Nonlinear Interrelationship Research Articles

A study of the nonlinear dynamic interrelationship between CO2 emissions and logistics sector output growth

Using provincial panel data spanning from 1995 to 2020, this paper examines the nonlinear interrelationship between logistics carbon dioxide emissions and output growth. For this purpose, we conduct a nonlinear co-integration test in heterogeneous panels; our results suggests a long-run relationship between these two variables. In contrast, adjustment to the equilibrium is inherently nonlinear. Furthermore, we estimate a panel smooth transition vector error correction model. Our major contributions, as we know, are the first study confirming the existence of a dynamic mechanism between logistics carbon dioxide emissions and regional output growth and the Environmental Kuznets Curve in China. Last, the interaction between logistics carbon dioxide emissions and regional output growth varies over the output growth in Chinese provinces and autonomous regions. Our results highlight the importance of considering possible nonlinearities in analyzing output-carbon dioxide emissions causality nexus and designing energy policies.

A deep reinforcement learning assisted simulated annealing algorithm for a maintenance planning problem

Maintenance planning aims to improve the reliability of assets, prevent the occurrence of asset failures, and reduce maintenance costs associated with downtime of assets and maintenance resources (such as spare parts and workforce). Thus, effective maintenance planning is instrumental in ensuring high asset availability with the minimum cost. Nevertheless, to find such optimal planning is a nontrivial task due to the (i) complex and usually nonlinear inter-relationship between different planning decisions (e.g., inventory level and workforce capacity), and (ii) stochastic nature of the system (e.g., random failures of parts installed in assets). To alleviate these challenges, we study a joint maintenance planning problem by considering several decisions simultaneously, including workforce planning, workforce training, and spare parts inventory management. We develop a hybrid solution algorithm (DRLSA\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\mathcal {DRLSA}$$\\end{document}) that is a combination of Double Deep Q-Network based Deep Reinforcement Learning (DRL) and Simulated Annealing (SA) algorithms. In each episode of the proposed algorithm, the best solution found by DRL is delivered to SA to be used as an initial solution, and the best solution of SA is delivered to DRL to be used as the initial state. Different from the traditional SA algorithms where neighborhood structures are selected only randomly, the DRL part of DRLSA\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\mathcal {DRLSA}$$\\end{document} learns to choose the best neighborhood structure to use based on experience gained from previous episodes. We compare the performance of the proposed solution algorithm with several well-known meta-heuristic algorithms, including Simulated Annealing, Genetic Algorithm (GA), and Variable Neighborhood Search (VNS). Further, we also develop a Machine Learning (ML) algorithm (i.e., K-Median) as another benchmark in which different properties of spare parts (e.g., failure rates, holding costs, and repair rates) are used as clustering features for the ML algorithm. Our study reveals that the DRLSA\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\mathcal {DRLSA}$$\\end{document} finds the optimal solutions for relatively small-size instances, and it has the potential to outperform traditional meta-heuristic and ML algorithms.

Civil airline fare prediction with a multi-attribute dual-stage attention mechanism

Airfare price prediction is one of the core facilities of the decision support system in civil aviation, which includes departure time, days of purchase in advance and flight airline. The traditional airfare price prediction system is limited by the nonlinear interrelationship of multiple factors and fails to deal with the impact of different time steps, resulting in low prediction accuracy. To address these challenges, this paper proposes a novel civil airline fare prediction system with a Multi-Attribute Dual-stage Attention (MADA) mechanism integrating different types of data extracted from the same dimension. In this method, the Seq2Seq model is used to add attention mechanisms to both the encoder and the decoder. The encoder attention mechanism extracts multi-attribute data from time series, which are optimized and filtered by the temporal attention mechanism in the decoder to capture the complex time dependence of the ticket price sequence. Extensive experiments with actual civil aviation data sets were performed, and the results suggested that MADA outperforms airfare prediction models based on the Auto-Regressive Integrated Moving Average (ARIMA), random forest, or deep learning models in MSE, RMSE, and MAE indicators. And from the results of a large amount of experimental data, it is proven that the prediction results of the MADA model proposed in this paper on different routes are at least 2.3% better than the other compared models.

