Davide Moro

In-Cylinder Pressure Reconstruction Based on Instantaneous Engine Speed Signal

This paper presents an original methodology for the instantaneous in-cylinder pressure waveform reconstruction in a spark-ignited internal combustion engine. The methodology is based on the existence of a linear correlation, characterized by frequency response functions, between in-cylinder pressure and engine speed signals. This correlation is experimentally verified and evaluated by simultaneous measurements of the above-mentioned quantities. The evaluation of different frequency response functions, one for each steady-state condition investigated, allows recovering the pressure waveform even under other engine running conditions (i.e., transients). In this way, during on-board operation, the pressure waveform could be recovered using only the engine speed signal, already present in current production electronic control units. In this paper the signal processing methodology and some experimental results, obtained during transient tests, are presented. The methodology could be interesting for the development of advanced engine control strategies aimed at the management of the torque generated by the engine. As an example, traction control in drive-by-wire systems could be a possible challenging application. The in-cylinder pressure reconstruction performed using the frequency response functions, in fact, allows the evaluation of the indicated torque. An important characteristic of this methodology is, furthermore, the diagnostic capability for the combustion process, that is guaranteed by the linear correlation between in-cylinder pressure and instantaneous engine speed waveforms. Also in presence of a misfiring cylinder, when the instantaneous engine speed waveform is strongly affected by the absence of combustion, the reconstructed in-cylinder pressure shows a good agreement with the measured one. The experimental tests have been conducted in a test cell using a four-cylinder production engine. It has to be noted, anyway, that the same methodology can be applied to engines with a higher number of cylinders.

Closed-loop individual cylinder air–fuel ratio control via UEGO signal spectral analysis

The paper presents the development and real time application of an original closed-loop individual cylinder AFR control system, based on a spectral analysis of the lambda sensor signal. The observation that any type of AFR disparity between the various cylinders is reflected in a specific harmonic content of the AFR signal spectrum, represents the starting point of the project. The results observed on a 4 cylinder Spark Ignition engine are encouraging, since in the investigated engine operating conditions the controller is able to reduce AFR inequality below 0.01 lambda. The paper also shows how the proposed controller can be applied to other engine configurations.

Automatic Combustion Phase Calibration With Extremum Seeking Approach

One of the most effective factors influencing performance, efficiency, and pollutant emissions of internal combustion engines is the combustion phasing: in gasoline engines electronic control units (ECUs) manage the spark advance (SA) in order to set the optimal combustion phase. Combustion control is assuming a crucial role in reducing engine tailpipe emissions and maximizing performance. The number of actuations influencing the combustion is increasing, and as a consequence, the calibration of control parameters is becoming challenging. One of the most effective factors influencing performance and efficiency is the combustion phasing: for gasoline engines, control variables such as SA, air-to-fuel ratio (AFR), variable valve timing (VVT), and exhaust gas recirculation (EGR) are mostly used to set the combustion phasing. The optimal control setting can be chosen according to a target function (cost or merit function), taking into account performance indicators, such as indicated mean effective pressure (IMEP), brake-specific fuel consumption (BSFC), pollutant emissions, or other indexes inherent to reliability issues, such as exhaust gas temperature or knock intensity. Many different approaches can be used to reach the best calibration settings: design of experiment (DOE) is a common option when many parameters influence the results, but other methodologies are in use: some of them are based on the knowledge of the controlled system behavior by means of models that are identified during the calibration process. The paper proposes the use of a different concept, based on the extremum seeking approach. The main idea consists in changing the values of each control parameter at the same time, identifying its effect on the monitored target function, and allowing to shift automatically the control setting towards the optimum solution throughout the calibration procedure. An original technique for the recognition of control parameters variations effect on the target function is introduced, based on spectral analysis. The methodology has been applied to data referring to different engines and operating conditions, using IMEP, exhaust temperature, and knock intensity for the definition of the target function and using SA and AFR as control variables. The approach proved to be efficient in reaching the optimum control setting, showing that the optimal setting can be achieved rapidly and consistently.

Combustion control using a Model-based MFB50 estimation methodology

Abstract Combustion control has always been a key factor in modern Internal Combustion Engines management systems, being even more crucial nowadays to improve drivability and reduce pollutant emissions. The position where 50% of fuel mass burned over an engine cycle is reached (MFB50) provides very important information about the effectiveness of combustion, such as the kind of combustion that is taking place. MFB50 can be evaluated using in-cylinder pressure sensors, even though at present, they are not widely used for on-board applications, as a result of problems due to measurement reliability and cost. Even if it might be possible to solve some of the problems related to measurement reliability in the future, pressure sensors will constitute a relevant part of the whole engine control system cost. This work presents a model based MFB50 estimation algorithm that requires no additional cost, because its based on the instantaneous engine speed measurement, which is already performed in modern engine control systems. This procedure, whose computational cost is compatible with a common engine control unit, has been applied successfully to a turbocharged Diesel engine mounted on-board a vehicle.

A methodology for increasing the signal to noise ratio for the misfire detection at high speed in a high performance engine

Abstract The paper presents a methodology for pre-processing the combustion time intervals, that is the basic signal used in misfire detection strategies, with the aim of increasing the signal-to-noise ratio to enable a more efficient misfire diagnosis, especially when the engine is running at high speeds and low loads. The performance of the basic misfire detection algorithm shows that in those engine operating conditions the background noise amplitude has approximately the same value of the information related to the misfire presence, thus hiding the misfire event that may not be detected. The proposed methodology is based on the correction of the combustion time signal cycle-by-cycle, using a vector of data that take into account the specific behavior of every cylinder. The vector of data for the combustion time correction is stored in a map inside the control unit and could be continuously updated with an auto-adaptive learning technique.

