Search

Turbulence in two-phase flows drives many important natural and engineering processes, from geophysical flows to nuclear power generation. Strong interphase coupling between the carrier fluid and disperse phase precludes the use of classical turbulence models developed for single-phase flows. In recent years, there has been an explosion of machine learning techniques for turbulence closure modeling, though many rely on augmenting existing models. In this work, we propose an approach that blends sparse regression and gene expression programming (GEP) to generate closed-form algebraic models from simulation data. Sparse regression is used to determine a minimum set of functional groups required to capture the physics, and GEP is used to automate the formulation of the coefficients and dependencies on operating conditions. The framework is demonstrated on homogeneous turbulent gas-particle flows in which two-way coupling generates and sustains carrier-phase turbulence. [ABSTRACT FROM AUTHOR]

Copyright of Rudarsko-Geolosko-Naftni Zbornik is the property of Faculty of Mining, Geology & Petroleum Engineering and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Strength and deformational behaviour of non-persistent jointed rock mass are the most important issues in the field of rock mechanics. In this paper, many slab shaped rock-like blocks containing two pre-existing parallel rough non-persistent joints were subjected to the uniaxial compressive load in order to investigate the effect of joint roughness coefficient, length of bridge, angle of bridge and inclination of joint on its mechanical responses through physical tests. Based on the test results, two models i.e. gene expression programming and linear regression analysis have been developed, and some statistical indexes were adopted to evaluate their performance. The results revealed that the gene expression programming is far superior to linear regression method, and its results are in agreement with the lab tests outcomes; moreover, parametric study of the suggested models indicated that joint inclination and joint roughness coefficient have the highest and lowest effective on the tests outcomes respectively. [ABSTRACT FROM AUTHOR]

A portfolio selection model is developed in this study using a new risk measure. The proposed risk measure is based on the fundamental value of stocks. For this purpose, a mathematical model is developed and transformed into an integer linear programming. In order to analyze its efficiency, the actual data of the Tehran Stock Exchange market are used in 12 scenarios to solve the proposed model. In order to evaluate the scenarios, data mining approaches are employed. Data mining methods used in this paper include Adaptive Neuro-Fuzzy Inference System (ANFIS), decision tree, random forest, Fisher Discriminant Analysis (FDA), and Gene Expression Programming (GEP). The best method for scenario evaluation is GEP based on numerical results. Hence, the market values are evaluated by this algorithm. Software packages like MATLAB, GEP xpero tools, and LINGO are used to solve the model. Different trends of market value and fundamental value volatility in the optimum stock portfolio are determined. It is possible to examine the optimum portfolio profitability in different scenarios. By using real-world data, trends are extracted and analyzed. Results show that the developed model can be effectively applied in bubble situations. [ABSTRACT FROM AUTHOR]

Estimation of reference évapotranspiration (ETO) is vibrantly required for estimating crop water requirement and budgeting irrigation scheduling. The beneficial use of water is of great importance due to shortage issues especially in developing countries like Pakistan. The Food and agricultural organization (FAO) developed a Peneman-Montieth (PM) method which can be globally considered as a standard method for estimation of ETOj but it requires numerous climatic data. Consequently, there is a need to find out the next best suitable method after PM method. The Multi-layer perceptron (MLP), Gene expression programing (GEP) and Radial basis function (RBF) were utilized to calculate ETO values. Monthly meteorological data of six different stations located in arid, semi-arid and humid regions of Pakistan covered from 1980 to 2015. Seventeen input combinations comprise of various climatic variables were developed to evaluate the impact on ETO. Of the available meteorological data, 70% was employed in training while remaining 30% used in testing process. The yielded values of the developed models were compared with the ETO estimated by PM method. The outcome of the study was also applied on some other climatic regions located in USA, New Zealand and China for numerous duration only three climatic parameters, namely, maximum temperature, mean relative humidity and wind velocity had a large positive effect on increasing the accuracy of estimating ETO. By comparing the eight performing indices, MLP among all the powerful predictive modeling techniques can also be considered as the superior alternative to the conventional methods in estimation of ET0. [ABSTRACT FROM AUTHOR]

Automated guided vehicles (AGV) with different carrying capacities are required for complex material handling in smart factories, which causes resource waste. To minimize the delay and reduce the cost of logistics systems, this paper proposes a dynamic scheduling method for self-organized AGVs (SAGV) in production logistics systems, where multiple identical SAGVs can communicate and freely combine with others as one vehicle to perform one task. Using an improved gene expression programming to learning dynamic dispatching rules, experimental results show that dispatching rules learned are efficient and the cost of logistic systems by using SAGVs is significantly reduced. [ABSTRACT FROM AUTHOR]

Sheet piles are of general water-retaining structures. The rate of discharge ow beneath sheet plies is an important parameter in the design of these structures. In this study, Gene Expression Programming (GEP) as an artificial intelligence method is used for developing a model to predict the discharge ow rate. The input parameters include the sheet pile height, upstream head, and hydraulic conductivity anisotropy ratio. In order to achieve better performance, the ow rate is normalized and selected as an output of the model. A database including 1000 cases created from the Scaled Boundary Finite Element Method (SBFEM) for the seepage beneath sheet plies is employed to develop the model. The GEP-based model predictions are shown in reasonable agreement with the simulated data, which indicates the efficiency of the developed model. The results of the sensitivity analysis indicate that the upstream head is the most inuential parameter in the discharge ow rate beneath the sheet piles. Furthermore, the outputs of the parametric analysis show the reasonable performance of the model in the prediction of normalized discharge ow rate. [ABSTRACT FROM AUTHOR]

Leaf Area Index (LAI) is an important structural feature of our ecosystem as it affects energy, carbon, and water exchanges between the land surface and overlying atmosphere. Global scale LAI datasets have been obtained by regression, heuristic data driven, and radiative transfer models using remotely sensed land surface reflectance data. However, the estimation of LAI from remotely sensed data is limited only to clear sky conditions. Also, it is problematic to estimate LAI in forests by using conventional remote sensing image analysis of multi-spectral data. Due to the above-mentioned shortcomings of estimating LAI from remotely sensed data, this study obtained LAI from meteorological data using the Gene Expression Programming (GEP) technique. The new approach was tested in different forest sites with broad-leaf and needle-leaf trees in USA. The results showed that the GEP technique can accurately estimate LAI from meteorological data in different forest sites. [ABSTRACT FROM AUTHOR]

