Neural Genetic Algorithm Research Articles

Hybrid identification method of coupled viscoplastic-damage constitutive parameters based on BP neural network and genetic algorithm

The constitutive model based on the theoretical framework of coupled viscoplastic-damage involves calibration of multiple and high coupling parameters. The inverse calibration by genetic algorithm (GA) with global search ability has some challenges as the dependence on the selection of the initial population, massive computation, and convergence inconsistency. To obtain statistical knowledge from state data to avoid subjective experience, a hybrid identification method based on the BP neural network and GA is proposed. A coupled viscoplastic-damage constitutive model based on the thermal deformation and microstructure evolution is established. The parameters in the model are determined based on the hybrid identification method. Two types of aluminum alloy sheets are selected to test the generalization, and mean square errors (RMSE) are 2.46 and 4.89, respectively. The results indicate that this method has higher accuracy than the inverse calibration method based on single optimization algorithm.

Artificial Neural Network Based Approach for Voltage Stability Analysis of for Sustained Operation of Power System

For reliable and secure power system, the stability analysis is recognized as an important problem. Voltage stability is the capacity of a power system to maintain steady acceptable voltages at all buses in the system. Voltage stability index (VSI) evaluation for a situation of power system can act as an accurate and fast indicator of the proximity of the system to voltage instability. Recently there has been considerable interest in intelligent methods based on artificial neural network (ANN), fuzzy logic and genetic algorithm to voltage stability assessment problem. ANN, with the ability to provide non-linear input/output mapping, parallel processing, learning and generalization have the potential to make them ideally suited for estimating VSI’s of a power system without solving the governing power system equations. This paper is to purpose an alternative method using ANN for finding the closeness of system operating point to voltage collapse that would be claimed to have better computational speed, accuracy, efficiency and reliability. Voltage Stability Analysis (VSA) using VSI is performed for different alternate loading strategies of power network building ANN models for every different scenario. The outcome found working satisfactorily in analyzing voltage stability problem, basically in ranking the network buses according vulnerability order.

A hybrid model for evaluating the sawability of stones through the performance of frame sawing machine

Modeling of stone processing, especially the sawability of stones (SS) has been studied extensively over the last decade. Artificial neural network (ANN) is an effective method to deal with complex engineering problems. However, ANN has some limitations such as trapped in local minima and cannot analyze the sensitivity of parameters. To fill the gaps, the primary objectives of this paper are to optimize ANN by genetic algorithm (GA) and introduce mean impact value (MIV) algorithm for parametric sensitivity analysis. Therefore, a hybrid model was proposed for predicting SS based on BP neural network (BPNN), MIV algorithm, and GA. Further, the main factors that affect SS were analyzed based on the grinding mechanism and fracture mechanics. The proposed model is not only meaningful to analyze SS, but provides a novel foundation for studying other stone procedures in future.

Estimating Physical Activity Energy Expenditure Using an Ensemble Model-Based Patch-Type Sensor Module

Chronic diseases, such as coronary artery disease and diabetes, are caused by inadequate physical activity and are the leading cause of increasing mortality and morbidity rates. Direct calorimetry by calorie production and indirect calorimetry by energy expenditure (EE) has been regarded as the best method for estimating the physical activity and EE. However, this method is inconvenient, owing to the use of an oxygen respiration measurement mask. In this study, we propose a model that estimates physical activity EE using an ensemble model that combines artificial neural networks and genetic algorithms using the data acquired from patch-type sensors. The proposed ensemble model achieved an accuracy of more than 92% (Root Mean Squared Error (RMSE) = 0.1893, R2 = 0.91, Mean Squared Error (MSE) = 0.014213, Mean Absolute Error (MAE) = 0.14020) by testing various structures through repeated experiments.

