Identification Of Hybrid Models Research Articles

White-Black-Box Hybrid Model Identification Based on RM-RF for Ship Maneuvering

A new system identification scheme based on model reference and random forest (RM-RF), was used to model ship maneuvering. First, the scheme establishes the relationship between a ship and the RM using the similarity rule. Second, a suitable RM was selected from public ship maneuvering models to avoid the tuning process in the empirical Maneuvering Modeling Group (MMG), model. Third, RF creates a relationship to map accelerations between the ship and the RM. Finally, the study case was implemented with fewer free-running model test data. The results show the feasibility of the identification modeling scheme and validated the generalizability of the proposed approach.

Design of Hybrid Model of Fertilizers Production Process for Automatic Control Purpose

An improved technique of hybrid modeling of fertilizers production processes for automatic control purpose is described. Design principles of mass balance systems for key components are outlined, and artificial neuron network (ANN) subsystems for modelling of absolute reaction rates are represented for ammonia production processes. The sensitivity equation approach was applied for parameters identification of a hybrid model for ammonia production processes. Possible difficulties and ways of practical implementation of the applied identification method for complex hybrid models for industrial chemical processes are outlined. The hybrid model identification results are presented and discussed. Ill. 5, bibl. 6 (in English; abstracts in English and Lithuanian). http://dx.doi.org/10.5755/j01.eee.114.8.704

Identification of semi-parametric hybrid process models

Hybrid models are mathematical models that comprise both mechanistic and black-box or data-driven components. Typically, the parameters in the mechanistic part of a hybrid model (if any) are assumed to be known. However in this research, a two-level approach is proposed for the identification of hybrid models where some parameters in the mechanistic part of the model are unknown. At the first level, the black-box component is identified using a regularization method with given values for the regularization and mechanistic parameters. At the second level, the regularization and mechanistic parameters are determined simultaneously and optimized according to a specific criterion placed on the predictive performance of the hybrid model. This approach is tested through the modelling of a toluene nitration process, where a support vector machine (SVM) model is used to represent the chemical kinetics, with the mass transfer-related mechanistic parameters being estimated simultaneously. The case study shows that good results can be obtained in terms of both the prediction of the process variables of interest and the estimates of the mechanistic parameters, when the measurement error in the training data is small whilst when the magnitude of the measurement error increases, the accuracy of the estimates of the mechanistic parameters decreases. However, the predictive performance of the resulting hybrid model in the latter case is still acceptable, and can be much better than that attained from the application of a pure black-box model under certain extrapolation conditions.

Incremental identification of hybrid process models

Missing knowledge about certain physical phenomena, such as chemical reaction kinetics or heat and mass transfer, hinders efficient and accurate modeling of complex processes. The so-called semi-empirical or hybrid modeling approach overcomes this obstacle by supplementing readily available physical model parts with empirical models representing insufficiently understood phenomena. However, hybrid model identification is still a challenging problem since usually (i) the complexity of each empirical model has to be adjusted, and (ii) many parameters have to be estimated by solving a large nonlinearly constrained optimization problem. In this paper we present an incremental hybrid model identification approach that decomposes the original problem into a sequence of simpler problems, such as data reconciliation, the solution of nonlinear equations, parameter estimation, and the separate identification of the empirical models by means of neural network training. The incremental approach is exemplified in a case study on the identification of a steady-state model of an ethylene glycol production process.

Identification of Hybrid Systems A Tutorial

This tutorial paper is concerned with the identification of hybrid models, i.e. dynamical models whose behavior is determined by interacting continuous and discrete dynamics. Methods specifically aimed at the identification of models with a hybrid structure are of very recent date. After discussing the main issues and difficulties connected with hybrid system identification, and giving an overview of the related literature, this paper focuses on four different approaches for the identification of switched affine and piecewise affine models, namely an algebraic procedure, a Bayesian procedure, a clustering-based procedure, and a bounded-error procedure. The main features of the selected procedures are presented, and possible interactions to still enhance their effectiveness are suggested.

The study of hybrid model identification, computation analysis and fault location for nonlinear dynamic circuits and systems

This paper presents the hybrid model identification for a class of nonlinear circuits and systems via a combination of the block-pulse function transform with the Volterra series. After discussing the method to establish the hybrid model and introducing the hybrid model identification, a set of relative formulas are derived for calculating the hybrid model and computing the Volterra series solution of nonlinear dynamic circuits and systems. In order to significantly reduce the computation cost for fault location, the paper presents a new fault diagnosis method based on multiple preset models that can be realized online. An example of identification simulation and fault diagnosis are given. Results show that the method has high accuracy and efficiency for fault location of nonlinear dynamic circuits and systems.

Real-time reservoir management: A multiscale adaptive optimization and control approach

We propose a decision-making approach for optimizing the profitability of hydrocarbon reservoirs. The proposed approach addresses the overwhelming complexity of the overall optimization problem by suggesting an oilfield operations hierarchy that entails different time scales. We discuss system identification, optimization, and control that are appropriate at various levels of the hierarchy and capitalize on the abilities of permanently instrumented and remotely actuated fields. Optimization is performed in real-time and is based on feedback. We provide details on real-time identification of hybrid models and their use at the scheduling and supervisory control levels. Case studies using field-calibrated simulation data demonstrate the applicability and value of the proposed approach. Directions for future development are given.

Neuro-fuzzy system modeling based on automatic fuzzy clustering

A neuro-fuzzy system model based on automatic fuzzy clustering is proposed. A hybrid model identification algorithm is also developed to decide the model structure and model parameters. The algorithm mainly includes three parts: 1) Automatic fuzzy C-means (AFCM), which is applied to generate fuzzy rules automatically, and then fix on the size of the neuro-fuzzy network, by which the complexity of system design is reducesd greatly at the price of the fitting capability; 2) Recursive least square estimation (RLSE). It is used to update the parameters of Takagi-Sugeno model, which is employed to describe the behavior of the system; 3) Gradient descent algorithm is also proposed for the fuzzy values according to the back propagation algorithm of neural network. Finally, modeling the dynamical equation of the two-link manipulator with the proposed approach is illustrated to validate the feasibility of the method.

Hybrid modelling of biotechnological processes using neural networks

The hybrid modelling approach for bioprocesses combines a neural network representation of the reaction rates with a mass–balance description of the reactor. A procedure for the identification of hybrid models is proposed and illustrated with an experimental case-study. The key feature is a state transformation which allows to identify separately the kinetic models of the reaction rates even if they occur simultaneously in the reactor.

Hybrid Modelling of Biotechnological Processes Using Neural Networks

The hybrid modelling approach for bioprocesses combines a neural network representation of the reaction rates with a mass-balance description of the reactor. A procedure for the identification of hybrid models is proposed and illustrated with an experimental case-study. The key feature is a state transformation which allows to identify separately the kinetic models of the reaction rates even if they occur simultaneously in the reactor.

Component Modal Synthesis Methods Based on Hybrid Models, Part II: Numerical Tests and Experimental Identification of Hybrid Models

A double modal synthesis method in which compatibility conditions at substructure interfaces are ensured by the introduction of loaded modes is presented in this study. These loaded modes, which are obtained by introducing mass loading or stiffness loading along the boundaries, are used to define generalized boundary coordinates. Thus the hybrid models presented in the first part of this study are developed so that they can be derived from test data as results of independent modal identifications. Unlike in classical modal synthesis methods, in this double modal synthesis method, it is not necessary to clamp the interfaces, which is always difficult to carry out during vibration tests. By introducing loaded modes, generalized boundary coordinates which represent boundary deformability in the frequency range under study can be defined.