• Energy Research
• Research software close
• AT close
• TH close

This is a source code for a study to appear in Applied Soft Computing. Abstract: non-linear regression is the primary tool for estimating kinetic models of chemical reactions. The default approach of minimizing the sum of squared residuals tends to underperform in the presence of systematic errors, non-normal distribution of residuals or identifiability issues such as a high correlation between parameters. Therefore, we argue for a careful choice of the fit criteria and propose new, concave loss functions. Together with regularization, they form a robust objective for the regression procedure. Discussion of the rationale behind the proposed approach and its effects is illustrated by laboratory data on the transesterification of palm oil. A dedicated simulation study complements qualitative examples. All of the top-performing methods use regularization. Concave loss functions were among the best in 6-7 out of 8 test cases, compared to 2-3 for the classical square loss confirming both statistical and practical usefulness of the novel fit criteria. This result holds for a variety of modern optimizers. In 76% of our simulations, we obtained results not significantly worse than the best, whereas methods currently used in the literature provide 38% for the relative and 0% for the square loss.

Software scripts required to generate the SECURES-Met dataset, divided into different use-cases. Generated SECURES-Met is available under https://doi.org/10.5281/zenodo.7907883 The project SECURES, in which this software was produced, was funded by the Climate and Energy Fund (Klima- und Energiefonds) under project number KR19AC0K17532.

1. The fasting endurance hypothesis (FEH) predicts strong selection for large body size in mammals living in environments where food supply is interrupted over prolonged periods of time. The Arctic is a highly seasonal and food restricted environment, but contrary to predictions from the FEH, empirical evidence shows that Arctic mammals are often smaller than their temperate conspecifics. Intraspecific studies integrating physiology and behaviour of different-sized individuals, may shed light on this paradox. 2. We tested the FEH in free-living Svalbard reindeer (Rangifer tarandus platyrhynchus). We measured daily energy expenditure (DEE), subcutaneous body temperature (Tsc) and activity levels during the late winter in 14 adult females with body masses ranging from 46.3 to 57.8 kg. Winter energy expenditure (WEE) and fasting endurance (FE) were modelled dynamically by combining these data with body composition measurements of culled individuals at the onset of winter (14 years, n = 140) and variation in activity level throughout winter (10 years, n = 70). 3. Mean DEE was 6.3±0.7 MJ day−1. Lean mass, Tsc and activity had significantly positive effects on DEE. Across all 140 individuals, mean FE was 85±17 days (range 48–137 days). In contrast to the predictions of the FEH, the dominant factor affecting FE was initial fat mass, while body mass and FE were not correlated. Furthermore, lean mass and fat mass were not correlated. FE was on average 80% (45 days) longer in fat than lean individuals of the same size. Reducing activity levels by ~16% or Tsc by ~5% increased FE by 7%, and 4%, respectively. 4. Our results fail to support the FEH. Rather, we demonstrate that (i) the size of fat reserves can be independent of lean mass and body size within a species, (ii) ecological and environmental variation influence FE via their effects on body composition, and (iii) physiological and behavioural adjustments can improve FE within individuals. Altogether, our results suggest that there is a selection in Svalbard reindeer to accumulate body fat, rather than to grow structurally large. The methods used to collect the data are described in Trondrud et al. 2021: "Fat storage influences fasting endurance more than body size in an ungulate. Accepted in Functional Ecology. Funding provided by: Norges ForskningsrådCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100005416Award Number: KLIMAFORSK 267613

This NetLogo model illustrates the cultural evolution of pro-environmental behaviour patterns. It illustrates how collective behaviour patterns evolve from interactions between agents and agents (in a social network) as well as agents and the affordances (action opportunities provided by the environment) within a niche. More specifically, the cultural evolution of behaviour patterns is understood in this model as a product of: 1. The landscape of affordances provided by the material environment, 2. Individual learning and habituation, 3. Social learning and network structure, 4. Personal states (such as habits and attitudes), and 5. Cultural niche construction, or the modulation of affordances within a niche. More particularly, the model illustrates how changes in the landscape of affordances can trigger nonlinear changes in collective behaviour patterns. Even linear changes in affordances can trigger nonlinear uptakes of collective behaviour patterns. The model also shows how several behavioural cultures can emerge from the same environment and even within the same network. The model is an elaboration of Kurt Lewin’s heuristic equation, *B = f(P, E)*, where behaviour (B) is a function (f) of the person (P) and the environment (E). The model introduces several feedback loops (1–5 above) to Lewin’s equation, and thus provides an entry-point into studying the evolution of dynamical and complex behavioural systems over time. The model should be considered an abstract model, since many of its parameters are unspecifiable due to limits to current understanding of human (social) behaviour. However, the model can be tuned to replicate real-world macro patterns, and can be used as a sandbox environment to locate tipping points in social systems. See ODD protocol (comes with download).

