Set Of Input-output Data Research Articles

Hydrogeological Experimental Site of Poitiers (France)

In hydrogeology, understanding the behavior of aquifers is an important issue. May it be for forecasting the availability of water on a long term or for estimating the impact of pollutants when injected at difierent regions, modeling of an aquifer is very useful for hydrogeologists. This paper gives an overlook of the difierent types of tests on a particularsite, namely the Hydrogeological Experimental Site (HES) of Poitiers (France). The aim is to show that the HES is a huge instrumented test bed allowing difierent types of experiments. Thus, system identification users can expect to gather large series of input-output datasets so as to experiment their own tools. Note that a large database exists concerning already-run tests, but what is interesting is that new experiments can be programmed so as to account for customized datasets. As a counterpart, the use of HES data should promote proper modeling of the underlying aquifer, thus contributing to the hydrological research field.

A Novel Approach to T-S Fuzzy Modeling of Nonlinear Dynamic Systems with Uncertainties using Symbolic Interval-Valued Outputs

A novel approach to Takagi-Sugeno (T-S) fuzzy modeling of a class of nonlineardynamic systems having variability in their outputs for the Nonlinear Output Error (NOE)case is addressed in this article. Multiple input-output datasets were obtained by repeating the identification experiment. The variability in the output time series is captured by defining the envelops of response at each time instant. These envelops actually provide the confidence interval based upper and lower bounds of the output time series using the extended Chebyshev’s Inequality. Different from the previous approach, in which two independent T-S fuzzy models were used for identifying each bound, a single T-S fuzzy model is identified in this work, which resulted in interval parameters for the antecedent and consequent variables. This is accomplished by first transforming the bounds into the symbolic interval-valued data and then using this data for identification. In order to get the expected value of the response, the estimated lower and upper bound time series of the identified T-S fuzzy model were averaged out at each time instant, as permitted by the extended Chebyshev’s Inequality. The proposed approach is demonstrated on an industrial diesel-engine electro-mechanical throttle valve.

DEVELOPING A MULTI-SCALE MULTI-REGION INPUT–OUTPUT MODEL

Many efforts have recently been devoted to developing global multi-region input–output (GMRIO) models. Unfortunately, the scales of GMRIO models do not allow them to capture the heterogeneity of regions within a single country. Multi-scale models can provide more comprehensive analyses capable of capturing the interdependencies of the global economy while preserving regional differences. The primary objective of this research is to develop methods for integrating multi-region input–output data sets from multiple spatial scales into multi-scale multi-region input–output (MSMRIO) models. These methods result in models that may have unusual features such as non-square trade coefficient matrices and a mix of industry-by-industry and commodity-by-commodity technical coefficients. To demonstrate the feasibility of MSMRIO modelling, a Canada-centric model was developed. This model includes 47 countries and Canada's 13 subnational regions. A MSMRIO model provides a tool to analyse global issues with a more spatially detailed focus.

Prediction of Squeeze Cast Density Using Fuzzy Logic Based Approaches

AbstractIn the present work, efforts are made to develop the input-output relationships for squeeze casting process by utilizing the fuzzy logic based approaches. Casting density in Squeeze casting is expressed as function of process parameters, such as time delay before pressurizing the metal, pressure durations, squeeze pressure, pouring temperature and die temperature. It is to be noted that, Mamdani based model and Takagi and Sugeno's model have been developed to model density in squeeze casting process. Manually constructed Mamdani based fuzzy logic controller and Takagi and Sugeno's based fuzzy logic controller have been used in approach 1 and approach 2 respectively. Training of FLC is carried with the help of five hundred input-output data set generated artificially through regression equations, obtained earlier by the same authors. The performance of the developed models was tested for both the linear and non-linear membership function distributions with the help of ten test cases. Moreover, the test data was collected by conducting the experiments and not used in training of FLCs. It is interesting to note that both approaches are capable to make accurate predictions. However, the performance of approach 2 with G bell shape membership function distribution is found to outperform approach 1 and other type of membership function distributions. The findings are useful to foundry-men, since it provides information on casting density in squeeze casting process for the different combination of process parameters without conducting any experiments.

