scholarly journals Identifiability and physical interpretability of hybrid, gray-box models - a case study

2021 ◽  
Vol 54 (3) ◽  
pp. 389-394
Author(s):  
M. Hotvedt ◽  
B. Grimstad ◽  
L. Imsland
Keyword(s):  
Box Models

ANN-RBF Hybrid Model for Spatiotemporal Estimation of Monthly Precipitation Case Study

2013 ◽  
Vol 4 (2) ◽  
pp. 1-16 ◽  
Author(s):  
Vahid Nourani ◽  
Ehsan Entezari ◽  
Peyman Yousefi
Keyword(s):  
Hybrid Model ◽  
Black Box ◽  
Monthly Precipitation ◽  
Study Region ◽  
Spatial Estimation ◽  
Box Models ◽  
Proposed Model ◽  
Temporal Estimation ◽  
Monthly Precipitation Data

For estimation of monthly precipitation, considering the intricacy and lack of accurate knowledge about the physical relationships, black box models usually are used because they produce more accurate values. In this article, a hybrid black box model, namely ANN-RBF, is proposed to estimate spatiotemporal value of monthly precipitation. In the first step a Multi Layer Perceptron (MLP) network is used for temporal estimation of monthly precipitation using the value of precipitation in previous months in the same gauging station. In the second step, Radial Basis Function (RBF) is used to estimate the value of precipitation in specific month and a spatial point within the study region, considering the value of monthly precipitation in other stations. In this regard, three commonly used RBFs’ Multi Quadric (MQ), Inverse Multi Quadric (IMQ) and Gaussian (Ga), are used for spatial estimation. Finally, the combination of these two steps leads to ANN-RBF hybrid model. The model is examined using monthly precipitation data of Ardabil plain located north western of Iran. All results show the reliable accuracy of ANN-RBF model for spatiotemporal estimation of precipitation. Furthermore, IMQ RBF yields more accurate results for spatial estimation in comparison with two other RBFs. The cross-validation scheme was also employed to validate the spatial estimation performance of the proposed model.

scholarly journals Detecting and modeling critical dependence structures between random inputs of computer models

2020 ◽  
Vol 8 (1) ◽  
pp. 263-297
Author(s):  
Nazih Benoumechiara ◽  
Nicolas Bousquet ◽  
Bertrand Michel ◽  
Philippe Saint-Pierre
Keyword(s):  
Computer Models ◽  
Dependence Structure ◽  
Uncertain Information ◽  
Industrial Case Study ◽  
Marginal Distributions ◽  
Limit Value ◽  
Box Models ◽  
Input Variables ◽  
Admissible Limit

AbstractUncertain information on input parameters of computer models is usually modeled by considering these parameters as random, and described by marginal distributions and a dependence structure of these variables. In numerous real-world applications, while information is mainly provided by marginal distributions, typically from samples, little is really known on the dependence structure itself. Faced with this problem of incomplete or missing information, risk studies that make use of these computer models are often conducted by considering independence of input variables, at the risk of including irrelevant situations. This approach is especially used when reliability functions are considered as black-box models. Such analyses remain weakened in absence of in-depth model exploration, at the possible price of a strong risk misestimation. Considering the frequent case where the reliability output is a quantile, this article provides a methodology to improve risk assessment, by exploring a set of pessimistic dependencies using a copula-based strategy. In dimension greater than two, a greedy algorithm is provided to build input regular vine copulas reaching a minimum quantile to which a reliability admissible limit value can be compared, by selecting pairwise components of sensitive influence on the result. The strategy is tested over toy models and a real industrial case-study. The results highlight that current approaches can provide non-conservative results.

scholarly journals Experiences from City-Scale Simulation of Thermal Grids

Resources ◽  
2021 ◽  
Vol 10 (2) ◽  
pp. 10
Author(s):  
Johan Simonsson ◽  
Khalid Tourkey Atta ◽  
Gerald Schweiger ◽  
Wolfgang Birk
Keyword(s):  
Renewable Energy Sources ◽  
District Heating ◽  
Lessons Learned ◽  
Spatial Behavior ◽  
Future Research ◽  
Heating And Cooling ◽  
Box Models ◽  
Future Research Directions ◽  
Lower Temperature

