Hybrid Intelligent System Research Articles

Mathematical modeling of hybrid intelligent system to longitudinal landing control design

Mathematical modeling of hybrid intelligent system to longitudinal landing control design

Adaptive surrogate modeling with evolutionary algorithm for well placement optimization in fractured reservoirs

Well placement optimization is a decisive task for the reliable design of field development plans. The use of optimization routines coupled to reservoir simulation models as an automatic tool is a popular practice, which could improve the decision-making process on well placement problems. However, despite the various automatic techniques developed, there is still a lack of robust computer-added optimization tool, which can solve the well placement problem with high accuracy in reasonable time while handling the technical constraints properly. In this paper, a hybrid intelligent system is proposed to deal with a real well placement problem with arbitrary well trajectories, complex model grids, and linear and nonlinear constraints. In this intelligent approach, a Genetic Algorithm (GA) combined with a hybrid constraint-handling strategy is applied in conjunction with a constrained space-filling sampling design, Gaussian Process (GP) surrogate model, and one proposed adaptive sampling routine. This self-adaptive framework allows to consecutively augment the quality of surrogate, enhance the accuracy of the process, and thus guide the optimization rapidly into the optimal solution. To demonstrate the efficiency of the developed method, a full-field reservoir case is considered. This case covers a real well placement project in a fractured unconventional reservoir of El Gassi, which is a mature field located in Hassi-Massoud, Algeria. The obtained results highlighted the effectiveness of the proposed approach for solving the real well placement problem with high accuracy in reasonable CPU-time. These auspicious features make it a reliable tool to be used on other real optimization projects.

Design of a hybrid intelligent system for the management of flood disaster risks

The frequency of occurrence and intensity of floods is a huge threat to environment, human existence, critical infrastructure and economy. Flood risk assessments depend on probabilistic approaches and suffer from non-existence of appropriate indices of acceptable risk, dearth of information and pieces of knowledge for explicit view and understanding of the characteristics and severity level of flood hazard. This paper proposes a hybridized intelligent framework comprising fuzzy logic (FL), neural network and genetic algorithm for clustering and visualization of flood data, prediction and classification of flood risks severity level. A multidimensional knowledge model of flood incidence using star, snowflake and facts constellation schemas was proposed for the knowledge warehouse. A six-layered adaptive neuro-fuzzy inference system implementing mamdani’s inference mechanism was design to evaluate input features based on fuzzy rules held in the multidimensional data model. The system is aimed at predicticting and classifying flood risk severity levels. The perception of emergency risk management is very important in modern society. Therefore, this work provides a framework for the practical applications of data mining techniques and tools to emergency risk management. The work would assist to identify locations with significant flood risk.

A Special Section on Hybrid Intelligent Systems and Applications in Medical Imaging and Health Informatics

<i>A Special Section on</i> Hybrid Intelligent Systems and Applications in Medical Imaging and Health Informatics

Optimized Artificial Neural Network for Biosignals Classification Using Genetic Algorithm

The artificial neural networks (ANNs) are increasingly being used to solve the problem of pattern recognition, but it is an arduous task for their designer to obtain the optimal topology to be used for ANN training since this is considered a very difficult problem. Even after several tests, the optimized topology may not be reached. A possible solution for this problem is the use of a hybrid intelligent system; an optimization technique is used together with the ANN in order to search for an optimized topology. This paper applies this concept, using the genetic algorithms for the optimization of the topology of a multilayer perceptron, used for the classification of wrist orientation, muscle contraction levels and subjective parameters of the voice. The data were preprocessed with wavelet packet transform. The tool presents promising results above 96% all the way up to 99% of total hits, with 98% and 90% reliability.

Hybrid intelligence system using fuzzy inference in cluster architecture for secured group communication

The complication of efficient group communication design in real-world application has been continuously increasing. More factors have to be taken into account continuously before any decision about the rekeying process in cluster architecture could be derived. In this scenario the decision-making process should be efficient and optimal; therefore, fuzzy inference system can be implemented for fast decision making in rekeying process. Secured and efficient group communication can be attained by deploying fuzzy-based inference system in hybrid clustered architecture of group communication. The aim of this proposed model is to reduce the communication and computation complexity of the system using fuzzy-based inference system. In this proposed model, a fuzzy inference system is designed with the function for membership in the group, and the manipulation on input–output relations is concluded. Moreover, by implementing the proposed system, complexity overhead is reduced due to efficient framework and minimized key management. Simulation is done to verify the efficiency of proposed model for the members, to subscribe the service pack with a requested time slot.