Modelling and prediction of complex non-linear processes by using Pareto multi-objective genetic programming

In this paper, a new multi-objective genetic programming (GP) with a diversity preserving mechanism and a real number alteration operator is presented and successfully used for Pareto optimal modelling of some complex non-linear systems using some input–output data. In this study, two different input–output data-sets of a non-linear mathematical model and of an explosive cutting process are considered separately in three-objective optimisation processes. The pertinent conflicting objective functions that have been considered for such Pareto optimisations are namely, training error (TE), prediction error (PE), and the length of tree (complexity of the network) (TL) of the GP models. Such three-objective optimisation implementations leads to some non-dominated choices of GP-type models for both cases representing the trade-offs among those objective functions. Therefore, optimal Pareto fronts of such GP models exhibit the trade-off among the corresponding conflicting objectives and, thus, provide different non-dominated optimal choices of GP-type models. Moreover, the results show that no significant optimality in TE and PE may occur when the TL of the corresponding GP model exceeds some values.

Comparative Study of R&D Spillover Effect for High Technology Industry and Traditional Industry

Based on economic distance weights matrix, using the R&D input-output panel data sets of high technological industry and traditional industry in 29 provinces in China from 2001 to 2010 and spatial econometric method,this paper analyzes their R&D spillover effects respectively between provinces. The results indicate: Even though R&D in the high technology industry in China are active, the R&D spillover in this industry is lower than that in the traditional industries; R&D capital contributes to R&D output more than R&D labor for high technological industry, while the traditional industry is just the opposite.

Soft Computing Methods for Thermo-Acoustic Simulation

In the present study, soft computing methods are employed for thermoacoustic simulation. A ducted Burke-Schumann diffusion flame is used as the heat source for a horizontal duct. First, a dynamic model is constructed from the input-output data sets (velocity forcing - heat release) generated from the Burke-Schumann flame using Comsol. An efficient and cheap model of heat source is obtained using dynamic fuzzy identification. The full thermoacoustic system is simulated in a time domain with the Galerkin method using the identified heat source model. Finally, dynamic neural networks are utilized for obtaining a dynamic fit for a set of operating conditions for the acoustic velocity at the heater location. The overall agreement between the outputs of the soft computing tools (fuzzy and neural network tools) with the Comsol and Galerkin solver is found to be satisfactory.

Genetic Programming in Groundwater Modeling

AbstractDetermination of water-table elevation corresponding to aquifer recharge or discharge is an important issue in sustainable groundwater management. This approach necessitates the application of numerical simulation models that may require substantial amounts of input data related to aquifer parameters and specifications, even for modeling only part of the aquifer, which makes the calculations expensive. Moreover, comprehensive aquifer modeling is a time-consuming and computationally intensive process. Artificial intelligence tools can replace simulation models and decrease computational efforts by using input and output data sets without considering complex relations of the system to be modeled. This paper employs an adaptive neural fuzzy inference system (ANFIS) and genetic programming (GP) as artificial intelligence tools to extract governing groundwater flow equations in Ghaen and Karaj aquifers in Iran. For both aquifers, several input-output data sets, for both training and testing data sets, ...

Hydrological model calibration for derived flood frequency analysis using stochastic rainfall and probability distributions of peak flows

Abstract. Derived flood frequency analysis allows the estimation of design floods with hydrological modeling for poorly observed basins considering change and taking into account flood protection measures. There are several possible choices regarding precipitation input, discharge output and consequently the calibration of the model. The objective of this study is to compare different calibration strategies for a hydrological model considering various types of rainfall input and runoff output data sets and to propose the most suitable approach. Event based and continuous, observed hourly rainfall data as well as disaggregated daily rainfall and stochastically generated hourly rainfall data are used as input for the model. As output, short hourly and longer daily continuous flow time series as well as probability distributions of annual maximum peak flow series are employed. The performance of the strategies is evaluated using the obtained different model parameter sets for continuous simulation of discharge in an independent validation period and by comparing the model derived flood frequency distributions with the observed one. The investigations are carried out for three mesoscale catchments in northern Germany with the hydrological model HEC-HMS (Hydrologic Engineering Center's Hydrologic Modeling System). The results show that (I) the same type of precipitation input data should be used for calibration and application of the hydrological model, (II) a model calibrated using a small sample of extreme values works quite well for the simulation of continuous time series with moderate length but not vice versa, and (III) the best performance with small uncertainty is obtained when stochastic precipitation data and the observed probability distribution of peak flows are used for model calibration. This outcome suggests to calibrate a hydrological model directly on probability distributions of observed peak flows using stochastic rainfall as input if its purpose is the application for derived flood frequency analysis.