Mining data streams require to cope with time, data size and possible concept drift constraints. Even more challenging is the case where, apart from the above, one has to deal with imbalanced data. Mining non stationary and imbalanced data streams is a relatively new area of research. In this paper, we propose the Gene Expression Programming (GEP) classifier with drift detection and data reuse for mining imbalanced data streams. GEP is used to evolve a complex expression tree returning predictions. Drift detector role is to signal the occurrence of drift which triggers inducing a new learner. Data reuse mechanism allows for improving the balance between minority and majority instances in a subset of data used for evolving the learner. The proposed approach is validated experimentally. The experiment results confirm that our classifier produces high-quality predictions. [ABSTRACT FROM AUTHOR]

The process of project evaluation is of vital importance for decision-making in organizations. In the particular case of IT projects, the historical average of successful projects is 30.7%, while renegotiated projects are 47.3% and cancelled projects are 22% [1]. These figures mean that huge budgets are affected every year by errors in planning or control and monitoring of projects, with an economic and social impact. The objective of this research is to evaluate the MCGEP evolutionary algorithm in different versions databases with information on the evaluation of IT projects. The aim is to determine the possibility of applying an evolutionary algorithm that uses programming of genetic expressions as opposed to others of greater use. [ABSTRACT FROM AUTHOR]

We present a machine learning based method for RANS modeling in the rotating frame of reference (RFR). The extended intrinsic mean spin tensor (EIMST) is adopted in a novel expansion of the evolution algorithm, named multi-dimensional gene expression programming (MGEP). Based on DNS data, a constrain free model for Reynolds stress is created by considering system rotating. The anisotropy behavior of Reynolds stress is considered in the model, which is then for the first time applied for modeling turbulent flow inside a rotating channel. Compared with the traditional RANS model, the new model can predict the non-symmetric profile of Reynolds stress. Meanwhile, the Taylor-Görtler vortex is captured in our simulations with the new model. It is demonstrated that the application of EIMST in MGEP can be successfully adopted for RANS modeling in the RFR. [ABSTRACT FROM AUTHOR]

• Corrector design for dynamized coefficients in the k-ε model. • Equation discovery through gene expression programming. • Pareto improvement contributed by increased model flexibility. • Management of the uncertainty arising from SRANS ill-conditioning. • Robustness, redistributability, and improved generalizability. Urban airflow and pollution modeling using steady Reynolds-averaged Navier-Stokes (SRANS) with two-equation models has faced accuracy challenges, limiting its reliability for sustainable city design. The low accuracy has been attributed to SRANS ill-conditioning, the linear eddy viscosity hypothesis, and uncertainty contributed by empirical formulations and coefficients. Many studies have attempted to modify the two-equation models specifically for urban problems by correcting the model formulations and calibrating the coefficients. However, these corrections often improve accuracy for targeted regions or quantities while diminishing it elsewhere. This stems from inflexibility in the model form. To mitigate the trade-offs and improve generalizability, this study introduces multiple dynamic correctors to the k-ε coefficients for expanded design space. In particular, the robustness and redistributability of the corrected model were considered by retaining the theoretical interpretation of turbulence and employing symbolic correctors. The gene expression programming technique was employed for equation discovery. The increased model flexibility mitigated the trade-offs in accuracies for different quantities in the training case. The generalizability of a corrected model was evaluated for airflow and dispersion around a single building, a building array, and a group of complex buildings. The corrected model performed consistently for the three flow types and exhibited generalizability. [ABSTRACT FROM AUTHOR]

Accurate cost estimation in tunneling is key to the project's success. Such information is critical for the early conceptual and design phases when key choices must be made. Numerous variables influence the cost of tunnel construction, and limited information is available at the early stages of design when the possible use of tunnels is being studied. Therefore, there is a limited number of models at engineers' disposal to develop a proper cost estimate for tunneling projects. This study aimed to offer a model for estimating the construction cost of drilling and blasting tunnels in the preliminary stage of a project. For this purpose, an optimized gene expression programming (GEP) method was used based on the study of 900 data points obtained from ten drilling and blasting tunnels, which were randomly split into the training (800 data points) and testing (100 data points) datasets. With the experience of previously constructed tunnels, eleven parameters were considered the most effective for the tunnel's construction cost. The best fit on predictions generated an equation for the optimized GEP model. Finally, by comparing the equation's outputs with the actual costs and its behavior with practice, its potential ability to estimate the construction cost of drilling and blasting tunnels was approved. Moreover, the Graphical User Interface (GUI) of the GEP model was created as a practical tool for estimating the construction cost of tunnels. This model can reduce tunnel uncertainties and give ML development in tunnel planning. • Investigation of the GEP performance to predict the construction cost of tunnels. • Employment of 900 datasets in the modeling, including eleven inputs. • Optimization of the GEP model to generate an estimation equation of tunnel cost. • Comparison of the equation's behavior with practice. • Sensitivity analysis of the input parameters on the equation's output. [ABSTRACT FROM AUTHOR]

As an evolutionary algorithm for automatic generation of computer programs, Gene expression programming (GEP) has received a lot of attention because it can generate concise and highly accurate solutions. However, GEP also has some unsatisfactory aspects, such as too many control parameters and a lack of balance between exploration and exploitation. Therefore, a new GEP with dual strategies and neighborhood search (DNSGEP) is proposed in this paper for symbolic regression problems. As the core part, a new multi-parent crossover operator with neighborhood search is proposed to replace the original transposition and recombination operators in GEP, which greatly reduces the excessive control parameters in GEP. To assist the multi-parent crossover operator in selecting parents, selection weight is used to assess the quality of individuals in the population. The multi-parent crossover operator combines selection weight to learn more from the better parents. In order to satisfy the different needs of global search and local search of the algorithm, two different crossover strategies are extended on the basis of the multi-parent crossover operator. Furthermore, a population distribution entropy strategy to GEP is proposed for determining when the algorithm should choose which crossover strategy to use. DNSGEP is compared with GEP and three GEP variants on 14 symbolic regression problems and 6 even parity problems. Experimental results show that DNSGEP has the success rate and accuracy that are fully superior to GEP, GEPADF and OGEP, not inferior to SLGEP. • The proposed algorithm can obtain higher precision solutions for most of problems. • A multi-parent crossover operator with neighborhood search is proposed. • A dual strategies mechanism is proposed to select crossover strategies. [ABSTRACT FROM AUTHOR]