An intelligent hybrid system for predicting the tortuosity of the pore system of fly ash concrete

In this paper, the soft computing approach is proposed to predict the transport tortuosity of the pore system of fly ash concrete as an important macroscopic parameter by building an intelligent hybrid system. This system is designed with a modular architecture composed of three neural network models optimized by genetic algorithms. The system is primarily based on the two main properties, compressive strength and porosity, which depend mainly on the concrete mix design, fly ash content and age. Performance of the proposed hybrid system is validated experimentally. Results indicate that concrete mixtures with 30% fly ash show the optimum transport tortuosity of the pore system. Moreover, results indicate also a threshold value of 33 for the transport tortuosity, above which both compressive strength and porosity are unimproved. This study concludes that the proposed system provides basis related to concrete durability and performs very well using a user-friendly interface to promote its effective utilisation.

A prediction model for overlying rock thickness of subsea tunnel: A hybrid intelligent system

The overlying rock thickness of a subsea tunnel controls its vertical line. It not only ensures the safety and stability of tunnel construction period and operation period, but also ensures the economy of subsea tunnel. In the current research, few papers give full consideration to the complex indicators of prediction the overlying rock thickness. However, in this study, a hybrid intelligent system was established to predict the overlying rock thickness of a subsea tunnel based on Qingdao Kiaochow Bay Subsea Tunnel, China. The sea depth, basic quality index of rock mass, soft soil layer thickness, permeability coefficient, and construction management level were selected as the main factors influencing the overlying rock thickness. Using the data obtained from project site exploration, objective weight factors were calculated using rough set theory, and subjective weight factors were calculated using the analytic hierarchy process. Furthermore, the combination of weights was obtained for each factor. Finally, the weight of influencing factors was incorporated into the extension model, and the overlying rock thickness of pending section was calculated. The results of overlying rock thickness prediction model are consistent with the actual value, indicating that the model has good engineering applicability and application value.

Computerized hybrid decision-making system for hormone replacement therapy in menopausal women.

The diversity of the results of different hormone replacement therapy (HRT) protocols and the fuzziness of the conclusions have caused problems in routine clinical practice. To develop an intelligent decision-making system for HRT specifically is appropriate as we use the abbrevation HRT in the background section in menopausal women in order to assist physicians. This study consisted of 179 peri- and post-menopausal patients who were admitted to Hacettepe University Hospital (between 1996 and 2001) with various menopausal complaints. Database variables used in this study were age, height, weight, menopause duration, clinical condition, HRT duration, and the laboratory test results. Our newly developed multiple-centered fuzzy clustering (MCFC) algorithm was applied to the medical data set to differentiate patient groups. Finally, a hybrid intelligent decision-making system was developed by combining knowledge-based algorithms and the MCFC algorithm results. We have used Fuzzy C-means, K-means, Hard C-means, similarity based clustering, and MCFC algorithms on the medical data set and have determined that the MCFC algorithm is the most appropriate algorithm for our medical model. We have defined 5 clusters and 16 cluster centers. A diagnostic phrase was assigned to each cluster center by the physician and these clusters together with knowledge-based algorithms were used for the decision-making system. We have developed a computerized hybrid decision-making system recommending HRT to peri- and post-menopausal women in order to assist and protect physicians.

Fuel Cell Output Current Prediction with a Hybrid Intelligent System

A fuel cell is a complex system, which produces electricity through an electrochemical reaction. For the formal application of control strategies on a fuel cell, it is very important to have a precise dynamic model of it. In this paper, a dynamic model of a real hydrogen fuel cell is obtained to predict its response. The data used in this paper to obtain the model have been acquired from a real fuel cell subjected to different load patterns by means of a programmable electronic load. Using this data, a nonlinear model based on a hybrid intelligent system is obtained. This hybrid model uses artificial neural networks to predict the output current of the fuel cell in a very precise way. The use of a hybrid scheme improves the performance of neural networks reducing to half the mean squared error obtained for a global model of the fuel cell.

Architecture of the Hybrid Intelligent Multi-agent System of Heterogeneous Thinking for Planning of Distribution Grid Restoration

Architecture of the Hybrid Intelligent Multi-agent System of Heterogeneous Thinking for Planning of Distribution Grid Restoration

Automated Detection of Parkinson's Disease Based on Multiple Types of Sustained Phonations Using Linear Discriminant Analysis and Genetically Optimized Neural Network.