Robust PID TS fuzzy control methodology based on gain and phase margins specifications

This paper proposes the analysis and design of TS fuzzy robust PID control based on gain and phase margins specifications for uncertain nonlinear dynamic systems with time delay. The input-output data set of the uncertain nonlinear time delay dynamic system is decomposed into several input-output spaces by Gustafson-Kessel GK clustering algorithm, which are used to compute several linear submodels by least squares algorithm, and grouped in a Takagi-Sugeno TS fuzzy inference system. From the gain and phase margins specifications, in the frequency domain, analytical formulas are derived for fuzzy model based robust PID control design via Paralel and Distributed Compensation PDC strategy, considering the influence of the time delay. The main contribution of the paper is the proposal of two theorems related to necessary and sufficient conditions for fuzzy robust control design. The first theorem shows that the robust PID controller in the i-th rule of the fuzzy robust PID controller guarantees the gain and phase margins specifications for the corresponding linear model. The second theorem shows that the robust PID controller in the i-th rule of the fuzzy robust PID controller guarantees the stability for all linear models. A simulation example illustrates the efficiency of the fuzzy controller for control of a single link robotic manipulator when compared to others control methods. The experimental results for real-time fuzzy robust PID control of a termic process are obtained to demonstrate the effectiveness and practical viability of the proposed strategy.

Balancing virtual land imports by a shift in the diet. Using a land balance approach to assess the sustainability of food consumption. Germany as an example

Nutrition is considered as one of the main drivers of global environmental change. Dietary patterns in particular, embedded in the international trade of foods and other biomass based commodities, determine the dimension of beneficial or harmful environmental impacts of the agri-food sector – both domestically and abroad. In this study we analysed different dietary scenarios from a virtual land flow perspective, based on representative consumption data for Germany in the years 2006 and 1985–89. Further we identified the consumer groups that would have to adapt most to balance Germany’s virtual land import and analysed the impact reduced food wastage. For the study, official data sets concerning production, trade and consumption were used. We derived land use data from environmentally extended input–output data sets and FAO statistics. The conversion of agricultural raw products to consumed commodities is based on official processing and composition data. Subgroup-specific intake data from the last representative National Nutrition Survey in Germany were used. We analysed 42 commodities, aggregated into 23 product groups, seven land use types and six nutrition scenarios. The results show that in the baseline scenario the average nutrition in the year 2006 leads to a virtual land import of 707m2p−1a−1, which represents 30% of the total nutrition-induced land demand of 2365m2p−1a−1. On the other hand, the German agri-food sector exports virtual land, in the form of commodities, equivalent to 262m2p−1a−1. In this paper we calculate that the resulting net import of virtual land could be balanced by way of a shift to an officially recommended diet and a reduction in the consumption of stimulants (cocoa, coffee, green/black tea, wine). A shift to an ovo-lacto-vegetarian or vegan diet would even lead to a positive virtual land balance (even with maintained consumption of stimulants). Moreover, we demonstrate that a shift in the average diet profile could lead to maintained or even expanded export competitiveness and simultaneously enable environmental benefits. Since such a diet shift complies with official dietary recommendations, it follows that public health benefits may well result. We show further that a reduction of avoidable food losses/wastage would not be sufficient to level out the virtual land balance of the average nutrition in Germany. Regarding the dietary developments in the last 20years, we argue that a dietary shift resulting in a zero land balance is within reach. The population groups that would have to be addressed most are younger and middle-aged men. Nevertheless, women’s land saving potentials should not be ignored neither. Due to the fact that a western-style diet prevails in Germany, we argue that our basic findings are applicable to other industrialised and densely populated countries.