Dynamic simulation of district heating and cooling networks has an increased importance in the transition towards renewable energy sources and lower temperature district heating grids, as both temporal and spatial behavior need to be considered. Even though much research and development has been performed in the field, there are several pitfalls and challenges towards dynamic district heating and cooling simulation for everyday use. This article presents the experiences from developing and working with a city-scale simulator of a district heating grid located in Luleå, Sweden. The grid model in the case study is a physics based white-box model, while consumer models are either data-driven black-box or gray-box models. The control system and operator models replicate the manual and automatic operation of the combined heat and power plant. Using the functional mock-up interface standard, a co-simulation environment integrates all the models. Further, the validation of the simulator is discussed. Lessons learned from the project are presented along with future research directions, corresponding to identified gaps and challenges.

scholarly journals Modelling for Digital Twins—Potential Role of Surrogate Models

Processes ◽  
2021 ◽  
Vol 9 (3) ◽  
pp. 476
Author(s):  
Ágnes Bárkányi ◽  
Tibor Chován ◽  
Sándor Németh ◽  
János Abonyi
Keyword(s):  
Model Building ◽  
Surrogate Models ◽  
Uncertain Information ◽  
Industrial Case Study ◽  
Continuous Processes ◽  
Digital Twin ◽  
Box Models ◽  
Digital Twins

The application of white box models in digital twins is often hindered by missing knowledge, uncertain information and computational difficulties. Our aim was to overview the difficulties and challenges regarding the modelling aspects of digital twin applications and to explore the fields where surrogate models can be utilised advantageously. In this sense, the paper discusses what types of surrogate models are suitable for different practical problems as well as introduces the appropriate techniques for building and using these models. A number of examples of digital twin applications from both continuous processes and discrete manufacturing are presented to underline the potentials of utilising surrogate models. The surrogate models and model-building methods are categorised according to the area of applications. The importance of keeping these models up to date through their whole model life cycle is also highlighted. An industrial case study is also presented to demonstrate the applicability of the concept.

Increasing System Robustness Through a Utility-Based Analysis

2013 ◽  
Author(s):  
Benjamin Baxter ◽  
Richard Malak
Keyword(s):  
Design Approach ◽  
System Model ◽  
Mars Rover ◽  
External Perturbations ◽  
Complex Engineered Systems ◽  
Box Models ◽  
Functional Models ◽  
System Robustness ◽  
Engineered Systems

Robustness is an important aspect of complex engineered systems. However, ambiguity in its definition can generate uncertainty among engineers about how to be sure that it is accounted for in their design. In this paper, robustness is defined as a property that allows a system to maintain its functions against anticipated internal and external perturbations. The important aspect of this definition is the focus of robustness on a per-function basis. A modified utility-based design approach is presented that provides a step by step implementation to help engineers ensure their designed systems meet their preferences and contain robust characteristics. The approach focuses on generating functional models of the proposed system. The functional models provide designers with insight into which internal and external perturbations should be included within the system model. An example case study is included to illustrate the steps of the modified utility-based design approach. The case study examines the entry, descent and landing of a Mars rover. The system and subsystems are modeled using black box models and EMS (Energy, Material, Signal) function structures. This allows the relevant internal and external perturbations to be modeled in the system model. The case study shows how using a utility-based analysis can produce a robust system.

scholarly journals Hydrodynamic modeling of stratified estuary: case study of the Jadro River (Croatia)

2015 ◽  
Vol 63 (1) ◽  
pp. 29-37 ◽  
Author(s):  
Igor Ljubenkov
Keyword(s):  
Sea Water ◽  
Field Measurements ◽  
Box Model ◽  
Salt Intrusion ◽  
Flushing Time ◽  
Box Models ◽  
Split Field ◽  
Stratified Estuary ◽  
Set Up

Abstract The Jadro River with total length of 4.3 km and average annual discharge of 7.9 m3 s-1 is a relatively small river on the east coast of the Adriatic Sea, close to Split. Field campaign measurements were made to estimate salt intrusion in the Jadro estuary in July 2012. This measurement confirmed the stratified character of the estuary where fresh water flows in a thin layer over denser sea water. Furthermore, a numerical model was set up for simulating unsteady stratified flow without mixing between the layers. The model is applied for the Jadro River and field measurements are used for calibration. In addition, the steady state of stratification within the estuary is analyzed by a box model which assumes mixing between layers. Results of the numerical and the box models were compared. The flushing time estimated with the box model is approximately 1.5 day for summer steady conditions. Numerical analysis however shows that the residence time is much larger owing to flow unsteadiness

scholarly journals Improved day ahead heating demand forecasting by online correction methods