Combustion Noise Real-Time Evaluation and Processing for Combustion Control

Newly developed Diesel engine control strategies are mainly aimed at pollutant emissions reduction, due to the increasing request for engine-out emissions and fuel consumption reduction. In order to reduce engine-out emissions, the development of closed-loop combustion control algorithms has become crucial. Modern closed-loop combustion control strategies are characterized by two main aspects: the use of high EGR rates (the goal being to obtain highly premixed combustions) and the control of the center of combustion. In order to achieve the target center of combustion, conventional combustion control algorithms correct the measured value by varying Main injection timing.It is possible to obtain a further reduction in pollutant emissions through a proper variation of the injection parameters. Modern Diesel engine injection systems allow designing injection patterns with many degrees of freedom, due to the large number of tuneable injection parameters (such as start and duration of each injection). Each injection parameter’s variation causes variations in the whole combustion process and, consequently, in pollutant emissions production. Injection parameters variations have a strong influence on other quantities that are related to combustion process effectiveness, such as noise radiated by the engine. This work presents a methodology that allows real-time evaluating combustion noise on-board a vehicle. The radiated noise can be evaluated through a proper in-cylinder pressure signal processing. Even though in-cylinder pressure sensor on-board installation is still uncommon, it is believed that in-cylinder pressure measurements will be regularly available on-board thanks to the newly developed piezo-resistive sensors.In order to set-up the methodology, several experimental tests have been performed on a 1.3 liter Diesel engine mounted in a test cell. The engine was run, in each operating condition, both activating and deactivating pre-injections, since pre-injections omission usually produces a decrease in pollutant emissions production (especially in particulate matter) and a simultaneous increase in engine noise. The investigation of the correlation between combustion process and engine noise can be used to set up a closed-loop algorithm for optimal combustion control based on engine noise prediction.Copyright

Development of a Misfire Detection’s Technique Based on an Engine’s Torsional Model

The recent OBD requirements enforce the misfire’s diagnosis and the isolation of the cylinder where the missing combustion took place. Most of the common-used techniques developed are based on the engine’s angular speed, that is derived by the signal usually measured with an inductive or Hall-effect sensor already used for the engine’s control. The presence of single or multiple misfires (several misfires within the same engine’s cycle) induces torsional vibration in the powertrain, requiring specific filtering of the diagnostic signal to avoid false alarms. This paper presents some preliminary results, related to a 4 cylinder 1.2 liter engine mounted on an eddy-current brake test bench, obtained by a new diagnosis technique based on two speed sensors, placed near the toothed wheels used respectively for the engine and current brake’s control. The signals coming from the two sensors, applied to an equation derived by a torsional model of the engine powertrain, allow to evaluate an index based on the difference between engine and brake’s torque that highlights the misfire presence. It will be shown that this index does not require any particular calibration procedure. Experimental tests, in which single and multiple misfires are induced in several operating conditions, show clearly the algorithm’s robustness in misfire detection, especially in multiple misfire tests, where the misfiring cylinders are exactly detected.Copyright

Remote Combustion Sensing Methodology for Non-Intrusive Cylinder Pressure Estimation in Diesel Engines

Abstract The pollutant emission reduction requested by future pollutant emission policies spawned a great deal of research in the field of combustion monitoring and closed-loop control. This is especially true for Diesel engine, since an important reduction in NOx and particular matter will be required. Many works demonstrate that a major engine-out emission reduction can be achieved through a closed-loop combustion control methodology based on cylinder pressure trace processing. However, the main obstacles to the use of cylinder pressure sensors for on-board application are measurement reliability over time and cost. Therefore, this paper describes the development of a methodology that allows estimating cylinder pressure through a proper combination of the information coming from low-cost sensors mounted on the engine.

Closed-Loop Individual Cylinder Air-Fuel Ratio Control Via UEGO Signal Spectral Analysis

The paper presents the development and real time application of an original closed-loop individual cylinder AFR control system, based on a spectral analysis of the lambda sensor signal. The observation that any type of AFR disparity between the various cylinders is reflected in a specific harmonic content of the AFR signal spectrum, represents the starting point of the project. The results observed on a 4 cylinder Spark Ignition engine are encouraging, since in the investigated engine operating conditions the controller is able to reduce AFR inequality below 0.01 lambda. The paper also shows how the proposed controller can be applied to other engine configurations.

A Methodology for Increasing the Signal to Noise Ratio for the Misfire Detection at High Speed in a High Performance Engine

Abstract The paper presents a methodology for the pre-processing of the combustion time intervals, that is the basic signal used in misfire detection strategies, with the aim of increasing the signal-to-noise ratio to enable a more efficient misfire diagnosis, especially when the engine is running at high speeds and low loads. The performance of the basic misfire detection algorithm shows that in those engine operating conditions the background noise amplitude has approximately the same value of the information related to the misfire presence, thus hiding the misfire event that may not be detected. The proposed methodology is based on the correction of the combustion time signal cycle-by-cycle, using a vector of data that take into account the specific behavior of every cylinder combustion. For each engine running condition, the vector of data for the combustion time correction is stored in a map inside the control unit and could be continuously updated with an auto-adaptive learning technique. Some results of the methodology are reported in the paper, and compared with the basic misfire detection algorithm performance.