Research on hybrid modelling techniques for complex karst geology

We compare the performance of different acoustic modeling techniques on the task of distributed speech recognition (DSR). The DSR technology is interesting for speech recognition tasks in mobile environments, where features are sent from a thin client to a server where the actual recognition is performed. The evaluation is done on the TI digits database which consists of single digits and digit-chains spoken by American-English talkers. We investigate clean speech and speech added with white noise. Our results show that new hybrid or discrete modeling techniques can outperform standard continuous systems on this task.

New trends in modeling and control of hybrid systems

A propulsion dynamics investigation of a single-screw powered ship, driven by two gas turbines in a combined gas and gas (COGAG) arrangement is presented. Distributed-lumped (hybrid) modelling techniques are applied to the propulsion drive-train consisting of two prime movers, a primary drive shaft, a reduction gearbox, a secondary drive shaft and a propeller. The study in this paper, unlike earlier studies, does not lump the drive-train components as discussed by Rubis [1] into analytically simple but unrepresentative pointwise models. By way of contrast, modelling and simulation techniques for distributed-lumped systems are presented herein, enabling the torsional vibration and stress analysis of the long drive shafts to be investigated with accuracy. As a result of this a torque control scheme can be derived for fuel flow regulation purposes, enabling superior speed and shaft torque control to be achieved. The results presented show the simulated dynamic responses associated with the propulsion unit, shaft and propeller following fuel flow changes.

The quality of measurements depends directly on the quality of the measurement system model. With this concern, novel hybrid modelling techniques have been formulated for model performance enhancement. Air-gauge focus system sensor modelling has been accomplished by theoretical and empirical data integration. These modelling techniques combine a priori knowledge in theoretical model elaboration, and to attain the enhanced levels of adequacy, accuracy and precision they approximate the exact unknown model, simultaneously by available theoretical and appropriate polynomial empirical functions. For such a hybrid model the solution of two approaches by means of linear transformation and successive linear and nonlinear transformations have been developed. The validations of elaborated air-gauge sensor models revealed that sensor hybrid model solving by successive linear and nonlinear transformations permits us to attain minimum discrepancy with empirical evidence for the whole region of interest for model predictor variables.

Modelling pressure drop in water treatment

Different empirical and hybrid modelling techniques were evaluated in order to improve the modelling compared with the present situation in the plate rolling mill at Rautaruukki, Raahe. Logged data from the production were used to develop new models and to compare them with each other and with the present model. The data describe the chemical composition and various process parameters of the rolling process, including thermal, geometric and kinematic parameters. Various neural networks and multivariate models were evaluated in order to take non‐linear effects into account. Hybrid methods use a combination of physically based models and empirical models. The simple physically based formula, force = area x yield strength factor, proved useful. A more advanced model solving von Karman's differential equation was evaluated as a stand‐alone model and as part of a hybrid model. The measured force of the previous rolling pass was tested as an additional input variable. Various aspects of the different modelling techniques are discussed and compared. Simple but improved models are described as suitable candidates to replace the present on‐line model.

Constructing predictive models to simulate complex bioprocess dynamics, particularly time‐varying (i.e., parameters varying over time) and history‐dependent (i.e., current kinetics dependent on historical culture conditions) behavior, has been a longstanding research challenge. Current advances in hybrid modeling offer a solution to this by integrating kinetic models with data‐driven techniques. This article proposes a novel two‐step framework: first (i) speculate and combine several possible kinetic model structures sourced from process and phenomenological knowledge, then (ii) identify the most likely kinetic model structure and its parameter values using model‐free Reinforcement Learning (RL). Specifically, Step 1 collates feasible history‐dependent model structures, then Step 2 uses RL to simultaneously identify the correct model structure and the time‐varying parameter trajectories. To demonstrate the performance of this framework, a range of in‐silico case studies were carried out. The results show that the proposed framework can efficiently construct high‐fidelity models to quantify both time‐varying and history‐dependent kinetic behaviors while minimizing the risks of over‐parametrization and over‐fitting. Finally, the primary advantages of the proposed framework and its limitation were thoroughly discussed in comparison to other existing hybrid modeling and model structure identification techniques, highlighting the potential of this framework for general bioprocess modeling.

