Improving Temperature Prediction Accuracy Using Kalman and Particle Filtering Methods

This paper demonstrates a comparative analysis of five machine learning (ML) algorithms for improving the signal processing time and temperature prediction accuracy in Brillouin optical time domain analysis (BOTDA) fiber sensor. The algorithms analyzed were generalized linear model (GLM), deep learning (DL), random forest (RF), gradient boosted trees (GBT), and support vector machine (SVM). In this proof-of-concept experiment, the performance of each algorithm was investigated by pairing Brillouin gain spectrum (BGS) with its corresponding temperature reading in the training dataset. It was found that all of the ML algorithms have significantly reduced the signal processing time to be between 3.5 and 655 times faster than the conventional Lorentzian curve fitting (LCF) method. Furthermore, the temperature prediction accuracy and temperature measurement precision made by some algorithms were comparable, and some were even better than the conventional LCF method. The results obtained from the experiments would provide some general idea in deploying ML algorithm for characterizing the Brillouin-based fiber sensor signals.

Download Full-text

Prognostics of lithium-ion batteries based on different dimensional state equations in the particle filtering method

Transactions of the Institute of Measurement and Control ◽

10.1177/0142331216642836 ◽

2016 ◽

Vol 39 (10) ◽

pp. 1537-1546 ◽

Cited By ~ 4

Author(s):

Xiaohong Su ◽

Shuai Wang ◽

Michael Pecht ◽

Peijun Ma ◽

Lingling Zhao

Keyword(s):

Lithium Ion Batteries ◽

Particle Filtering ◽

Prediction Accuracy ◽

Lithium Ion ◽

Exponential Model ◽

State Equation ◽

Remaining Useful Life ◽

State Equations ◽

Battery Capacity ◽

Useful Life

Accurate prediction of the remaining useful life of lithium-ion batteries plays a significant role in various devices and many researchers have focused on lithium-ion battery reliability and prognosis. A particle filter (PF) is an effective filter for estimation and prediction of time series data where model structure is available. The prediction accuracy of a PF depends on two key factors: parameter initialization and the state equation. In this paper, parameters are estimated using a PF and two empirical exponential models, i.e. the exponential model and improved exponential model, are used to track the battery capacity degradation; each model uses a different state equation. Experiments were performed to compare prediction accuracy using the related parameters estimation model with that using the capacity decline model; this paper compares the effects of the different state equations on the lithium-ion battery remaining useful life prediction. The experimental results show the merits of the capacity decline model based on particle filtering. The capacity decline model PF is more suitable for estimating the battery capacity trend in the long term.

Download Full-text

Integrate Uncertainty in the Process of Prognostics for Electronics

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.69.132 ◽

2011 ◽

Vol 69 ◽

pp. 132-137

Author(s):

Yi Wang ◽

Ming Qing Xiao ◽

Jia Yong Fang

Keyword(s):

Particle Filtering ◽

Prediction Accuracy ◽

Damage Model ◽

Optimal Estimation ◽

Cumulative Damage ◽

Bayesian Filter ◽

Filter System ◽

Time Stress ◽

Dynamic Damage

Elements of uncertainty in the electronics Prognostics process were studied. A method for electronics Dynamic Damage Optimal Estimation and prognostics based on Particle Filtering were proposed. Under the effect of time stress, the electronics cumulative damage is the result of the continuous effect of the stress, as a result, a HMM based electronics dynamic damage model was built at first place, analytical results of uncertainties in the process of prognostics were given and thus a Bayesian based filter system was built. Bayesian Filter change the problem of uncertainty into an optimal estimation processes as a result, the optimal estimation was fetched by applying the particle filtering into the estimation. The experiment case study proved that the proposed method can eliminate the uncertainties caused by measurement and the system effectively and improve the RUL prediction accuracy.

Download Full-text

River ice and water temperature prediction on the Danube

Hungarian Geographical Bulletin ◽

10.15201/hungeobull.70.3.1 ◽

2021 ◽

Vol 70 (3) ◽

pp. 201-214

Author(s):

Zoltán Árpád Liptay ◽

◽

Szabolcs Czigány ◽

Ervin Pirkhoffer ◽

◽

...

Keyword(s):

Water Temperature ◽

Prediction Accuracy ◽

Ice Formation ◽

Coefficient Of Determination ◽

Mean Square ◽

Sensible Heat ◽

Weighted Mean ◽

Temperature Prediction ◽

Temperature Calculation

This paper presents a modification of the theory of weighted mean temperatures for rivers. Rodhe, B. (1952) assumed the dominance of sensible heat transfer on ice formation. We aimed to improve the method for the evaluation of ice and water temperature based on a relatively low number of inputs. We further developed the model by introducing the effect of pre-existing ice, hence increasing the accuracy of the model on the timing of ice disappearance. Prediction accuracy of ±1 day was reached for the timing of the appearance of ice. Additional outputs have also been added to the model, including the termination of ice and the prediction of water temperature. The temperature calculation had a coefficient of determination of 95 percent, and a root mean square error of 1.33 °C during the calibration period without the use of observed water temperatures. The validation was carried out in a forecasting situation, and the results were compared to the energy balance.