Many complex systems in the real world evolve with time. These dynamic systems are often modeled by ordinary differential equations in mathematics. The inverse problem of ordinary differential equations is to convert the observed data of a physical system into a mathematical model in terms of ordinary differential equations. Then the model may be used to predict the future behavior of the physical system being modeled. Genetic programming has been taken as a solver of this inverse problem. Similar to genetic programming, gene expression programming could do the same job since it has a similar ability of establishing the model of ordinary differential systems. Nevertheless, such research is seldom studied before. This paper is one of the first attempts to apply gene expression programming for solving the inverse problem of ordinary differential equations. Based on a statistic observation of traditional gene expression programming, an improvement is made in our algorithm, that is, genetic operators should act more often on the dominant part of genes than on the recessive part. This may help maintain population diversity and also speed up the convergence of the algorithm. Experiments show that this improved algorithm performs much better than genetic programming and traditional gene expression programming in terms of running time and prediction precision. [ABSTRACT FROM AUTHOR]

The paper studies how to improve the trading decision of Taiwan stocks with the information of US stock market. Our method first aligns the trading days between Taiwan and US stock markets. Next, the similarity between the portfolio index (PI, constructed from 100 Taiwan stocks) and one of the US stock indices, the Dow Jones Industrial Average (DJIA), NASDAQ composite index (NASDAQ), or Standard & Poor’s 500 (S&P 500), is computed, respectively. The trading signals of PI or each US stock index are generated by the method of Lee et al. Finally, the consensus signals of PI are determined by the majority vote scheme with the weighted functions, calculated from the similarity. The testing period of PI starts from 2000/1/4 to 2017/12/29, totally 4480 days. As the experimental results show, the index combination (PI, DJIA, NASDAQ) with the weighted function W (4) is considered to be the best combination for trading PI. Its average annualized return (cumulative return) achieves 15.03% (1170.42%), which is better than the method of Lee et al. 13.88% (947.65%), and the buy-and-hold strategy 9.85% (442.90%). [ABSTRACT FROM AUTHOR]

Supplier evaluation and selection is a complicated multiple criteria decision-making process which affects supply chain management (SCM) directly. Recent studies emphasize that artificial intelligence approaches obtain better performance than conventional methods in evaluating the suppliers' performance and determining the best suppliers. Hence, this study proposes a new robust genetic-based intelligent approach, namely gene expression programming (GEP), to improve the supplier selection process for a supply chain and to cope with the drawback of the other intelligent approaches in this area. The applicability of this method was exhibited by a case study in the textile manufacturing industry. To show the performance of the mathematical-genetic model, comparisons with four intelligent techniques, namely multi-layer perceptron (MLP) neural network, radial basis function (RBF) neural network, adaptive neuro-fuzzy inference system (ANFIS) and support vector machine (SVM), were performed. The results derived from the intelligent approaches were compared by using a collected dataset from a textile factory. The obtained results demonstrated that first the GEP-based model provides a mathematical model for the suppliers' performance based on the determined criteria, and the developed GEP model is more accurate than the four other intelligent models in terms of accuracy in performance estimation. In addition, to verify the validity of the developed model, different statistical tests were used and the results showed that the GEP model is statistically powerful. [ABSTRACT FROM AUTHOR]

Designing suitable behavioral rules of agents so as to generate realistic behaviors is a fundamental and challenging task in many forms of computational modeling. This paper proposes a novel methodology to automatically generate a descriptive model, in the form of behavioral rules, from video data of human crowds. In the proposed methodology, the problem of modeling crowd behaviors is formulated as a symbolic regression problem and the self-learning gene expression programming is utilized to solve the problem and automatically obtain behavioral rules that match data. To evaluate its effectiveness, we apply the proposed method to generate a model from a video dataset in Switzerland and then test the generality of the model by validating against video data from the United States. The results demonstrate that, based on the observed movement of people in one scenario, the proposed methodology can automatically construct a general model capable of describing the crowd dynamics of another scenario in a different context (e.g., Switzerland vs. U.S.) as long as that the crowd behavior patterns are similar. [ABSTRACT FROM AUTHOR]

In this paper, four formulas are proposed via gene expression programming (GEP)-based models and regression analysis (RA) to predict the flexural strength (fs) values of mortars containing different mineral admixtures that are ground granulated blast-furnace slag (GGBFS), silica fume (SF) and fly ash (FA) at different ages. Three formulas obtained from the GEP-I, GEP-II and GEP-III models are constituted to predict the fs values from the age of specimen, water-binder ratio and compressive strength. Besides, one formula obtained from the RA is constituted to predict the fs values from the compressive strength. To achieve these formulas in the GEP and RA models, 972 data of the experimental studies presented with mortar mixtures were gathered from the literatures. 734 data of the experimental studies are divided without pre-planned for these formulas achieved from the training and testing sets of GEP and RA models. Beside, these formulas are validated with 238 data of experimental studies un-employed in training and testing sets. The fs results obtained from the training, testing and validation sets of these formulas are compared with the results obtained from the experimental studies and the formulas given in the literature for concrete. These comparisons show that the results of the formulas obtained from the GEP and RA models appear to well compatible with the experimental results and find to be very credible according to the results of other formulas. [ABSTRACT FROM AUTHOR]

Increasing draught seasons and lack of access to potable water reserves have been the major risks threatening water authorities and governments over the recent years. Therefore, long term water forecasts are receiving much more attention nowadays. Unlike the conventional projection of historical water demand, researchers have tried to implement sophisticated mathematical models to predict demand of water. Gene expression programming (GEP), as a relatively new forecasting technique, remains to be explored in this endeavor. The main purpose of this research was to assess the performance of GEP models using wavelet decomposition with 2 transfer functions ( db2 and haar ) and 3 levels. Results of this study showed GEP models can be highly sensitive to wavelet decomposition if all combinations of proper lag times are used as inputs feeding these models. [ABSTRACT FROM AUTHOR]