Probabilistic modelling and analysis of stand-alone hybrid power systems

As a part of the Hybrid Intelligent Algorithm, a model based on an ANN (artificial neural network) has been proposed in this paper to represent hybrid system behaviour considering the uncertainty related to wind speed and solar radiation, battery bank lifetime, and fuel prices. The Hybrid Intelligent Algorithm suggests a combination of probabilistic analysis based on a Monte Carlo simulation approach and artificial neural network training embedded in a genetic algorithm optimisation model. The installation of a typical hybrid system was analysed. Probabilistic analysis was used to generate an input–output dataset of 519 samples that was later used to train the ANNs to reduce the computational effort required. The generalisation ability of the ANNs was measured in terms of RMSE (Root Mean Square Error), MBE (Mean Bias Error), MAE (Mean Absolute Error), and R-squared estimators using another data group of 200 samples. The results obtained from the estimation of the expected energy not supplied, the probability of a determined reliability level, and the estimation of expected value of net present cost show that the presented model is able to represent the main characteristics of a typical hybrid power system under uncertain operating conditions.

Modeling and monitoring of nonlinear multi-mode processes

In this paper, a new subspace separation method is proposed and a new modeling and monitoring approach in multi-mode processes based on subspace separation is proposed. The existing modeling methods have the following disadvantages: (1) different monitoring models are built in view of each different mode, which needs each mode to be able to offer fully complete modeling reference data. (2) The connection between each mode is ignored, which could be useful in process modeling and monitoring. The proposed method has the following advantages: (1) the common subspace is extracted and the monitoring performance of multi-mode processes is significantly improved. (2) New subspace separation is used to establish an integrated monitoring system, which would simplify the monitoring model structure and enhance its reliability. (3) The direct relationship of input dataset and output dataset is considered in the multi-mode processes, which is crucial for the complex industry process. Experiment results show effectiveness of the proposed method.

Viewing Economy as a Network: An Exploration Through Input-Output Model

National Innovation System is a departure from viewing entrepreneurship as individual-agent based phenomenon. Viewing economy as a network system gives a perspective to make interventions for innovation at systemic level. In this paper, by identifying the network properties of the Indian economy, using input-output data sets, we provide evidence for co-movement of economic growth, employment and betweenness of the sectors in input-output network of Indian economy. We argue that by structurally repositioning the sectors of economy (by generating new ties or strengthening existing but weak ties), it is possible to create opportunities for innovation, which will have multiplier effects through chains of economy. By examining the structure of the networked Indian economic system, we identify integration of services particularly that of transportation services to be the primary challenge for growth and innovation.

Subspace identification analysis of RFX and T2R reversed-field pinches

Input–output datasets from two magnetic confinement fusion (MCF) experiments of the reversed-field pinch (RFP) type are examined. The RFP datasets, which are samples of the distributed magnetic field dynamics, are naturally divided into many smaller batches due to the pulsed-plasma operation of the experiments. The two RFP experiments considered are (i) EXTRAP T2R (T2R) with 64 inputs and 64 outputs and (ii) RFX-mod (RFX) with 192 inputs and 192 outputs. Both T2R and RFX are magnetohydrodynamically unstable and operates under magnetic feedback with optional dither injection. Using subspace system identification techniques and randomised cross-validation (CV) methods to minimise the generalisation error, state-space orders of the empirical systems are suggested. These system orders are compared to “stabilisation diagrams” commonly used in experimental modal analysis practice. The relation of the CV system order to the decay of the singular values from the subspace method is observed. Both (i) stable vacuum diffusion and (ii) unstable plasma response datasets are analysed. Apparent simulation and prediction errors are quantified for both cases using a deviation-accounted-for index. These results are purely data-driven. A simple approach towards exploitation of the subspace techniques for finite-element model refinement and data confrontation is presented.

