Using physics to extend the range of machine learning models for an aerodynamic, hydraulic and combusting system: The toy model concept

Brahma, Indranil; Robert, J.; Freid, Bradley

doi:10.1016/j.egyai.2021.100113

Cited by 10 publications

(3 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In turn, Gregov [17] proposed a new approach for modeling and predictive analysis of the GEROLER hydraulic motor based on two types of multi-layer ANNs: static feed-forward and dynamic NARX. A toy model concept applicable to machine learning models, including neural networks, was proposed by Brahma [18]. Functional capabilities of the concept in the example of a hydraulic turbine efficiency prediction were presented.…”

Section: Neural Predictionmentioning

confidence: 99%

Artificial Intelligence Methods in Hydraulic System Design

Filo

2023

Energies

View full text Add to dashboard Cite

Reducing energy consumption and increasing operational efficiency are currently among the leading research topics in the design of hydraulic systems. In recent years, hydraulic system modeling and design techniques have rapidly expanded, especially using artificial intelligence methods. Due to the variety of algorithms, methods, and tools of artificial intelligence, it is possible to consider the prospects and directions of their further development. The analysis of the most recent publications allowed three leading technologies to be indicated, including artificial neural networks, evolutionary algorithms, and fuzzy logic. This article summarizes their current applications in the research, main advantages, and limitations, as well as expected directions for further development.

show abstract

Section: Neural Predictionmentioning

confidence: 99%

Artificial Intelligence Methods in Hydraulic System Design

Filo

2023

Energies

View full text Add to dashboard Cite

show abstract

“…Tan et al used extremum seeking and online optimization of engine model parameters to maximize IMEP [14]; an average error of 0.17 bar was achieved with this online-based strategy. The integration of other unique data-driven, empirical, and semi-empirical methods has also been explored in [15,16], where physics-based models were utilized to extend the range of machine learning predictions via the space transform. The method was effectively demonstrated for predictions of aerodynamic forces and hydraulic turbine efficiencies, as well as diesel engine calibration and combustion engine emissions.…”

Section: Introductionmentioning

confidence: 99%

“…The literature demonstrates that leveraging physics to inform ANN models could enhance predictions of engine metrics and, as such, improve real-time combustion phasing control methods without the strict need for in-cylinder sensors. While prior work using either direct ANNs or integrated ANN and physics-based approaches to predict critical engine metrics has shown promise [1,[5][6][7]13,15,16,[18][19][20][21][22][23][24], some challenges with uncertainty and complexity still need to be addressed. Some existing methods assume linear relationships and, as such, are simpler to use for control, but have a more limited ability to extrapolate, and encounter greater inaccuracy due to the nonlinearity of engine processes [7,13,18,22].…”

Section: Introductionmentioning

confidence: 99%

A Hybrid Physics-Based and Stochastic Neural Network Model Structure for Diesel Engine Combustion Events

Ankobea-Ansah

Hall

2022

Vehicles

View full text Add to dashboard Cite

Estimation of combustion phasing and power production is essential to ensuring proper combustion and load control. However, archetypal control-oriented physics-based combustion models can become computationally expensive if highly accurate predictive capabilities are achieved. Artificial neural network (ANN) models, on the other hand, may provide superior predictive and computational capabilities. However, using classical ANNs for model-based prediction and control can be challenging, since their heuristic and deterministic black-box nature may make them intractable or create instabilities. In this paper, a hybridized modeling framework that leverages the advantages of both physics-based and stochastic neural network modeling approaches is utilized to capture CA50 (the timing when 50% of the fuel energy has been released) along with indicated mean effective pressure (IMEP). The performance of the hybridized framework is compared to a classical ANN and a physics-based-only framework in a stochastic environment. To ensure high robustness and low computational burden in the hybrid framework, the CA50 input parameters along with IMEP are captured with a Bayesian regularized ANN (BRANN) and then integrated into an overall physics-based 0D Wiebe model. The outputs of the hybridized CA50 and IMEP models are then successively fine-tuned with BRANN transfer learning models (TLMs). The study shows that in the presence of a Gaussian-distributed model uncertainty, the proposed hybridized model framework can achieve an RMSE of 1.3 × 10−5 CAD and 4.37 kPa with a 45.4 and 3.6 s total model runtime for CA50 and IMEP, respectively, for over 200 steady-state engine operating conditions. As such, this model framework may be a useful tool for real-time combustion control where in-cylinder feedback is limited.

show abstract