Shock response prediction of the typical structure in spacecraft based on the hybrid modeling techniques

SummarySurrogate modeling techniques have been increasingly developed for optimization and uncertainty quantification problems in many engineering fields. The development of surrogates requires modeling high‐dimensional and nonsmooth functions with limited information. To this end, the hybrid surrogate modeling method, where different surrogate models are combined, offers an effective solution. In this paper, a new hybrid modeling technique is proposed by combining polynomial chaos expansion and kernel function in a sparse Bayesian learning framework. The proposed hybrid model possesses both the global characteristic advantage of polynomial chaos expansion and the local characteristic advantage of the Gaussian kernel. The parameterized priors are utilized to encourage the sparsity of the model. Moreover, an optimization algorithm aiming at maximizing Bayesian evidence is proposed for parameter optimization. To assess the performance of the proposed method, a detailed comparison is made with the well‐established PC‐Kriging technique. The results show that the proposed method is superior in terms of accuracy and robustness.

A direct hybrid finite element–wave based modelling technique for efficient analysis of poroelastic materials in steady-state acoustic problems

Hybrid systems are systems in which the discrete dynamics and the continuous dynamics not only coexist but also interact with each other. The modeling of such hybrid systems is different from the convention modeling techniques followed for purely discrete or purely continuous systems. Development of techniques for Modeling, system identification and controller synthesis of hybrid systems is an upcoming research area. In this paper a three tank benchmark system is modeled using hybrid modeling technique and the modeled system is simulated using SIMULINK and a controller is designed for the hybrid system using STATEFLOW in MATLAB.

This paper demonstrates the execution investigation of the automatic continuous speech recognition system for Kannada language using hybrid modelling techniques. The well-known modelling techniques, in particular, deep neural system (DNN), hidden Markov display (HMM), subspace Gaussian mixture model (SGMM) and maximum mutual information (MMI) have been combined to form the hybrid modelling for speech recognition. The persistent Kannada speech information is gathered from 1600 speakers (960 males and 640 females) of the age bunch in the scope of 8 years-80 years. The speech information is acquired from different geographical regions of the Karnataka state under certifiable condition. It comprises of 20,000 words that spread 30 locale. The point of this paper is to examine the execution of hybrid modelling techniques with regards to Kannada speech recognition. Kaldi toolbox is utilized for the implementation of this system, in which Mel frequency cepstral coefficient (MFCC) is used as a feature extraction procedure. The word error rate (WRR) is the error metric used to determine the efficiency of the automatic speech recognition (ASR) system. The experimental results demonstrate that the recognition rate got through the combination of DNN and HMM is better over other hybrid ASR modelling strategies.

This paper investigates the design and verification of a hybrid model predictive controller of a damping multi-mode switching damper for application in vehicle suspensions. Since the damping mode switches induce different modes of operation, the vehicle suspension system including this damper poses challenging hybrid control problem. To solve this problem, a novel approach to the modelling and controller design problem is proposed based on hybrid modelling and model predictive control techniques. The vehicle suspension system with the damping multi-mode switching damper is formulated as a mixed logical dynamical model comprising continuous and discrete system inputs. Based on this model, a constrained optimal control problem is solved to manage the switching sequences of the damping mode with respect to the suspension performance requirements. Numerical simulation results demonstrate the effectiveness of the proposed control methodology finally.

This paper illustrates the application of hybrid modeling and model predictive control techniques to the management of air-to-fuel ratio and torque in advanced technology gasoline direct-injection stratified-charge (DISC) engines. A DISC engine is an example of a constrained hybrid dynamical system, because it can operate in two distinct modes (stratified and homogeneous) and because the mode-dependent constraints on the air-to-fuel ratio and on the spark timing need to be enforced during its operation to avoid misfire, knock, and high combustion variability. In this paper, we approximate the DISC engine dynamics as a two-mode discrete-time switched affine system. Using this approximation, we tune a hybrid model predictive controller with integral action based on online mixed-integer quadratic optimization, and show the effectiveness of the approach through simulations. Then, using an offline multiparametric optimization procedure, we convert the controller into an equivalent explicit piecewise affine form that is easily implementable in an automotive microcontroller through a lookup table of linear gains

This paper illustrates the application of hybrid modeling and model predictive control techniques to the water purge management in a fuel cell with dead-end anode. The anode water flow dynamics are approximated as a two-mode discrete-time switched affine system that describes the propagation of water inside the gas diffusion layer, the spilling into the channel and consequent filling and plugging the channel. Using this dynamical approximation, a hybrid model predictive controller based on on-line mixed-integer quadratic optimization is tuned, and the effectiveness of the approach is shown through simulations with a high-fidelity model. Then, using an off-line multiparametric optimization procedure, the controller is converted into an equivalent piecewise affine form which is easily implementable even in an embedded controller through a lookup table of affine gains.