Copyright of Informatica (03505596) is the property of Slovene Society Informatika and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The intelligent system applications require automated data-driven modeling tools. The performance consistency of modeling tools is very essential to reduce the need for human intervention. Classical Gene Expression Programmings (GEPs) employ predefined genetic rules for the node-based evolution of expression trees in the absence of optimal numerical values of constant terminals, and these shortcomings can limit the search efficiency of expression trees. To alleviate negative impacts of these limitations on the data-driven GEP modeling performance, a Differential Evolutionary Chromosomal GEP (DEC-GEP) algorithm is suggested. The DEC-GEP utilizes the Differential Evolution (DE) algorithm for the optimization of a complete genotype of expression trees. For this purpose, a modifier gene container, which stores numerical values of constant terminals, is appended to the frame of GEP chromosome, and this modified chromosome structure enables simultaneous optimization of expression tree genotypes together with numerical values of constant terminals. Besides, the DEC-GEP algorithm can benefit from exploration and exploitation capabilities of the DE algorithm for more efficient evolution of GEP expression trees. To investigate consistency of the DEC-GEP algorithm in a data-driven modeling application, an experimental study was conducted for soft calibration of the low-cost, solid-state sensor array measurements, and results indicated that the DEC-GEP could yield dependable CO concentration estimation models for electronic nose applications. • This study introduces a complete chromosomal Gene Expression Programming (GEP) scheme. • A modifier gene container is added to chromosome to store values of constants. • Differential evolution optimizes complete chromosome to yield better expression trees. • Differential evolution optimizes complete chromosome to yield better expression trees. • This GEP is used for generation of CO concentration estimation models. [ABSTRACT FROM AUTHOR]

The use of lightweight structural concrete to reduce the weight of earthquake resistant buildings is very useful, and utilizing non-destructive tests for determining the concrete strength of these structures is essential. Ultrasonic pulse velocity test is one of the main methods of non-destructive testing in the assessment of compressive strength of concrete in the service area. The aim of this study is estimation of the compressive strength of lightweight aggregate concrete by offering suitable mathematical formulations. For this purpose, many samples of three different types of lightweight aggregate concrete are made and tested by concrete breaking machine and ultrasonic pulse velocity instrument. Then, some relationships are presented to predict the compressive strength of the concrete by using ultrasonic test results. Two different methods of fitting exponential function and Gene Expression Programming (GEP) are utilized to find the proper relationships. The results show that the exponential function has high accuracy in estimating the compressive strength of lightweight aggregate concrete made with natural pumices. But, it does not have good results for concrete made of expanded clay. However, the GEP has high accuracy for all the materials, and the experimental results are perfectly compatible with the test results. [ABSTRACT FROM AUTHOR]

A set of multivariate equations have been developed using gene expression programming (GEP) based symbolic regression technique to generate the flow quantiles of flow duration curve (FDC) in the ungauged catchments in the East Coast of Peninsular Malaysia. The equations were derived from four to seven candidate explanatory variables prepared from climatic, geomorphologic, geographic characteristics, soil properties, and land use and land cover information. Support vector machine (SVM) was used to optimize the best combinations for calibration and validation of GEP models from the data available in thirteen gauged catchments in the study area. Seven flow percentiles namely 0.05, 0.10, 0.25, 0.50, 0.75, 0.90, and 0.95 as well as extreme, maximum, minimum and mean annual flows were identified to develop a framework for predicting various flow metrics. Obtained results revealed that nonlinear regression equations developed using GEP can generate FDCs in ungauged catchments of East Coast of Peninsular Malaysia with an efficiency of up to 0.92. [ABSTRACT FROM AUTHOR]

The open, interconnected, and shared operational characteristics of the energy Internet introduce more sophisticated cybersecurity attacks. How to accurately detect these cyber attacks is crucial for energy Internet security protection. Existing machine learning-based intrusion detection algorithms cannot cope with the continuous increase of network traffic and features in the energy Internet. And convolutional neural networks (CNN) can be a good solution for the descending and optimal selection of high-dimensional intrusion features. Unfortunately, traditional convolutional neural networks have complex structures with many parameters and are prone to fall into local optimality. To fill the gap of CNN, in this paper, we use a gene expression programming (GEP) to optimize the parameters of CNN and propose an intrusion detection algorithm based on GEP-CNN (GCNN-IDS). Our key idea is to avoid the convolutional neural network from falling into local optimum by designing a new code on GEP and fitness function to optimize the parameters of the CNN using the global search capability of GEP. The experimental results on two benchmark datasets and a real dataset substantiate that the detection accuracy of the optimized CNN-based intrusion detection algorithm (ICNN-IDS) reaches up to 0.9143 under different parameter combinations; meanwhile, compared with other algorithms, the detection accuracy, precision, recall, F 1 and false detection rate of the intrusion detection model proposed in this paper reach 0.9897, 0.99, 0.98, 0.97 and 0.0126, respectively. • An intrusion detection model on CNN is proposed. • The intrusion detection on GEP-CNN is proposed. • The performances of our model are superior to the other models. [ABSTRACT FROM AUTHOR]

Gene expression programming (GEP) is a commonly used approach for solving symbolic regression (SR) problems. However, GEP often falls into a local optimum. GEP also randomly searches for results and could revisit low value search spaces, which impacts its performance. To overcome the two problems, we propose a new algorithm using an adversarial bandit (AB) technique to enhance GEP and the new algorithm is named AB-GEP. AB-GEP segments the mathematical expression space into many subspaces. It then leverages a new search space selection method, AvgExp3, to enhance the population jump between subspaces. This prevents the algorithm from falling into a local optimum. AvgExp3 dynamically estimates a subspace by the rewards generated in the search space. The dynamical reward estimation makes AB-GEP more adaptive to reward changes and it also provides guidance for AB-GEP to choose a subspace that could contain the correct results. This study proves that AvgExp3 is an unbiased estimation of the average rewards of subspace, and its variance is lower than the standard Exp3 method. The evaluation on two benchmark datasets shows that AB-GEP can maintain better population diversity and obtain better results than three traditional GEPs, GEP, SL-GEP, and SPJ-GEP. AB-GEP ranks top in 50% of the 32 benchmarks, while one of the traditional GEPs only gets no more than 20% of the best results. We release our code at https://github.com/kgae-cup/ab-gep. [ABSTRACT FROM AUTHOR]

Precise evaluation of pure hydrocarbon and natural gas viscosity is vital for reliable reservoir characterization, simulation, transportation and optimum consumption. The most trustable sources of pure hydrocarbon and natural gas viscosity values are laboratory experiments. The need of new methods becomes important when there is not enough experimental data for specific composition, pressure, and temperature conditions. In this study, a promising approach is utilized for the prediction of viscosities of pure hydrocarbons as well as gas mixtures containing heavy hydrocarbon components and impurities such as carbon dioxide, nitrogen, helium, and hydrocarbon sulfide using over 3800 data sets. Gene Expression Programming (GEP) is employed to develop a general model for pure and natural gas viscosity. The proposed model is a function of pseudo reduced pressure, pseudo reduced temperature, molecular weight and density. In addition, comparative studies are performed between the results obtained by the GEP model and previously published empirical correlations. To this end, statistical and graphical error analyses are used simultaneously. The results obtained show a value of 4.9% for average absolute percent relative error which is a measure of relative absolute deviation from the experimental data. The results also propose that standard deviation as a sign of data scattering is only 0.0870. These observations illustrate that the GEP model is more robust, reliable and consistent than the existing correlations for prediction of pure and natural gas viscosity. Finally, the relevancy factor shows that molecular weight has the greatest effect on gas viscosity. [ABSTRACT FROM AUTHOR]