Software repository for the paper: Thanheiser, S.; Haider, M. Particle Mass Diffusion Model for Level Control of Bubbling Fluidized Beds with Horizontal Particle Flow Powder Technology 2023

This repository belongs to the paper 'Pathway to a land-neutral expansion of Brazilian renewable fuel production'. The description of the data and a manual of how to install and use the simulation can be found in how-to.pdf Abstract Biofuels are currently the only available bulk renewable fuel. They have, however, limited expansion potential due to high land requirements and associated risks for biodiversity, food security, and land conflicts. We therefore propose to increase output from ethanol refineries in a land-neutral methanol pathway: surplus CO2-streams from fermentation are combined with H2 from renewably powered electrolysis to synthesize methanol. We illustrate this pathway with the Brazilian sugarcane ethanol industry using a spatio-temporal model. The fuel output of existing ethanol generation facilities can be increased by 43%-49% or 100TWh without using additional land. This amount is sufficient to cover projected growth in Brazilian biofuel demand in 2030. We identify a trade-off between renewable energy generation technologies: wind power requires the least amount of land whereas a mix of wind and solar costs the least. In the cheapest scenario, green methanol is competitive to fossil methanol at an average carbon price of 95€/tCO2.

This database studies the performance inconsistency on the biomass HHV proximate analysis. The research null hypothesis is the consistency in the rank of a biomass HHV model. Fifteen biomass models are trained and tested in four datasets. In each dataset, the rank invariability of these 15 models indicates the performance consistency. The database includes the datasets and source codes to analyze the performance consistency of the biomass HHV. These datasets are stored in tabular on an excel workbook. The source codes are the biomass HHV machine learning model through the MATLAB Objected Orient Program (OOP). These models consist of eight regressions, four supervised learnings, and three neural networks. An excel workbook, "BiomassDataSetProximate.xlsx," collects the research datasets in six worksheets. The first worksheet, "Proximate," contains 803 HHV data from 17 pieces of literature. The names of the worksheet column indicate the elements of the proximate analysis on a % dry basis. The HHV column refers to the higher heating value in MJ/kg. The following worksheet, "Full Residuals," backups the model testing's residuals based on the 20-fold cross-validations. The article verifies the performance consistency through these residuals. The other worksheets present the literature datasets implemented to train and test the model performance in many pieces of literature. A file named "SourceCodeProximate.rar" collects the MATLAB machine learning models implemented in the article. The list of the folders in this file is the class structure of the machine learning models. These classes extend the features of the original MATLAB's Statistics and Machine Learning Toolbox to support, e.g., the k-fold cross-validation. The MATLAB script, "runStudyProximate.m," is the article's main program (Kijkarncharoensin & Innet, 2021) to analyze the performance consistency of the biomass HHV model through the proximate analysis. The script instantly loads the datasets from the excel workbook and automatically fits the biomass model through the OOP classes. The first section of the MATLAB script generates the most accurate model by optimizing the model's higher parameters. It takes a few hours for the first run to train the machine learning model via the trial and error process. The trained models can be saved in MATLAB .mat file and loaded back to the MATLAB workspace. The remaining script, separated by the script section break, performs the residual analysis to inspect the performance consistency. Furthermore, the figure of the biomass data in the 3D scatter plot, and the box plots of the prediction residuals are exhibited. Finally, the interpretations of these results are examined in the author's article. Reference : Kijkarncharoensin, A., & Innet, S. (2021). The Performance Inconsistency of the Biomass Higher Heating Value (HHV) Models: The Proximate Analysis [Manuscript in preparation]. University of the Thai Chamber of Commerce.

This MATLAB capsule is an Objected Orient Program of a novel approach for the biomass HHV prediction model based on the proximate analysis. The proposed prediction methodology combines unsupervised and supervised learning to classify the biomass data in k regimes and independently fit the clustered data with the lasso, a shrinkage regression. Two feature sets, {FC,VM,ASH,HHV} and {VM/FC, ASH/VM, FC/ASH,HHV}, can be used as the predictors depending on the user requirements. The hold-out, k-fold, and resubstitution evaluate the model compared with the 12 literature models. In addition, the present capsule embeds with the literature datasets for the performance comparisons. Moreover, the presented code generates several graphical outputs to analyze the computational results, including clustering and residual analysis. Finally, the lasso shrinkage coefficients given on the specific lambda and the optimal ones are also exhibited in the capsule outputs. Reference [1.] Akara Kijkarncharoensin, “An unsupervised learning technique to classify the biomass thermal properties on the proximate analysis,” working paper, The University of the Thai Chamber of Commerce, Bangkok, Thailand. [2.] Akara Kijkarncharoensin, “A regime-switching biomass higher heating value lasso model based on the proximate analysis”, working paper, The University of the Thai Chamber of Commerce, Bangkok, Thailand.

This represents the second official release of Tools for Energy Model Optimization and Analysis (Temoa). It coincides with the release of the first Open Energy Outlook report.