Modified Principal Component Analysis for Identifying Key Environmental Indicators and Application to a Large-Scale Tidal Flat Reclamation

Identification of the key environmental indicators (KEIs) from a large number of environmental variables is important for environmental management in tidal flat reclamation areas. In this study, a modified principal component analysis approach (MPCA) has been developed for determining the KEIs. The MPCA accounts for the two important attributes of the environmental variables: pollution status and temporal variation, in addition to the commonly considered numerical divergence attribute. It also incorporates the distance correlation (dCor) to replace the Pearson’s correlation to measure the nonlinear interrelationship between the variables. The proposed method was applied to the Tiaozini sand shoal, a large-scale tidal flat reclamation region in China. Five KEIs were identified as dissolved inorganic nitrogen, Cd, petroleum in the water column, Hg, and total organic carbon in the sediment. The identified KEIs were shown to respond well to the biodiversity of phytoplankton. This demonstrated that the identified KEIs adequately represent the environmental condition in the coastal marine system. Therefore, the MPCA is a practicable method for extracting effective indicators that have key roles in the coastal and marine environment.

Artificial Intelligence Techniques for Predicting and Mapping Daily Pan Evaporation

In this study, Artificial Intelligence techniques such as Artificial Neural Network (ANN), Model Tree (MT) and Genetic Programming (GP) are used to develop daily pan evaporation time-series (TS) prediction and cause-effect (CE) mapping models. Ten years of observed daily meteorological data such as maximum temperature, minimum temperature, relative humidity, sunshine hours, dew point temperature and pan evaporation are used for developing the models. For each technique, several models are developed by changing the number of inputs and other model parameters. The performance of each model is evaluated using standard statistical measures such as Mean Square Error, Mean Absolute Error, Normalized Mean Square Error and correlation coefficient (R). The results showed that daily TS-GP (4) model predicted better with a correlation coefficient of 0.959 than other TS models. Among various CE models, CE-ANN (6-10-1) resulted better than MT and GP models with a correlation coefficient of 0.881. Because of the complex non-linear inter-relationship among various meteorological variables, CE mapping models could not achieve the performance of TS models. From this study, it was found that GP performs better for recognizing single pattern (time series modelling), whereas ANN is better for modelling multiple patterns (cause-effect modelling) in the data.

RBFNN Versus Empirical Models for Lag Time Prediction in Tropical Humid Rivers

Lag time (Lt) reflects the speed at which a river basin responds to rainfall (RF) events and is influenced by many hydrological parameters such as RF and stream flow (SF). These two parameters are represented by four variables, namely peak RF intensity, previous 48-h rainfall, peak SF and previous 48-h SF. In fact, lag time is highly stochastic in nature and its relation with the mentioned four variables is highly nonlinear interrelationship. The main objective of this study is to develop a model to estimate the Lt between upstream and downstream stations in tropical humid rivers. The graphical hydrological approach (HGA) has been used to estimate the Lt based on 95 RF-SF and considered as the references value for the proposed model evaluation. Linear, non-linear and Radial Basis Function Neural Network (RBFNN) methods have been developed successfully for Selangor River basin. The results show that the RBFNN outperformed the linear and the non-linear model and could achieve correlation coefficient (r) between the observed Lt and predicted Lt equal to 0.979 while r for the linear and the non-linear model equal to 0.519 and 0.631, receptively. Furthermore, the RBFNN model could attain minimum root mean square error (RMSE) between the observed Lt and predicted Lt equal to 1.23 while RMSE for the linear and the non-linear model equal to 1.9 and 2.02, receptively. The proposed RBFNN model significantly abridges the estimation of Lt values and avoids the essential need for comprehensive description of all parameters affecting on its value.