Selectivity index and separation efficiency prediction in industrial magnetic separation process using a hybrid neural genetic algorithm

It is essential to know the process efficiency in the industrial magnetic separation process under different operating conditions because it is required to control the process parameters to optimize the process efficiency. To our knowledge, there is no information about using artificial intelligence for modeling the magnetic separation process. Hence, finding a robust and more accurate estimation method for predicting the separation efficiency and selectivity index is still necessary. In this regard, a feed-forward neural network was developed to predict the separation efficiency and selectivity index. This model was trained to present a predictive model based on the percentage of iron, iron oxide and sulfur in mill feed and cobber feed, 80% passing size in mill feed and cobber feed and plant capacity. Therefore, this work aims to develop an intelligent technique based on an artificial neural network and a hybrid neural-genetic algorithm for modeling the concentration process. Results indicated that the values of mean square error and coefficient of determination for the testing phase were obtained 0.635 and 0.86 for selectivity index and of 4.646 and 0.84 for separation efficiency, respectively. In order to improve the performance of neural network, genetic algorithm was used to optimize the weights and biases of neural network. The results of modeling with GA-ANN technique indicated that the mean square error and coefficient of determination for the testing phase were achieved by 0.276 and 0.95 for selectivity index and of 1.782 and 0.92 for separation efficiency, respectively. The other statistical criteria for the GA-ANN model were better than those of the ANN model.

Optimization of diesel engine dual-variable geometry turbocharger regulated two-stage turbocharging system based on radial basis function neural network-quantum genetic algorithm

ABSTRACT Aiming at the problem of power and economy reduction in high-altitude operation of diesel engines, a radial basis function neural network-quantum genetic algorithm (RBFNN-QGA) optimization method is proposed to optimize diesel engine dual-variable geometry turbocharger (VGT) regulated two-stage turbocharging (RTST) system. Firstly, the GT-power simulation model of dual-VGT RTST diesel engine was established, and then calibrated with test data. The latin hypercube sampling (LHS) algorithm was selected for the experimental design, and the back propagation neural network (BPNN), RBFNN, and extreme learning machine (ELM) were used to establish three diesel engine high-altitude performance prediction models based on the sample data, to compare the ability of the three neural networks in predicting the performance of diesel engines. Finally, RBFNN is coupled with QGA to establish the optimization model of dual-VGT RTST system at different altitudes with torque, fuel consumption rate as the optimization target, and vortex front row temperature and in-cylinder combustion pressure as the constraints. The study results show that three kinds of neural networks have achieved satisfactory prediction results, BPNN and RBFNN are better than ELM in prediction accuracy, and ELM training is faster than BPNN and RBFNN; QGA converges faster and has higher optimization efficiency than GA, and the maximum torque of the optimized dual-VGT diesel engine is restored to more than 85% of the plain level, compared with the single-turbocharged diesel engine and the single-VGT two-stage turbocharged diesel engine, the low-speed torque of the dual-VGT RTST diesel engine is increased by 384% and 100%. This optimization model has better optimization effect and efficiency.

Optimum research on the temperature of the ship stern-shaft mechanical seal end faces based on finite element coupled analysis

Abstract Due to the complicated working conditions of the ship stern-shaft mechanical seals, it is very difficult to evaluate and optimize the temperature on the sealed end faces. In the paper, one FEA model of the small taper convergent stern-shaft mechanical seals was put forward, and the nonlinear end-face liquid film pressure distribution was derived from different end-face gaps based on the simplified Reynolds equation as the end-face input parameters in FEA model. To prove the correction of the FEA results, one experiment was designed and completed. At the same time, combined with a parametrically coupled FEA model and an orthogonal experimental design, one PSO-BP-GA method based on the combination of particle swarm optimization (PSO), BP neural network and genetic algorithm (GA) was proposed to optimize the temperature of the sealed end faces. The results show that aiming at the prediction accuracy of the temperature on the sealed end faces, the PSO-BP algorithm is higher than BP algorithm and GA-BP algorithm, which error interval is shortened about 40% and 29%, respectively. Subsequently, PSO-BP-GA algorithm is obviously better than BP-GA algorithm and GA-BP-GA algorithm, which validates its effectiveness and feasibility. It will help to lay a theoretical foundation for temperature evaluation on the sealed end faces.