Accurate determination of pressure-volume-temperature (PVT) properties of petroleum reservoirs is essential in material balance calculations, inflow performance, well-test analysis, reservoir simulation, etc. Ideally, those data should be obtained experimentally; however, experimental measurements require accurate and enough sampling, and are time consuming, expensive and tedious. Therefore, seeking for a simple, reliable and accurate model for prediction of PVT properties of petroleum systems is of a vital importance. In this communication, a large PVT data bank, covering a wide range of thermodynamic conditions was collected from variety of geographical locations around the world. Afterward, gene expression programming (GEP) was employed to develop a universal model for solution gas: oil ratio. The proposed model is a function of bubble point pressure, gas specific gravity, and oil API gravity, and has a very simple format with only one tuning parameter. The proposed model was compared to both explicit and implicit models available in literature for prediction of solution gas: oil ratio, using statistical and graphical error analyses. The results of this study indicate that the proposed model is more accurate, reliable and efficient compared to all other published correlations. [ABSTRACT FROM AUTHOR]

In the present work, the performance and emission parameters of a single cylinder four-stroke CRDI engine under CNG-diesel dual-fuel mode have been modelled by Gene Expression Programming. Based on the experimental data, GEP model was developed to predict BSFCeq, BTE, NO x , PM and HC. Load, fuel injection pressure and CNG energy share were chosen as input parameters for the model. The developed GEP model was capable of predicting the performance and emission parameters with commendable accuracy as observed from correlation coefficients within the range of 0.999368–0.999999. Mean absolute percentage error in the range of 0.036–1.09% along with noticeably low root mean square errors provided an acceptable index of the robustness of the predicted accuracy. In addition, the obtained results were also compared with an ANN model developed on the same parametric ranges wherein the GEP model was observed to be superior in predicting the desired response variables. [ABSTRACT FROM AUTHOR]

In this paper, we first find out some good trading strategies from the historical series and apply them in the future. The profitable strategies are trained out by the gene expression programming (GEP), which involves some well-known stock technical indicators as features. Our data set collects the 100 stocks with the top capital from the listed companies in the Taiwan stock market. Accordingly, we build a new series called portfolio index as the investment target. For each trading day, we search for some similar template intervals from the historical data and pick out the pertained trading strategies from the strategy pool. These strategies are validated by the return during a few days before the trading day to check whether each of them is suitable or not. Then these suitable strategies decide the buying or selling consensus signal with the majority vote on the trading day. The training period is from 1996/1/6 to 2012/12/28, and the testing period is from 2000/1/4 to 2012/12/28. Two simulation experiments are performed. In experiment 1, the best average accumulated return is 548.97% (average annualized return is 15.47%). In experiment 2, we increase the diversity of trading strategies with more training. The best average accumulated return is increased to 685.31% (average annualized return is 17.18%). These two results are much better than that of the buy-and-hold strategy, whose return is 287.00%. [ABSTRACT FROM AUTHOR]

Highlights: [•] One method for guiding the investment in mutual funds is proposed. [•] The buying timing and selling timing are decided by the trading functions generated by GEP. [•] The funds in the portfolio are selected by the ranking of Sortino ratios. [•] EQ and MV models are for capital allocation, and the annual profit higher than 12% in both models. [•] SPA test for ensuring our method can earn positive returns without data snooping. [ABSTRACT FROM AUTHOR]

This study compares capabilities of two different modelling methodologies for predicting breaking strength of rotor spun yarns. Forty eight yarn samples were produced considering variations in three drawing frame parameters namely break draft, delivery speed, and distance between back and middle rolls. Several topologies with different architectures were trained to get the best adaptive neuro-fuzzy inference system (ANFIS) and gene expression programming (GEP) models. Prediction performance of the GEP model was compared with that of ANFIS using root mean square error (RMSE) and correlation coefficient (R²-Value) parameters on the test data. Results show that, the GEP model has a significant priority over the ANFIS model in term of prediction accuracy. The correlation coefficient (R²-value) and root mean square error for the GEP model were 0.87 and 0.35 respectively, while these parameters were 0.48 and 0.53 for the ANFIS model. Also, a mathematical formula was developed with high degree of accuracy using GEP algorithm to predict the breaking strength of the yarns. This advantage is not accessible in the ANFIS model. [ABSTRACT FROM AUTHOR]

The accuracy of artificial neural networks (ANNs), adaptive neuro-fuzzy inference system (ANFIS) and gene expression programming (GEP) in modeling dissolved oxygen (DO) concentration was investigated in this study. Water temperature, specific conductance, pH, discharge and DO concentration data from South Platte River at Englewood, Colorado were used. Various input combinations of these data were tried as inputs to the ANN and ANFIS methods. The ANN and ANFIS models with the water temperature, specific conductance, pH and discharge input parameters performed the best. The optimal GEP model was obtained for the best input combination and compared with the ANN and ANFIS models with respect to correlation coefficient, root mean square error, mean absolute error and mean absolute relative error criteria. Results revealed that the GEP model performed better than the ANN and ANFIS models in modeling DO concentration. [ABSTRACT FROM AUTHOR]

Unsaturated soil shear strength can be determined using e ective stress relation that depends on the effective stress parameter. Several models have been developed in the past few years to estimate this parameter. In this research, the Gene Expression Programming (GEP) is used as an Artificial Intelligence (AI) method for developing a model to predict the e ective stress parameter, using effcient parameters. The principal advantage of the GEP approach is its ability to generate powerful prediction equations without any prior assumption on the possible form of the functional relationship. The input terminal set consists of net confining pressure, suction, Soil Water Characteristic Curve (SWCC) fitting parameter, bubbling pressure, residual and saturated volumetric water content. The output terminal set has one member, which is the effective stress parameter. An experimental database obtained from the literature is employed to develop the model. Comparison of the model prediction with the actual data, as well as other investigators, indicates a very good performance and ability of model. Sensitivity and parametric analyses are conducted to verify the results. It is also shown that soil suction is the most inuential parameter in the effective stress parameter of unsaturated soils. [ABSTRACT FROM AUTHOR]

