scholarly journals Tdnn-Based Engine In-Cylinder Pressure Estimation from Shaft Velocity Spectral Representation

Sensors ◽  
2021 ◽  
Vol 21 (6) ◽  
pp. 2186
Author(s):  
Andrés F. Valencia-Duque ◽  
David A. Cárdenas-Peña ◽  
Andrés M. Álvarez-Meza ◽  
Álvaro A. Orozco-Gutiérrez ◽  
Héctor F. Quintero-Riaza
Keyword(s):  
Neural Network ◽  
Time Delay ◽  
Angular Velocity ◽  
Spectral Representation ◽  
Combustion Engine ◽  
Pulse Response ◽  
Cylinder Pressure ◽  
Direct Recording ◽  
Processing Steps ◽  
Engine Condition

Pressure is one of the essential variables to give information about engine condition and monitoring. Direct recording of this signal is complex and invasive, while angular velocity can be measured. Nonetheless, the challenge is to predict the cylinder pressure using the shaft kinematics accurately. In this paper, a time-delay neural network (TDNN), interpreted as a finite pulse response (FIR) filter, is proposed to estimate the in-cylinder pressure of a single-cylinder internal combustion engine (ICE) from fluctuations in shaft angular velocity. The experiments are conducted over data obtained from an ICE operating in 12 different states by changing the angular velocity and load. The TDNN’s delay is adjusted to get the highest possible correlation-based score. Our methodology can predict pressure with an R2 >0.9, avoiding complicated pre-processing steps.

scholarly journals Cylinder Pressure Prediction of An HCCI Engine Using Deep Learning

2021 ◽  
Vol 34 (1) ◽  
Author(s):  
Halit Yaşar ◽  
Gültekin Çağıl ◽  
Orhan Torkul ◽  
Merve Şişci
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Deep Learning ◽  
High Performance ◽  
Internal Combustion Engines ◽  
Deep Neural Network ◽  
Combustion Engine ◽  
Internal Combustion ◽  
Cylinder Pressure ◽  
Hcci Engine

AbstractEngine tests are both costly and time consuming in developing a new internal combustion engine. Therefore, it is of great importance to predict engine characteristics with high accuracy using artificial intelligence. Thus, it is possible to reduce engine testing costs and speed up the engine development process. Deep Learning is an effective artificial intelligence method that shows high performance in many research areas through its ability to learn high-level hidden features in data samples. The present paper describes a method to predict the cylinder pressure of a Homogeneous Charge Compression Ignition (HCCI) engine for various excess air coefficients by using Deep Neural Network, which is one of the Deep Learning methods and is based on the Artificial Neural Network (ANN). The Deep Learning results were compared with the ANN and experimental results. The results show that the difference between experimental and the Deep Neural Network (DNN) results were less than 1%. The best results were obtained by Deep Learning method. The cylinder pressure was predicted with a maximum accuracy of 97.83% of the experimental value by using ANN. On the other hand, the accuracy value was increased up to 99.84% using DNN. These results show that the DNN method can be used effectively to predict cylinder pressures of internal combustion engines.

A Time Delay Neural Network with Shared Weight Self-Attention for Small-Footprint Keyword Spotting

2019 ◽  
Author(s):  
Ye Bai ◽  
Jiangyan Yi ◽  
Jianhua Tao ◽  
Zhengqi Wen ◽  
Zhengkun Tian ◽  
...  
Keyword(s):  
Neural Network ◽  
Time Delay ◽  
Keyword Spotting ◽  
Small Footprint
Sign in / Sign up

Export Citation Format

Share Document