2019 ◽  
Author(s):  
Felix Bünning ◽  
Andrew Bollinger ◽  
Philipp Heer ◽  
Roy S. Smith ◽  
John Lygeros
Keyword(s):  
Machine Learning ◽  
Error Correction ◽  
Energy Demand ◽  
Demand Forecasting ◽  
Coefficient Of Determination ◽  
Average Error ◽  
Small Scale ◽  
Level Control ◽  
Box Models

To reduce the heating and cooling energy demand of buildings and districts novel control strategies are constantly being developed that require information on the future demand of the controlled entity. Demand forecasting is commonly done with deterministic white box models or fitted grey-box models, however, recently more and more data and machine learning based approaches are being developed. All approaches have weaknesses: white-box models require major modelling effort, grey-box approaches are limited by their model or parameter complexity and machine learning is dependent on hyperparameters, some of which are randomly chosen, and therefore considered unreliable. Here we develop a forecasting approach based on Artificial Neural Networks (ANN) and introduce error correction methods based on online learning and the learned autocorrelation of the forecasting error. We compare the approach to other regression based and grey-box methods in a real case study of a small-scale district energy system with mixed use and unknown lower-level control. We show that the proposed method outperforms the other forecasting methods in terms of average error and coefficient of determination. We further demonstrate that in our case study the error correction methods significantly reduce variance in ANN performance created by randomly initialized parameters in the networks.

scholarly journals Explainable Machine Learning Framework for Image Classification Problems: Case Study on Glioma Cancer Prediction

2020 ◽  
Vol 6 (6) ◽  
pp. 37
Author(s):  
Emmanuel Pintelas ◽  
Meletis Liaskos ◽  
Ioannis E. Livieris ◽  
Sotiris Kotsiantis ◽  
Panagiotis Pintelas
Keyword(s):  
Machine Learning ◽  
Image Classification ◽  
Prediction Accuracy ◽  
Classification Problems ◽  
Cancer Prediction ◽  
Learning Framework ◽  
Box Models ◽  
Black Box Models

Image classification is a very popular machine learning domain in which deep convolutional neural networks have mainly emerged on such applications. These networks manage to achieve remarkable performance in terms of prediction accuracy but they are considered as black box models since they lack the ability to interpret their inner working mechanism and explain the main reasoning of their predictions. There is a variety of real world tasks, such as medical applications, in which interpretability and explainability play a significant role. Making decisions on critical issues such as cancer prediction utilizing black box models in order to achieve high prediction accuracy but without provision for any sort of explanation for its prediction, accuracy cannot be considered as sufficient and ethnically acceptable. Reasoning and explanation is essential in order to trust these models and support such critical predictions. Nevertheless, the definition and the validation of the quality of a prediction model’s explanation can be considered in general extremely subjective and unclear. In this work, an accurate and interpretable machine learning framework is proposed, for image classification problems able to make high quality explanations. For this task, it is developed a feature extraction and explanation extraction framework, proposing also three basic general conditions which validate the quality of any model’s prediction explanation for any application domain. The feature extraction framework will extract and create transparent and meaningful high level features for images, while the explanation extraction framework will be responsible for creating good explanations relying on these extracted features and the prediction model’s inner function with respect to the proposed conditions. As a case study application, brain tumor magnetic resonance images were utilized for predicting glioma cancer. Our results demonstrate the efficiency of the proposed model since it managed to achieve sufficient prediction accuracy being also interpretable and explainable in simple human terms.

The Genesis and Internal Dynamics of El Salvador's People's Revolutionary Army, 1970–1976

2014 ◽  
Vol 38 (01) ◽  
pp. 102-129
Author(s):  
ALBERTO MARTÍN ÁLVAREZ ◽  
EUDALD CORTINA ORERO
Keyword(s):  
Latin America ◽  
El Salvador ◽  
Early Years ◽  
Left Wing ◽  
Internal Dynamics ◽  
Internal Coherence ◽  
The 1960S

AbstractUsing interviews with former militants and previously unpublished documents, this article traces the genesis and internal dynamics of the Ejército Revolucionario del Pueblo (People's Revolutionary Army, ERP) in El Salvador during the early years of its existence (1970–6). This period was marked by the inability of the ERP to maintain internal coherence or any consensus on revolutionary strategy, which led to a series of splits and internal fights over control of the organisation. The evidence marshalled in this case study sheds new light on the origins of the armed Salvadorean Left and thus contributes to a wider understanding of the processes of formation and internal dynamics of armed left-wing groups that emerged from the 1960s onwards in Latin America.

Sign in / Sign up

Export Citation Format

Share Document