Seven years of hourly temperature and electric load data for British Columbia in western Canada were used to compare two statistical methods, artificial neural networks (ANN) and gene expression programming (GEP), to produce hour-ahead load forecasts. Two linear control methods (persistence and multiple linear regression) were used for verification purposes. A two-stage (predictor-corrector) approach was used. The first stage used a single regression model that applied weather and calendar data for the previous hour to predict load for any hour of the day, and the second stage applied different corrections every hour, based on a high correlation of today's load with yesterday's load for any given hour. By excluding the day-before variables, the two-stage method reduced the total number of variables in the first-stage regression and gave better results. The first five years of data were used for training (regression finding) and validation (comparative testing of candidate functions to reduce overfitting) and the last two for verification (scoring with independent data). It was found that both non-linear methods worked better than the linear methods for the first stage. All methods worked well for the second stage; hence, persistence is recommended for the second stage because it is easiest to do. After both stages, the load error was less than 0.6% of the load magnitude for hour-ahead forecasts. When used iteratively to create forecasts up to 24 hours ahead, errors grew to about 3.5% of the load magnitude for both GEP and ANNs. We also experimented by training the statistical methods with a shorter period (one year) of past data to examine the over-fitting problem. Overall, ANNs were more useful in fitting the curves to robust data, and GEP was superior for short datasets and was less sensitive to the length of the dataset. We also found that a time-nested electric load forecasting model with different lead times kept maximum load errors to 2.1% of the load magnitude out to a horizon of 24 forecasts. RÉSUMÉ[Traduit par la rédaction] Nous avons utilisé sept années de données de température et de charge électrique pour la Colombie-Britannique dans l'ouest du Canada pour comparer deux méthodes statistiques, les réseaux neuronaux artificiels (ANN) et la programmation génétique (GEP), pour produire des prévisions de charge une heure à l'avance. Nous nous sommes servis de deux méthodes de vérification linéaires (persistance et régression linéaire multiple) à des fins de vérification. Nous avons utilisé une approche en deux étapes (prédicteur–correcteur). Dans la première étape, un modèle de régression simple appliquait des données météorologiques et de calendrier pour l'heure précédente pour prévoir la charge à n'importe quelle heure de la journée et la deuxième étape appliquait à chaque heure différentes corrections basées sur une forte corrélation de la charge d'aujourd'hui avec la charge d'hier pour une heure quelconque. En excluant les variables du jour précédent, la méthode à deux étapes a réduit le nombre total de variables dans la régression de la première étape et a donné de meilleurs résultats. Nous avons utilisé les cinq premières années de données pour l'apprentissage (détermination de la régression) et la validation (essais comparatifs de fonctions candidates pour réduire le surapprentissage) et les deux dernières pour la vérification (établissement d'un score à l'aide de données indépendantes). Il ressort que les deux méthodes non linéaires ont mieux fonctionné que les méthodes linéaires pour la première étape. Toutes les méthodes ont bien fonctionné pour la deuxième étape; la persistance est donc recommandée pour la deuxième étape parce qu'elle est plus facile à appliquer. Après les deux étapes, l'erreur sur la charge était inférieure à 0,6% de la grandeur de la charge pour les prévisions d'une heure à l'avance. Lorsque utilisées de façon itérative pour produire des prévisions de jusqu’à 24 heures, les erreurs ont augmenté jusqu’à environ 3,5% de la grandeur de la charge tant pour la GEP que les ANN. Nous avons aussi expérimenté en entraînant les méthodes statistiques avec une période plus courte (un an) de données passées pour examiner le problème du surapprentissage. Dans l'ensemble, les ANN étaient plus utiles pour ajuster les courbes aux données robustes alors que la GEP était meilleure pour les courts ensembles de données et était moins sensible à la longueur de l'ensemble de données. Nous avons aussi trouvé qu'un modèle de prévision de charge électrique emboîté dans le temps avec différent temps de prévision a limité les erreurs sur la charge à 2,1% de la grandeur de la charge jusqu’à un horizon de 24 prévisions. [ABSTRACT FROM AUTHOR]

Abstract: Water distribution system (WDS), as the important and expensive infrastructure, plays a crucial role in supplying water for the citizens living in the urban areas. The huge cost of the system drives researchers to seek the optimal cost design. During the last several decades, evolutionary methods such as genetic algorithms (GA), ant colony optimization algorithms (ACO) and simulated annealing (SA) have been found to explore the optimal combination of pipe diameters for design of WDS. However, in the case of complex distribution network, those methods suffer with low computational efficiency. In this paper, the gene expression programming (GEP) algorithms, a method based on the genetic algorithms, is adopted to solve the optimal design of WDS. This method is applied to the Hanoi benchmark network, the result of 6.081 million costs and 15,000 evaluations is combined with that obtained with GA and SA. From the results, it is observed that the proposed algorithm is an advanced alternative for the design of WDS, taking the computational efficiency and the ability of finding global or near-global optimal solutions into consideration. [Copyright &y& Elsevier]

Exploring relationships between characteristics of a yarn and influencing factors is momentous subject to optimize the selection of the variables. Different modelling methodologies have been used to predict spun yarn properties. Developing a prediction approach with higher degree of precision is a subject that has received attention by the researchers. In the last decade, Artificial Neural Network (ANN) has been developed successfully for textile nonlinear processes. In spite of the precision, ANN is a black box and does not indicate inter-relationship between input and output parameters. Hence, Gene Expression Programming (GEP) is presented here as an intelligent algorithm to predict breaking strength of rotor spun yarns based on draw frame parameters as one of the most important stages in spinning line. Forty eight samples were produced and different models were evaluated. Prediction performance of the GEP was compared with that of ANN using Mean Square Error (MSE) and correlation coefficient (R²-Value) parameters on test data. The results showed a better capability of the GEP model in comparison to the ANN model. The R²-value and MSE were 97% and 0.071 respectively which means desirable predictive power of GEP algorithm. Finally, an equation was extracted to predict breaking strength of the yarns with a high degree of accuracy using GEP algorithm. [ABSTRACT FROM AUTHOR]

