Search results

This study aims to investigate the possible use of a deep neural network (DNN) as an inverse solver.

Different models based on DNNs are designed and proposed for the resolution of inverse electromagnetic problems either as fast solvers for the direct problem or as straightforward inverse problem solvers, with reference to the TEAM 25 benchmark problem for the sake of exemplification.

Using DNNs as straightforward inverse problem solvers has relevant advantages in terms of promptness but requires a careful treatment of the underlying problem ill-posedness.

This work is one of the first attempts to exploit DNNs for inverse problem resolution in low-frequency electromagnetism. Results on the TEAM 25 test problem show the potential effectiveness of the approach but also highlight the need for a careful choice of the training data set.

The purpose of this paper is to develop a transform method and a deep learning model to identify the inner surface shape based on the measurement temperature at the outer boundary of the pipe.

The training process is assisted by the finite element method (FEM) simulation which solves the direct problem for the data preparation. To avoid re-meshing the domain when the inner surface shape varies, a new transform method is proposed to transform the shape identification problem into the effective thermal conductivity identification problem. The deep learning model is established to set up the relationship between the measurement temperature and the effective thermal conductivity. Then the unknown geometry shape is acquired by the mapping between the inner shape and the effective thermal conductivity through the inverse transform method.

The new method is successfully applied to identify the internal boundary of a pipe with eccentric circle, ellipse and nephroid inner geometries. The results show that as the measurement points increased and the measurement error decreased, the results became more accurate. The position of the measurement point and mesh density of the FEM model have less effect on the results.

The deep learning model and the transform method are developed to identify the pipe inner surface shape. There is no need to re-mesh the domain during the computation progress. The results show that the proposed method is a fast and an accurate tool for identifying the pipe inner surface.

This paper discusses an algorithm for phase change front identification in continuous casting. The problem is formulated as an inverse geometry problem, and the solution procedure utilizes temperature measurements inside the solid phase and sensitivity coefficients. The proposed algorithms make use of the boundary element method, with cubic boundary elements and Bezier splines employed for modelling the interface between the solid and liquid phases. A case study of continuous casting of copper is solved to demonstrate the main features of the proposed algorithms.

The purpose of this paper is to present an efficient algorithm based on a multi‐level adaptive mesh refinement strategy for the solution of ill‐posed inverse heat conduction problems arising in pool boiling using few temperature observations.

The stable solution of the inverse problem is obtained by applying the conjugate gradient method for the normal equation method together with a discrepancy stopping rule. The resulting three‐dimensional direct, adjoin and sensitivity problems are solved numerically by a space‐time finite element method. A multi‐level computational approach, which uses an a posteriori error estimator to adaptively refine the meshes on different levels, is proposed to speed up the entire inverse solution procedure.

This systematic approach can efficiently solve the large‐scale inverse problem considered without losing necessary detail in the estimated quantities. It is shown that the choice of different termination parameters in the discrepancy stopping conditions for each level is crucial for obtaining a good overall estimation quality. The proposed algorithm has also been applied to real experimental data in pool boiling. It shows high computational efficiency and good estimation quality.

The high efficiency of the approach presented in the paper allows the fast processing of experimental data at many operating conditions along the entire boiling curve, which has been considered previously as computationally intractable. The present study is the authors' first step towards a systematic approach to consider an adaptive mesh refinement for the solution of large‐scale inverse boiling problems.

The purpose of this paper is to reduce the destructive effects of existing unavoidable noises contaminating temperature data in inverse heat conduction problems (IHCP) utilizing the wavelets.

For noise reduction, sensor data were treated as input to the filter bank used for signal decomposition and implementation of discrete wavelet transform. This is followed by the application of wavelet denoising algorithm that is applied on the wavelet coefficients of signal components at different resolution levels. Both noisy and de‐noised measurement temperatures are then used as input data to a numerical experiment of IHCP. The inverse problem deals with an estimation of unknown surface heat flux in a 2D slab and is solved by the variable metric method.

Comparison of estimated heat fluxes obtained using denoised data with those using original sensor data indicates that noise reduction by wavelet has a potential to be a powerful tool for improvement of IHCP results.

Noise reduction using wavelets, while it can be implemented very easily, may also significantly relegate (or even eliminate) conventional regularization schemes commonly used in IHCP.

The purpose of this paper is to present an inverse analysis procedure based on a coupled numerical formulation through which the coefficients describing non‐linear thermal properties of blood perfusion may be identified.

The coupled numerical technique involves a combination of the dual reciprocity boundary element method (DRBEM) and a genetic algorithm (GA) for the solution of the Pennes bioheat equation. Both linear and quadratic temperature‐dependent variations are considered for the blood perfusion.

The proposed DRBEM formulation requires no internal discretisation and, in this case, no internal nodes either, apart from those defining the interface tissue/tumour. It is seen that the skin temperature variation changes as the blood perfusion increases, and in certain cases flat or nearly flat curves are produced. The proposed algorithm has difficulty to identify the perfusion parameters in these cases, although a more advanced genetic algorithm may provide improved results.