Solar Power Prediction Using Interval Type-2 TSK Modeling

The random nature of solar energy resources is one of the obstacles to their large-scale proliferation in power systems. The most practical way to predict this renewable source of energy is to use meteorological data. However, such data are usually provided in a qualitative form that cannot be exploited using traditional quantitative methods but which can be modeled using fuzzy logic. This paper proposes type-1 and interval type-2 Takagi-Sugeno-Kang (TSK) fuzzy systems for the modeling and prediction of solar power plants. The paper considers TSK models with type-1 antecedents and crisp consequents, type-1 antecedents and consequents, and type-2 antecedents and crisp consequents. The design methodology for tuning the antecedents and consequents of each model is described. Finally, input-output data sets from a solar plant are used to obtain the three TSK models and their prediction results are compared to results from the literature. The results show that type-2 TSK models with type2 antecedents and crisp consequents provide the best performance based on the solar plant data.

An approach to fault diagnosis of nonlinear systems using neural networks with invariance to Fourier transform

A neural network with Gauss-Hermite polynomial activation functions is used for approximating the nonlinear system’s dynamics out of a set of input-output data. Thus the output of the neural network provides a series expansion that takes the form of a weighted sum of Gauss-Hermite basis functions. Knowing that the Gauss-Hermite basis functions satisfy the orthogonality property and remain unchanged under the Fourier transform, subjected only to a change of scale, one has that the considered neural network provides the spectral analysis of the output of the monitored system. Actually, the squares of the weights of the output layer of the neural network denote the distribution of energy into the associated spectral components for the output signal of the monitored nonlinear system. By observing changes in the amplitude of the aforementioned spectral components one can have also an indication about malfunctioning of the monitored system and can detect the existence of failures. Moreover, since specific faults are associated with amplitude changes of specific spectral components of the system fault isolation can be also performed.

Characterization of the Hysteresis Duhem Model

The Duhem model, widely used in structural, electrical and mechanical engineering, gives an analytical description of a smooth hysteretic behavior. In practice, the Duhem model is mostly used within the following black-box approach: given a set of experimental input-output data, how to tune the model so that its output matches the experimental data. It may happen that a Duhem model presents a good match with the experimental real data for a specific input, but does not necessarily keep significant physical properties which are inherent to the real data, independently of the exciting input. This paper presents a characterization of different classes of Duhem models in terms of their consistency with the hysteresis behavior.

Consistency of the Duhem Model with Hysteresis

The Duhem model, widely used in structural, electrical, and mechanical engineering, gives an analytical description of a smooth hysteretic behavior. In practice, the Duhem model is mostly used within the following black-box approach: given a set of experimental input-output data, how to tune the model so that its output matches the experimental data. It may happen that a Duhem model presents a good match with the experimental real data for a specific input but does not necessarily keep significant physical properties which are inherent to the real data, independent of the exciting input. This paper presents a characterization of different classes of Duhem models in terms of their consistency with the hysteresis behavior.

Extended Kalman Filter Based Learning Algorithm for Type-2 Fuzzy Logic Systems and Its Experimental Evaluation

In this paper, the use of extended Kalman filter for the optimization of the parameters of type-2 fuzzy logic systems is proposed. The type-2 fuzzy logic system considered in this study benefits from a novel type-2 fuzzy membership function which has certain values on both ends of the support and the kernel, and uncertain values on other parts of the support. To have a comparison of the extended Kalman filter with other existing methods in the literature, particle swarm optimization and gradient descent-based methods are used. The proposed type-2 fuzzy neuro structure is tested on different noisy input-output data sets, and it is shown that extended Kalman filter has a better performance as compared to the gradient descent-based methods. Although the performance of the proposed method is comparable with the particle swarm optimization method, it is faster and more efficient than the particle swarm optimization method. Moreover, the simulation results show that the proposed novel type-2 fuzzy membership function with the extended Kalman filter has noise rejection property. Kalman filter is also used to train the parameters of type-2 fuzzy logic system in a feedback error learning scheme. Then, it is used to control a real-time laboratory setup ABS and satisfactory results are obtained.