Objective: Parkinson’s disease (PD) is a serious neurodegenerative disorder. It is reported that most of PD patients have voice impairments. But these voice impairments are not perceptible to common listeners. Therefore, different machine learning methods have been developed for automated PD detection. However, these methods either lack generalization and clinically significant classification performance or face the problem of subject overlap. Methods: To overcome the problems discussed above, we attempt to develop a hybrid intelligent system that can automatically perform acoustic analysis of voice signals in order to detect PD. The proposed intelligent system uses linear discriminant analysis (LDA) for dimensionality reduction and genetic algorithm (GA) for hyperparameters optimization of neural network (NN) which is used as a predictive model. Moreover, to avoid subject overlap, we use leave one subject out (LOSO) validation. Results: The proposed method namely LDA-NN-GA is evaluated in numerical experiments on multiple types of sustained phonations data in terms of accuracy, sensitivity, specificity, and Matthew correlation coefficient. It achieves classification accuracy of 95% on training database and 100% on testing database using all the extracted features. However, as the dataset is imbalanced in terms of gender, thus, to obtain unbiased results, we eliminated the gender dependent features and obtained accuracy of 80% for training database and 82.14% for testing database, which seems to be more unbiased results. Conclusion: Compared with the previous machine learning methods, the proposed LDA-NN-GA method shows better performance and lower complexity. Clinical Impact: The experimental results suggest that the proposed automated diagnostic system has the potential to classify PD patients from healthy subjects. Additionally, in future the proposed method can also be exploited for prodromal and differential diagnosis, which are considered challenging tasks.

Rock tensile strength prediction using empirical and soft computing approaches

The tensile strength of rocks plays an important role in site investigation during the initial phase of many construction projects. However, laboratory tests as direct methods for tensile strength prediction are not considered efficient in terms of time and cost. The present study is focused on developing new hybrid intelligent models for rock tensile strength prediction. Rock samples which were gathered from a tunnel in Malaysia were subjected to testing in the laboratory to prepare a database comprising 80 datasets for analysis. These rock index tests include Schmidt hammer test, dry density test, and point load test, together with the Brazilian tensile strength (BTS). Moreover, simple regression analysis was conducted and the results showed that a model with multiple inputs needs to be developed for tensile strength prediction. Subsequently, three hybrid intelligent systems, namely, an imperialist competitive algorithm (ICA)-artificial neural network (ANN), a genetic algorithm (GA)-ANN, and a particle swarm optimization (PSO)-ANN, were developed for estimating the tensile strength of rocks. In fact, the ICA, PSO, and GA were employed for weight and bias adjustments in ANN models. The values of the variance accounted for (VAF), coefficient of determination (R2), and root mean square error (RMSE) were obtained for the evaluation of the developed hybrid models. For the training datasets of the GA-ANN, PSO-ANN, and ICA-ANN models, R2 values of 0.911, 0.933, and 0.923, respectively, were obtained. On the other hand, R2 values of the testing datasets of the GA-ANN, PSO-ANN, and ICA-ANN models were 0.912, 0.929, and 0.920, respectively. The results showed that, although all the developed predictive models can predict the BTS within acceptable levels of accuracy, the PSO-ANN predictive model provided the best performance. Consequently, it was concluded that, due to the higher capability of the PSO-ANN model, it could be introduced as a new model for the prediction of the tensile strength of rocks.

A Hybrid Intelligent Multiagent System for the Remote Control of Solar Farms

ABSTRACTThis paper describes a multiagent architecture integrated system designed to supervise infrastructures in solar farms. The system enables monitoring the environment by means of sensor networks that are in charge of collecting data. It is designed using a hybrid model composed of an inference engine and an ontology. The former makes the system intelligent, while the latter structures knowledge. We have also developed a tool to configure and use the multiagent system in a simple and intuitive way.

Making AI Ready for the Wild: The Hybrid Intelligence Unicorn Hunter

This demonstration paper proposes a hybrid intelligence system that combines the complementary strengths of human and machines for complex decision-making problems that require human “gut feeling” (i.e. success prediction for startups). The architecture extends principles of previous interactive machine learning systems by using continuous input from an expert crowd and explicitly leveraging the advantages of collective intelligence. This approach allows to augment machine learning techniques for generating features, intuitive and analytic labeling as well as troubleshooting.

OA014] 4D lung tumor motion modelling using dynamic MLC tracking and EPID feedback

Purpose Accurate delivery of the prescribed dose to moving lung tumors is a key challenge in radiotherapy. Tumor tracking involves real-time specifying the target and correcting the geometry in order to compensate for the respiratory motion, that’s why tracking the tumor requires caution. This study proposes an approach to generate a model to track lung tumor motion by controlling dynamic multileaf collimators without any marker. Methods For this purpose, 4D-CT images of 10 patients were used and all the slices which contained the tumor were contoured for all patients. The first four phases of 4D-CT images which contained tumors were selected as input of the software, and the next six phases were considered as the output. A hybrid intelligent system, ANFIS (Adaptive neuro fuzzy inference system), is used to predict motion of lung tumor. The root mean square error (RMSE) was used to investigate the accuracy of ANFIS performance in tumor motion prediction. For modeling of respiratory motion the end-exhale phases of these images have been considered as the reference and were analyzed using the neuro-fuzzy method to predict the magnitude of displacement of the lung tumor. Then, the predicted data were used to determine the leaf motion in MLC. Finally, the trained algorithm was figured out using Shaper software to show how the MLCs can track the moving tumor and then imported on the Varian linac equipped with EPID. Results The results showed that the RMSE did not have a major variation. Also, there was a good agreement between the images obtained by EPID and Shaper for a respiratory cycle. Conclusions The data in the 4DCT images were used for motion tracking instead of using markers that leads to more information of tumor motion with respect to methods based on marker location. This developed approach can track the moving tumor with MLC based on the 4D modelling and so it can improve treatment accuracy, dose conformity and sparing of healthy tissues because of low error in margins that can be ignored. Therefore, this method can work more accurate compared with the gating and invasive approaches using markers.