Download Full-text

Improved Particle Swarm Optimization for Sea Surface Temperature Prediction

Energies ◽

10.3390/en13061369 ◽

2020 ◽

Vol 13 (6) ◽

pp. 1369 ◽

Cited By ~ 1

Author(s):

Qi He ◽

Cheng Zha ◽

Wei Song ◽

Zengzhou Hao ◽

Yanling Du ◽

...

Keyword(s):

Time Series ◽

Particle Swarm Optimization ◽

Surface Temperature ◽

Sea Surface Temperature ◽

Prediction Accuracy ◽

Particle Swarm ◽

Sea Surface ◽

Temperature Prediction ◽

Swarm Optimization ◽

Improved Particle Swarm Optimization

The Sea Surface Temperature (SST) is one of the key factors affecting ocean climate change. Hence, Sea Surface Temperature Prediction (SSTP) is of great significance to the study of navigation and meteorology. However, SST data is well-known to suffer from high levels of redundant information, which makes it very difficult to realize accurate predictions, for instance when using time-series regression. This paper constructs a simple yet effective SSTP model, dubbed DSL (given its origination from methods known as DTW, SVM and LSPSO). DSL is based on time-series similarity measure, multiple pattern learning and parameter optimization. It consists of three parts: (1) using Dynamic Time Warping (DTW) to mine the similarities in historical SST series; (2) training a Support Vector Machine (SVM) using the top-k similar patterns, deriving a robust SSTP model that offers a 5-day prediction window based on multiple SST input sequences; and (3) developing an improved Particle Swarm Optimization (PSO) method, dubbed LSPSO, which uses a local search strategy to achieve the combined requirement of prediction accuracy and efficiency. Our method strives for optimal model parameters (pattern length and interval step) and is suited for long-term series, leading to significant improvements in SST trend predictions. Our experimental validation shows a 16.7% reduction in prediction error, at a 76% gain in operating efficiency. We also achieve a significant improvement in prediction accuracy of non-stationary SST time series, compared to DTW, SVM, DS (i.e., DTW + SVM), and a recent deep learning method dubbed Long-Short Term Memory (LSTM).

Download Full-text

Particle Filtering approach to parameter estimate and temperature prediction of satellite

2008 7th World Congress on Intelligent Control and Automation ◽

10.1109/wcica.2008.4593401 ◽

2008 ◽

Cited By ~ 1

Author(s):

Pang Li-ping ◽

Qu Hongquan

Keyword(s):

Particle Filtering ◽

Parameter Estimate ◽

Temperature Prediction ◽

Filtering Approach

Download Full-text

A Closed-Form Solution for Temperature Profiles in Selective Laser Melting of Metal Additive Manufacturing

Materials Science Forum ◽

10.4028/www.scientific.net/msf.982.98 ◽

2020 ◽

Vol 982 ◽

pp. 98-105

Author(s):

Steven Y. Liang ◽

Jin Qiang Ning ◽

Elham Mirkoohi

Keyword(s):

Finite Element ◽

Heat Source ◽

Closed Form ◽

Heat Loss ◽

Heat Sink ◽

Prediction Accuracy ◽

Closed Form Solution ◽

Form Solution ◽

Laser Melting ◽

Temperature Prediction

This paper presents a closed-form solution for the temperature prediction in selective laser melting (SLM). This solution is developed for the three-dimensional temperature prediction with consideration of heat input from a moving laser heat source, and heat loss from convection and radiation on the part top boundary. The consideration of heat transfer boundary condition and latent heat in the closed-form solution leads to an improvement on the understanding of thermal development and prediction accuracy in SLM, and thus the usefulness of the analytical model in the temperature prediction in real applications. A moving point heat source solution is used to calculate the temperature rise due to the heat input. A heat sink solution is used to calculate the temperature drop due to heat loss from convection and radiation on the part boundary. The heat sink solution is modified from a heat source solution with equivalent power due to heat loss from convection and radiation, and zero-moving velocity. The temperature solution is then constructed from the superposition of the linear heat source solution and linear heat sink solution. Latent heat is considered using a heat integration method. Ti-6Al-4V is chosen to test the presented model with the assumption of isotropic and homogeneous material. The predicted molten pool dimensions are compared to the documented values from the finite element method and experiments in the literature. The presented model has improved prediction accuracy and significantly higher computational efficiency compared to the finite element model.

Download Full-text