The coupled technique allows accurate inverse solutions of the Pennes bioheat equation for quantitative diagnostics on the physiological conditions of biological bodies and for optimisation of hyperthermia for cancer therapy.

The proposed technique can be used to guide hyperthermia cancer treatment, which normally involves heating tissue to 42‐43°C. When heated up to this range of temperatures, the blood flow in normal tissues, e.g. skin and muscle, increases significantly, while blood flow in the tumour zone decreases. Therefore, the consideration of temperature‐dependent blood perfusion in this case is not only essential for the correct modelling of the problem, but also should provide larger skin temperature variations, making the identification problem easier.

The purpose of this paper is to propose a numerical procedure for discrete identification of the missing part of the domain boundary in a heat conduction problem. A new approach to sensitivity analysis is intended to give a better understanding of the influence of measurement error on boundary reconstruction.

The solution of Laplace’s equation is obtained using the Trefftz method, and then each of the sought boundary points can be derived numerically from a nonlinear equation. The sensitivity analysis comes down to the analytical evaluation of a sensitivity factor.

The proposed method very accurately recovers the unknown boundary, including irregular shapes. Even a very large number of the boundary points can be determined without causing computational problems. The sensitivity factor provides quantitative assessment of the relationship between the temperature measurement errors and boundary identification errors. The numerical examples show that some boundary reconstruction problems are error-sensitive by nature but such problems can be recognized with the use of a sensitive factor.

The present approach based on the Trefftz method separates, in terms of computation, specification of the coefficients appearing in the Trefftz method and missing coordinates of the sought boundary points. Due to introducing a sensitivity factor, a more profound sensitivity analysis was successfully conducted.

Many a times, the information about the boundary heat flux is obtained only through inverse approach by locating the thermocouple or temperature sensor in accessible boundary. Most of the work reported in literature for the estimation of unknown parameters is based on heat conduction model. Inverse approach using conjugate heat transfer is found inadequate in literature. Therefore, the purpose of the paper is to develop a 3D conjugate heat transfer model without model reduction for the estimation of heat flux and heat transfer coefficient from the measured temperatures.

A 3 D conjugate fin heat transfer model is solved using commercial software for the known boundary conditions. Navier–Stokes equation is solved to obtain the necessary temperature distribution of the fin. Later, the complete model is replaced with neural network to expedite the computations of the forward problem. For the inverse approach, genetic algorithm (GA) and particle swarm optimization (PSO) are applied to estimate the unknown parameters. Eventually, a hybrid algorithm is proposed by combining PSO with Broyden–Fletcher–Goldfarb–Shanno (BFGS) method that outperforms GA and PSO.

The authors demonstrate that the evolutionary algorithms can be used to obtain accurate results from simulated measurements. Efficacy of the hybrid algorithm is established using real time measurements. The hybrid algorithm (PSO-BFGS) is more efficient in the estimation of unknown parameters for experimentally measured temperature data compared to GA and PSO algorithms.

Surrogate model using ANN based on computational fluid dynamics simulations and in-house steady state fin experiments to estimate the heat flux and heat transfer coefficient separately using GA, PSO and PSO-BFGS.

The purpose of this paper is to develop a new genetic optimization strategy which provides computationally more efficient and accurate solutions, and to provide practically applicable optimization method in radar cross‐section (RCS) minimization problems.

The problem of RCS minimization for three‐dimensional air vehicle is considered. New computationally efficient optimization tool; neural networks (NNs) coupled multi‐frequency vibrational genetic algorithm (NN‐coupled VGA^m) is based on genetic algorithm (GA) search strategy together with NNs. The results include RCS minimization problem of an air vehicle under structural and aero dynamical‐related geometry constraints.

For the demonstration problem considered, remarkable reduction in the computational time has been accomplished.

The results reported in this paper suggest an efficient GA optimization methodology for engineering problems.

Owing to reduction in computational time, the new method provides a shorter design cycle for engineering problems.

The study of heat transfer mechanisms is an area of great interest because of various applications that can be developed. Mathematically, these phenomena are usually represented by partial differential equations associated with initial and boundary conditions. In general, the resolution of these problems requires using numerical techniques through discretization of boundary and internal points of the domain considered, implying a high computational cost. As an alternative to reducing computational costs, various approaches based on meshless (or meshfree) methods have been evaluated in the literature. In this contribution, the purpose of this paper is to formulate and solve direct and inverse problems applied to Laplace’s equation (steady state and bi-dimensional) considering different geometries and regularization techniques. For this purpose, the method of fundamental solutions is associated to Tikhonov regularization or the singular value decomposition method for solving the direct problem and the differential Evolution algorithm is considered as an optimization tool for solving the inverse problem. From the obtained results, it was observed that using a regularization technique is very important for obtaining a reliable solution. Concerning the inverse problem, it was concluded that the results obtained by the proposed methodology were considered satisfactory, as even with different levels of noise, good estimates for design variables in proposed inverse problems were obtained.

In this contribution, the method of fundamental solution is used to solve inverse problems considering the Laplace equation.

In general, the proposed methodology was able to solve inverse problems considering different geometries.

The association between the differential evolution algorithm and the method of fundamental solutions is the major contribution.