GNRCS: Hybrid Classification System based on Neutrosophic Logic and Genetic Algorithm

In this paper, we present a hybrid intelligent system based on Neutrosophic Logic (NL). In conjunction with Genetic Algorithm(GA) for classification. The neutrosophic logic is adapted for representing different forms of knowledge. GA is used to refine the generated neutrosophic rules.

Combining predictors for multi-layer architecture of adaptive fuzzy inference system

To solve problems with a Sugeno adaptive fuzzy neural network using training data, it is necessary to select the appropriate combination of input characteristics of the sub-adaptive neuro-fuzzy inference system (ANFIS) and to determine the appropriate topology. The multi-layer architecture of a sub-ANFIS (MLA-ANFIS) is a good model for prediction problems and solves them modularity. Since, the combination of several predictors is the current focus in the construction of hybrid intelligent systems; we created many solutions to combine machine learning methods, namely ANFIS, support vector machine (SVM), deep neural network (DNN), naive Bayes (NB), linear regression (LR), extreme learning machine (ELM), and decision tree (DT) mixed predictors, and ensemble bootstrap aggregation based on MLA-ANFIS in order to discover the optimal model of combined predictors based on the MLA-ANFIS with a combination of input features entered in the MLA-ANFIS. We implemented our approaches on 365-day concrete compressive strength, thoracic surgery, fertility diagnosis, breast, energy, and glass identification datasets from UCI. The experimental results prove that the combining predictors for the MLA-ANFIS show performance improvements compared to the pure MLA-ANFIS method.

Applying various hybrid intelligent systems to evaluate and predict slope stability under static and dynamic conditions

The evaluation and precise prediction of safety factor (SF) of slopes can be useful in designing/analyzing these important structures. In this study, an attempt has been made to evaluate/predict SF of many homogenous slopes in static and dynamic conditions through applying various hybrid intelligent systems namely imperialist competitive algorithm (ICA)-artificial neural network (ANN), genetic algorithm (GA)-ANN, particle swarm optimization (PSO)-ANN and artificial bee colony (ABC)-ANN. In fact, ICA, PSO, GA and ABC were used to adjust weights and biases of ANN model. In order to achieve the aim of this study, a database composed of 699 datasets with 5 model inputs including slope gradient, slope height, friction angle of soil, soil cohesion and peak ground acceleration and one output (SF) was established. Several parametric investigations were conducted in order to determine the most effective factors of GA, ICA, ABC and PSO algorithms. The obtained results of hybrid models were check considering two performance indices, i.e., root-mean-square error and coefficient of determination $$(R^{2})$$ . To evaluate capability of all hybrid models, a new system of ranking, i.e., the color intensity rating, was developed. As a result, although all predictive models are able to approximate slope SF values, PSO-ANN predictive model can perform better compared to others. Based on $$R^{2}$$ , values of (0.969, 0.957, 0.980 and 0.920) were found for testing of ICA-ANN, ABC-ANN, PSO-ANN and GA-ANN predictive models, respectively, which show higher efficiency of the PSO-ANN model in predicting slope SF values.

Short-term prediction of market-clearing price of electricity in the presence of wind power plants by a hybrid intelligent system

This paper provides a new hybrid intelligent method for short-term prediction of the market-clearing price of electricity in the presence of wind power plants. The proposed method uses a data filtering technique based on wavelet transform and a radial basis function neural network, which is utilized for primary prediction. The main prediction engine comprises three MLP neural networks with different learning algorithms. To get rid of local minimums and to optimize the all neural networks, the meta-heuristic Imperialist Competitive Algorithm method is used. The input data for network training belong to the Nord Pool power market. The information includes a complete set of the historical record on electricity price and wind power generation. Moreover, the simultaneous impact of wind power generation is analyzed to predict the market-clearing price. Besides, the correlation coefficient factor is provided to consider the impact of wind power in forecasting the electricity price. Simulation results show the supremacy of the proposed method over other methods, to which it has been compared in this study. Also, the prediction error decreases significantly.