Abstract: Fault detection and isolation in rotating machinery is very important from an industrial viewpoint as it can help in maintenance activities and significantly reduce the down-time of the machine, resulting in major cost savings. Traditional methods have been found to be not very accurate. Soft computing based methods are now being increasingly employed for the purpose. The proposed method is based on a genetic programming technique which is known as gene expression programming (GEP). GEP is somewhat a new member of the genetic programming family. The main objective of this paper is to compare the classification accuracy of the proposed evolutionary computing based method with other pattern classification approaches such as support vector machine (SVM), Wavelet-GEP, and proximal support vector machine (PSVM). For this purpose, six states viz., normal, bearing fault, impeller fault, seal fault, impeller and bearing fault together, cavitation are simulated on centrifugal pump. Decision tree algorithm is used to select the features. The results obtained using GEP is compared with the performance of Wavelet-GEP, support vector machine (SVM) and proximal support vector machine (PSVM) based classifiers. It is observed that both GEP and SVM equally outperform the other two classifiers (PSVM and Wavelet-GEP) considered in the present study. [Copyright &y& Elsevier]

Abstract: Mining the interesting function from the very complex and large data sets is an important research direction in DM and KDD. Traditional methods depend on experience and professional knowledge and involve subjectivity and blindness when determine function type and can not work well with complex function expression. Gene expression programming (GEP) is an adaptive genotype/phenotype evolutionary algorithm. Function mining method basted on GEP can overcome the shortcomings of the traditional methods. To overcome the deficiencies of GEP in maintaining population diversity and protection the optimal solution, we propose an improved gene expression programming algorithm, called GEP-FM. The new features implemented in GEP-FM are: a new gene structure adopted head, body, tail; self-adaptive mutation operator and a new fitness function. GEP-FM was applied to difficult identification function mining problems. The results show that GEP-FM has good performance and is superior to traditional algorithm and original GEP, the function mined by GEP-FM has a higher fitted value and prediction accuracy. [Copyright &y& Elsevier]

Abstract: Genetic/evolutionary methods are frequently used to deal with complex adaptive systems. The classic example is a Genetic Algorithm. A Genetic Algorithm uses a simple linear representation for possible solutions to a problem. This is usually a bit vector. Unfortunately, the natural representation for many problems is a tree structure. In order to deal with these types of problems many evolutionary methods make use of tree structures directly. Gene Expression Programming is a new, popular evolutionary technique that deals with these types of problems by using a linear representation for trees. In this paper we present and evaluate Robust Gene Expression Programming (RGEP). This technique is a simplification of Gene Expression Programming that is equally efficient and powerful. The underlying representation of a solution to a problem in RGEP is a bit vector as in Genetic Algorithms. It has fewer and simpler operators than those of Gene Expression Programming. We describe the basic technique, discuss its advantages over related methods, and evaluate its effectiveness on example problems. [Copyright &y& Elsevier]

Immune-inspired intrusion detection is a promising technology for network security, and well known for its diversity, adaptation, self-tolerance, etc. However, scalability and coverage are two major drawbacks of the immune-inspired intrusion detection systems (IIDSes). In this paper, we propose an IIDS framework, named GEP-IIDS, with improved basic system elements to address these two problems. First, an additional bio-inspired technique, gene expression programming (GEP), is introduced in detector (corresponding to detection rules) representation. In addition, inspired by the avidity model of immunology, new avidity/affinity functions taking the priority of attributes into account are given. Based on the above two improved elements, we also propose a novel immune algorithm that is capable of integrating two bio-inspired mechanisms (i.e., negative selection and positive selection) by using a balance factor. Finally, a pruning algorithm is given to reduce redundant detectors that consume footprint and detection time but do not contribute to improving performance. Our experimental results show the feasibility and effectiveness of our solution to handle the scalability and coverage problems of IIDS. [ABSTRACT FROM AUTHOR]

Abstract: Multisite Land Use Allocation Problem (MLUA) refers to the problem of allocating more than one land use type in an area. MLUA problem is one of the truly NP Complete (combinatorial optimization) problems. To cope with this type of problems, intelligent techniques such as genetic algorithms, and simulated annealing, have been used. Research in the area of Spatial Decision Support Systems (SDSS) for resource allocation issues, a new scientific area of information system applications developed to support semi-structured or unstructured spatial decisions, has recently generated attention for integrating Artificial Intelligence (AI) techniques with Geographic Information Systems (GIS). In this paper we demonstrate how GIS can be integrated with Gene Expression Programming (GEP), a recently developed AI approach, for solving MLUA problems. The feasibility of the proposed approach in solving MLUA problems was checked using a fictive case study. The results indicated that the proposed approach gives good and satisfactory results. [Copyright &y& Elsevier]

Abstract: Gene Expression Programming (GEP) is a new technique of evolutionary algorithm that implements genome/phoneme representation in computing programs. Due to its power in global search, it is widely applied in symbolic regression. However, little work has been done to apply it to real parameter optimization yet. This paper proposes a real parameter optimization method named Uniform-Constants based GEP (UC-GEP). In UC-GEP, the constant domain directly participates in the evolution. Our research conducted extensive experiments over nine benchmark functions from the IEEE Congress on Evolutionary Computation 2005 and compared the results to three other algorithms namely Meta-Constants based GEP (MC-GEP), Meta-Uniform-Constants based GEP (MUC-GEP), and the Floating Point Genetic Algorithm (FP-GA). For simplicity, all GEP methods adopt a one-tier index gene structure. The results demonstrate the optimal performance of our UC-GEP in solving multimodal problems and show that at least one GEP method outperforms FP-GA on all test functions with higher computational complexity. [Copyright &y& Elsevier]

Predication is one of the fundamental tasks of data mining. In recent years, Artificial Intelligence techniques are widely being used in data mining applications where conventional statistical methods were used such as Regression and classification. The aim of this work is to show the applicability of Gene Expression Programming (GEP), a recently developed Al technique, for hydraulic data prediction and to evaluate its performance by comparing it with Multiple Linear Regression (MLR). Both GEP and MLR were used to model the hydraulic jump over a roughened bed using very large series of experimental data that contain all the important flow and roughness parameters such as the initial Froude number, the height of roughness ratio, the length of roughness ratio, the initial length ratio (from the gate) and the roughness density. The results show that GEP is a promising Al approach for hydraulic data prediction. [ABSTRACT FROM AUTHOR]

Currently, the only widely available tool for controlling the SARS-CoV-2 pandemic is nonpharmacological interventions (NPIs). Coronavirus aerosols are around 0.3–2 µm in diameter (0.9 m in mass). The present study used artificial intelligence such as gene expression programming (GEP) and genetic algorithms (GA) were used to predict and optimize the diameter of Nylon-6,6 nanofibers via electrospinning for protection against coronavirus. It is suggested that using the controlled experimental conditions such as concentration of nylon-6,6 (16% wt/v), applied voltage (26 kV), working distance (18 cm) and injection rate (0.2 mL/h) have resulted the diameter of nylon-6,6 nanofibers about 55.8 nm. Coronavirus face masks could use the obtained diameter and electrostatic interaction between viral particles and naofibers as active layers. [ABSTRACT FROM AUTHOR]

