normal engine operating points are moving into higher load regions, typically requiring over-fuelling to maintain exhaust gas temperatures within component protection limits and retarded ignition timings in order to mitigate knock and pre-ignition events. These two mechanisms are counterproductive, since the retarded ignition timing delays combustion, in turn raising exhaust gas temperature. A key process being used to inhibit the occurrence of these knock and pre-ignition phenomena is cooled exhaust gas recirculation (EGR). Cooled EGR lowers temperatures during the combustion process, reducing the possibility of knock, and can thus reduce or eliminate the need for over-fuelling. It has also been shown to reduce exhaust gas temperature and improve combustion efficiency through improved combustion phasing. This paper reviews data collected on ‘ULTRABOOST’, a collaborative research project which is co-funded by the Technology Strategy Board, the UK's innovation agency. The project is based on a prototype, heavily downsized 2.0L DISI engine running under controlled boost conditions, featuring low pressure or ‘long route’ EGR. This allows for the EGR feed gas to be fed from either pre- or post-catalyst sampling points. This paper will present data collected at equivalent operating conditions using both catalyzed and un-catalyzed exhaust gas to feed the EGR system. Some comparison of equivalent BMEP levels will be made but this is not the focus of the investigation. Current standards utilize a CO2 ratio for EGR measurement, but the results presented highlight inaccuracies in this measurement method when repeatable test conditions for pre- and post-catalyst EGR routes are required due to the significant changes in gas composition pre- and post-catalyst. The test engine was artificially boosted utilizing a Combustion Air Handling Unit to simulate commercially available turbo-machinery, which allowed for the catalyst effects to be assessed in isolation. Results from the engine are presented for constant intake manifold pressure and constant EGR mass flow to highlight the effects of the EGR feed gas composition on a range of engine performance metrics.

Likelihood-Based Control of Engine Knock

Engine knock is an undesirable phenomenon, which requires feedback control in order to maximize engine efficiency and avoid damage to the engine. In this paper, an analysis of experimental data is used to provide further evidence that knock behaves as a cyclically uncorrelated random process. It is argued that all knock controllers are therefore ultimately stochastic in nature and that the knock control problem is best undertaken within a stochastic framework. The properties of knock events are discussed and, based on these properties, a new likelihood-based stochastic knock controller is presented. The new controller achieves a significantly improved regulatory response relative to conventional strategies, while also maintaining a rapid transient response. It is therefore possible to operate closer to the knock limit without increasing the risk of engine damage.