A robust surrogate model of a solid oxide cell based on an adaptive polynomial approximation method

Chi, Yingtian; Qiu, Yiwei; Lin, Jin; Song, Yonghua; Li, Wenying; Hu, Qian; Mu, Shujun; Liu, Min

doi:10.1016/j.ijhydene.2020.09.116

Cited by 17 publications

(5 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the fully-coupled model, the cell-level fine model is coupled with the stack-level model by providing the electrochemical heat source term and mass source terms of O2, H2, and H2O, while the stack-level model provides the distribution of 𝑇, 𝑝 𝑂 2 , 𝑝 𝐻 2 , 𝑝 𝐻 2 𝑂 as the boundary conditions for cell-level models. To avoid computing the cell-level non-uniform distributions of mass and heat sources which do not significantly affect the stack-level temperature and mass distributions, the APA data-driven method, proposed by our group in [9], is used to build surrogate models for single cells. In other words, the celllevel fine model is simulated under several sampling points (different values of 𝑇 𝑎𝑣 , 𝐼, 𝐹𝑈, 𝑝 𝑂 2 ,𝑎𝑣 ), and the simulation results are used to identify the relationships between 𝑄 ℎ𝑒𝑎𝑡 , 𝑄 𝑚𝑎𝑠𝑠,𝑂 2 , 𝑉 𝑐𝑒𝑙𝑙 and 𝑇 𝑎𝑣 , 𝐼, 𝐹𝑈, 𝑝 𝑂 2 ,𝑎𝑣 , formulated as follows.…”

Section: Fully-coupled 3d Models and The Simplification Methods Based...mentioning

confidence: 99%

“…The first generation of FCSOC is electrolyte-supported, resulting in a low power density and a working temperature of around 800℃. In [9], we reported the latest anode-supported generation of FCSOC, lowering the working temperature to 600℃ and improving the power density. In this paper, the stacking scheme for the anode-supported FCSOCs is introduced for the first time.…”

Section: Introductionmentioning

confidence: 99%

“…= 𝑎 3,0 + 𝑎 3,1 𝑇 𝑎𝑣 + 𝑎 3,2 𝐼 + 𝑎 3,3 𝐹𝑈 + 𝑎 3,4 𝑝 𝑂 2 ,𝑎𝑣 + ⋯ 𝑎 𝑖,𝑗 is the coefficient of the 𝑗th term in the 𝑖th model. The number of terms and values of 𝑎 𝑖,𝑗 are adaptively determined by the APA method, whose detail is described in[9]. 𝑄 𝑚𝑎𝑠𝑠,𝑂 2 is directly calculated by the current 𝐼 since 𝐼 represents the electrochemical reaction rate.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Simplified Three-dimensional Models for Studying the Non-uniformity inside the Flat-chip Solid Oxide Cell Stacks

Yingtian¹,

Chi²,

Qiu³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Fully-coupled 3d Models and The Simplification Methods Based...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Simplified Three-dimensional Models for Studying the Non-uniformity inside the Flat-chip Solid Oxide Cell Stacks

Yingtian¹,

Chi²,

Qiu³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In order to solve this problem, the APA method is selected to explore the relationship among I r_rms , I S_rms , P Fe and the resonant parameters, which utilizes sampling points to construct a polynomial as the surrogate model to predict the output. Besides, it has been theoretically proven in the approximation theory that the most accurate polynomial can be constructed using a set of deterministic sampling points, if the relationship between inputs and outputs satisfies certain requirements on smoothness [23].…”

Section: Surrogate Model Based On Adaptive Polynomial Approximation M...mentioning

confidence: 99%

Multi‐objectives optimization of parameter design for LLC converter based on data‐driven surrogate model

Zhou

Zhang

Luo

et al. 2023

IET Power Electronics

Self Cite

View full text Add to dashboard Cite

Due to high computation complexity, traditional design methods for LLC converter usually consider limited kinds of performance, and the process for searching the optimal parameter scheme is discrete, which might cause the missing of real optimal solution. In order to solve these problems, a multi‐objectives optimization design method for the LLC converter is proposed in this paper. According to the numerical calculation method, a closed parameter space for resonant parameters is established under the constraints of operation mode, switching frequency, zero voltage switching (ZVS), and voltage stress of resonant capacitor. Besides, the root‐mean‐square (RMS) value of resonant current and secondary current, as well as the core loss of the transformer, are selected as the optimization objectives. Non‐dominated sorting genetic algorithm (NSGA‐II) is utilized to deal with these multi‐objective formulations. In order to reduce the computation complexity, a data‐driven method, called adaptive polynomial approximation (APA), is selected to obtain the explicit expressions of the parameter space and the optimization objectives. Then, by substituting this simplified surrogate model into the NSGA‐II algorithm, the optimal parameter scheme is obtained. The comprehensive comparison analysis is performed, and a 400 W/48 V experimental prototype was built to verify the theoretical analysis, which shows that the proposed method features the highest efficiency of 95%.

show abstract

“…The exact accuracy of the Kalman filtering approach 27 lies in an appropriate and valid model of the cells, but accurate cell models are difficult to obtain. Data-driven methods 28 are approaches for estimating SOC building on machine learning policies, which include methods of neural network (NN), 29 deep learning (DL) method, 30 support vector machine, 31 and fuzzy logic method. 32 Of all these, it is the neural network method that is the widest in use.…”

Section: Introductionmentioning

confidence: 99%

Novel Improved Particle Swarm Optimization-Extreme Learning Machine Algorithm for State of Charge Estimation of Lithium-Ion Batteries

Zhang

Wang

et al. 2022

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Incisively estimating the state of charge (SOC) of lithium-ion batteries is essential to ensure the safe and stable operation of a battery management system. Neural network methods do not depend on a specific lithium-ion battery model and are able to mirror the lithium-ion battery’s nonlinear relationships, thus receiving widespread attention; however, traditional neural network methods exhibit a long training time and low accuracy in estimating SOC. This paper presents an original algorithm of an improved particle swarm optimization (IPSO) extreme learning machine (ELM) neural network, improving the particle swarm algorithm using nonlinear inertia weights to enhance the global optimization seeking capability of ELM for solving the problem of poor precision of previous battery SOC estimation. The lithium-ion battery voltage and current are the input variables of the model, while SOC is used as the output variable. The results of the experiments revealed that the root-mean-square estimation errors of the proposed IPSO-ELM algorithm for SOC are within 0.31, 0.32, and 0.14% of the root mean square under the hybrid pulse power characteristic (HPPC), the Beijing bus dynamic stress test (BBDST), and the dynamic stress test (DST) operating conditions. Compared with the prediction results of the PSO-ELM and ELM neural networks, the simulation results prove that the SOC optimization method in this paper possesses superior precision and overcomes the shortcomings of traditional neural networks.

show abstract

A robust surrogate model of a solid oxide cell based on an adaptive polynomial approximation method

Cited by 17 publications

References 49 publications

Simplified Three-dimensional Models for Studying the Non-uniformity inside the Flat-chip Solid Oxide Cell Stacks

Simplified Three-dimensional Models for Studying the Non-uniformity inside the Flat-chip Solid Oxide Cell Stacks

Multi‐objectives optimization of parameter design for LLC converter based on data‐driven surrogate model

Novel Improved Particle Swarm Optimization-Extreme Learning Machine Algorithm for State of Charge Estimation of Lithium-Ion Batteries

Contact Info

Product

Resources

About