Current dynamic markets require manufacturing industries to organize a robust plan to cope with uncertain demand planning. This work addresses the mixed-model multi-manned assembly line balancing under uncertain demand and aims to optimize the assembly line configuration by a robust mixed-integer linear programming (MILP) model and a robust solution generation mechanism embedded with dispatching rules. The proposed model relaxes the cycle time constraint and designs robust sequencing constraints and objective functions to ensure the line configuration can meet all the demand plans. Furthermore, two solution generation mechanisms, including a task-operator-sequence and an operator-task-sequence, are designed. To quickly find a suitable line configuration, a gene expression programming (GEP) approach with multi-attribute representation is proposed to obtain efficient dispatching rules which are ultimately embedded into the solution generation mechanisms. Experimental results show that solving the proposed MILP model mathematically is effective when tackling small and medium-scale instances. However, for large instances, the dispatching rules generated by the GEP have significant superiority over traditional heuristic rules and those rules mined by a genetic programming algorithm. • Uncertain demand are first considered into mixed-model multi-manned assembly line balancing problems. • A novel robust MILP model with cycle time relaxation and robust objective function is developed. • Two robust heuristic solution generation embedded with evolutionary-generated rules are proposed to obtain robust solutions. • A gene expression programming with multi-attribute representation to mine efficient dispatching rules. [ABSTRACT FROM AUTHOR]

Evolutionary computation (EC) is a widely used computational intelligence that facilitates the formulation of a range of complex engineering problems. This study tackled two hybrid EC techniques based on genetic programming (GP) for ground motion prediction equations (GMPEs). The first method coupled regression analysis with multi-objective genetic programming. In this way, the strategy was maximizing the accuracy and minimizing the models' complexity simultaneously. The second approach incorporated mesh adaptive direct search (MADS) into gene expression programming to optimize the obtained coefficients. A big data set provided by the Pacific Earthquake Engineering Research Centre (PEER) was used for the model development. Two explicit formulations were developed during this effort. In those formulae, we correlated spectral acceleration to a set of seismological parameters, including the period of vibration, magnitude, the closest distance to the fault ruptured area, shear wave velocity averaged over the top 30 meters, and style of faulting. The GP-based models are verified by a comprehensive comparison with the most well-known methods for GMPEs. The results show that the proposed models are quite simple and straightforward. The high degrees of accuracy of the predictions are competitive with the NGA complex models. Correlations of the predicted data using GEP-MADs and MOGP-R models with the real observations seem to be better than those available in the literature. Three statistical measures for GMPEs, such as E (%), LLH, and EDR index, confirmed those observations. • Evolutionary Computation was applied to ground-motion prediction equations. • Two hybrid approaches were utilized to develop explicit formula. • First, a regression analysis combined with multi-objective genetic programming. • Second, an adaptive direct search hybridized with gene expression programming. • The data provided by Pacific Earthquake Engineering Research Centre was utilized. [ABSTRACT FROM AUTHOR]

Aquifer encroachments toward the producing gas zones can substantially decrease the volumetric gas flowrate due to the bypassing of large gas volumes in the porous media. It is well-known that the main parameter monitoring the cumulative gas production is the residual gas saturation; therefore, a complete information about strong aquifer drive gas reservoirs is required relying on both operational and petrophysical parameters. Despite the abundance of studies, attempting to establish comprehensive and adequately precise models for estimating the residual gas saturations is still challenging. In present work, the new method namely, Gene Expression Programming (GEP), was used to propose 9 correlations for force, spontaneous imbibition and the overall waterflood conditions. For this reason, a comprehensive database including 897 residual gas saturation data were adopted from the open source literatures. The variation ranges of the employed dataset include 0.013 to 1 for initial gas saturation, 0.001 to 12950 mD for absolute permeability of rock, 0.008 to 0.494 for porosity (as the input parameters) and 0 to 0.79 for residual gas saturation (as the model output). Numerous types of error analyses such as statistical and visual tools are employed to assess the prediction capability of the new methods developed here as compared with a commonly applied literature model. As a result, it is perceived that the General I, Imbibition I and Force I are the most effective correlations with the root mean square error (RMSE) of 0.09, 0.06 and 0.14, consecutively. The results of the literature model are less efficient than our newly suggested methods. Moreover, the so-called technique of sensitivity analysis revealed that the porosity and permeability have insignificant impacts on predicting the residual gas saturation in the case of spontaneous imbibition. To end with, the results of this study can be the good nominees for fast, accurate and inexpensive estimation of the residual gas saturation especially during reservoir engineering simulation and calculation at the time in which the whole information may not be available to the researches. Relative impact value of each input variable on the estimation of residual gas saturation during different waterflood conditions for the best models proposed here. Image 1 • Using GEP strategy, several empirical models for predicting residual gas saturation were developed for the first time. • Several statistical quality measures and graphical illustrations were used to assess the model performance. • The new methods were compared with the commonly applied literature correlations. • Consequently, the new methods in this study show the best agreement with the measured data. • The developed new methods can be a powerful tool for applying in gas reservoir simulators. [ABSTRACT FROM AUTHOR]

Complex and flexible access mode, and frequent data interaction bring about large security risks to data transmission for active distribution networks. How to ensure data security is critical to the safe and stable operation of active distribution networks. Traditional methods, like access control, data encryption, and text filtering based on intelligent algorithms, are difficult to ensure the security of dynamically increased and high-dimensional numerical data transmission in active distribution networks. In this paper, we first propose a rough feature selection algorithm based on the average importance measurement (RFS-AIM) to simplify the complexity of data recognition. Then, we propose a sensitive data recognition function mining algorithm based on RFS-AIM and improved gene expression programming (SDR-IGEP) where population update operation is constructed by chromosome similarity based on the Jaccard coefficient. The operation avoids local convergence of the gene express programming by increasing individual diversity in the new population. Finally, we present a new incremental mining algorithm for a sensitive data recognition function based on global function fitting (ISDR-GFF) by using a grain granulation model for incremental datasets. The experimental results on IEEE benchmark datasets and real datasets show that the algorithms proposed in this paper outperform the state-of-the-art algorithms in terms of the average running time, precision, recall, F 1 index, accuracy, specificity and speedup on all experimental datasets. [ABSTRACT FROM AUTHOR]