Optimizing Fracture Design Based on a Novel RBF Neural Network

The factors affecting performance of fractured wells are analyzed in this work. The static and dynamic geologic data of fractured well and fracturing treatment parameters obtained from 51 fractured wells in sand reservoirs of Zhongyuan oilfield are analyzed by applying the grey correlation method. Ten parameters are screened, including penetrability, porosity, net thickness, oil saturation, water cut, average daily production, and injection rate, amount cementing front spacer, amount sand-carrying agent and amount sand. With the novel Radial Basis Function neural network model based on immune principles, 13 parameters of 42 wells out of 51 are used as the input samples and the stimulation ratios as the output samples. The nonlinear interrelationship between the input samples and output samples are investigated, and a productivity prediction model of optimizing fracture design is established. The data of the rest 7 wells are used to test the model. The results show that the relative errors are all less than 7%, which proves that the novel Radial Basis Function neural network model based on immune principles has less calculation, high precision and good generalization ability.

A Nonlinear Interrelationship Between Period Length and the Amount of Activity - Age-Dependent changes

The circadian activity rhythm undergoes changes in the course of postnatal development. Experiments without external time cues were performed to characterize the endogenous component and to investigate any age-dependent changes. Female laboratory mice were used. At the beginning of the experiment they were 3 (juvenile), 23 (adult) or 72 (senile) weeks old. Animals were kept in climatic chambers (constant darkness, food and water ad libitum, temperature: 22±2°C, rel. humidity: 55±5%). Locomotor activity was recorded continuously using infrared detectors. The data were stored and analysed by means of the “Chronobiology Kit” (Stanford University). The mean period lengths were not statistically different between age groups. The stability of the spontaneous activity rhythms was highest in adult mice, however. The mean activity/day decreased from juvenile to senile mice. A nonlinear interrelationship between period length and amount of activity was obtained. At lower activity levels the period length became shorter with increasing activity; at higher levels it became longer again. The general shape of the curve was similar in all age groups. With respect to the nonlinear curve, one could not establish a general age dependency of period length. At similar ranges of activity the period length would be shortest in senile animals. Taking into account, however, the decline with age of the amount of activity the period of old mice could be shorter than, equal to or longer than that of adult mice. The results show that the endogenous component of the circadian activity rhythm, including feedback loops, matures and stabilizes from the juvenile to the adult. An expected loss of stability in senile mice was not demonstrated, probably due to a high variance of the animals’ biological age. These age-dependent changes contribute to the changes of circadian activity rhythms obtained under entrained conditions.

Regulation of mitochondrial resting state respiration: Slip, leak, heterogeneity?

The contribution of molecular slippage of proton pumps, of proton leak and of coupling heterogeneity of mitochondrial population to the well-known non-linear interrelationship between resting state respiration and the protonmotive force is discussed in view of the following experimental findings. (1) After blocking mitochondrial respiration with cyanide, the rate of dissipation of the membrane potential is non-linearly dependent on the actual membrane potential, similarly to the resting state respiration in mitochondria titrated with small amounts of an inhibitor. In contrast, ΔpH dissipates proportionally to its actual value. (2) The rate of electron flow from succinate to ferricyanide depends upon the protonmotive force, similarly to the flow from succinate to oxygen. This strongly suggests that the H +/e − stoichiometry in complexes III and IV of the respiratory chain is constant. (3) Mitochondria ‘in situ’, in permeabilized Ehrlich ascites cells, exhibit the same non-linear flux/force relationship as isolated mitochondria. These results strongly suggest that the non-ohmic characteristics of the inner mitochondrial membrane, with respect to protons driven by the membrane potential but not by the concentration gradient, is the main factor responsible for the nonlinear flux/force relationship in resting state mitochondria.

Stokes particle in an inhomogeneous turbulent stream

The motion of a small (Stokes) particle suspended in an inhomogeneous turbulent, incompressible viscous stream is considered It is shown that the nonlinear interrelationship between the disordered vibrational and translational particle motions in an inhomogeneous field of turbulent pulsations specifies considerable systematic particle displacement and the transformation of the turbulent medium motion into directed motion. An estimate is made of the effects for a particle suspended in a medium whose pulsation field is stationary and varies in one direction.