Comparison of different optimization techniques for microwave-assisted biodiesel production

ABSTRACT Microwave-assisted biodiesel production from palm oil is becoming more and more important because of the greatly shortened reaction time and high energy-efficiency. In this novel research, an experimental study was conducted to optimize the experimental conditions of microwave-assisted biodiesel production. Response surface methodology (RSM), back propagation artificial neural network (BP-ANN) and Genetic algorithm improved BP-ANN (BPANN-GA) and were established and compared to optimize four operating parameters, including methanol/oil molar ratio (5:1–15:1), catalyst amount (0.75 –1.25 wt.%), temperature (60–70°C) and microwave time (20 –50 min). RSM, BP-ANN and BPANN-GA models were considered to be reliable in predicting biodiesel yield, BPANN-GA model (R2 = 0.9957) showed better accuracy than the RSM model (R2 = 0.9823) and BP-ANN model (R2 = 0.9867). The optimized process parameters were 14.9 methanol/oil molar ratio, 0.84 wt.% catalyst amount, 69°C temperature and microwave time of 49.3 min, the corresponding yield was 98.15%. The produced biodiesel properties were compared with ASTM D6571 standards. The results of the research are constructive for the optimization of microwave-assisted biodiesel experimental conditions when processing a small sample database.

Optimization of Helical Coil Tube Heat Exchanger: A Systematic Review

Helical coil heat exchanger is used in the food industries, air cooling industries etc.., due to its advantage of structural compact and high heat transfer coefficient. Many existing researches have been carried to improve the heat transfer efficiency and pressure drop. Mathematical model of the heat exchanger has been used to validate the performance of the heat exchanger. This paper involves in review the researches related to the optimization of helical coil tube heat exchanger. Some of the optimization methods such as Taguchi method, Genetic Algorithm (GA) and Multi-Objective Genetic Algorithm (MOGA) is used. From the analysis, Artificial Neural Network (ANN) and GA method has efficient performance in modelling and optimization of heat exchanger. Although, some methods involved in economic optimization of heat exchanger and these method has lower performance.

Bankruptcy Prediction: A Model Based on Cash Flow Ratios: Evidence From Selected European Countries

The importance of assessing the financial distress risk of a company is a topic that has been of central value in many different economic fields and since a long time. Until the twenty-first century, most of the studies were concentrated primarily on using mathematical and statistical methods to assess the health of businesses. Many of these studies employed either accounting-based ratios or cash flow-based ratios; even if there is not a unique conclusion, the use of cash flows seems to improve the predictive capacity of the models significantly. Especially in the last twenty-five years, methods derived from different fields started to be applied in forecasting corporate failures, such as artificial neural networks, genetic algorithms, and fuzzy logic.The objective of this study was to test the goodness of the discriminatory power of ratios based only on cash flows using a model that employs genetic algorithms and fuzzy logic. Five countries (Germany, Spain, France, Great Britain, Italy) and five Nace macro sectors (Agriculture, Industry, Services, Construction, Commerce and Food) have been considered in the analysis for a total of around 719-thousand companies. The model has proven to be well-performing on most of the countries and sectors that have been tested. The results obtained are almost all adequate; in particular, in Germany and Spain, results have been particularly good.The main weaknesses of this work are the limited availability of financial data in some countries and the time delay from the reporting of financial statement to the availability of the data through web services. It means that a large-scale risk assessment requires – being useful for the public and the private sectors – greater and faster disclosure of information at European level, and standardization of financial information transparency among countries.

Path Planning and Setup Orientation for Automated Dimensional Inspection Using Coordinate Measuring Machines

Inspection planning is considered an essential practice in the manufacturing industries because it ensures enhanced product quality and productivity. A reasonable inspection plan, which can reduce inspection costs and achieve high customer satisfaction, is therefore very important in the production industry. Considerations such as preparations for part inspection, measuring machines, and their setups as well as the measurement path are described in an inspection plan which is subsequently translated into part inspection machine language. Therefore, the measurement of any component using a coordinate measuring machine (CMM) is the final step preceded by several other procedures, such as the preparation of the part setup and the generation of the probe path. Effective measurement of components using CMM can only be done if the preceding steps are properly optimized to automate the whole inspection process. This paper has proposed a method based on artificial intelligence techniques, namely, artificial neural network (ANN) and genetic algorithm (GA), for fine-tuning output from the different steps to achieve an efficient inspection plan. A case study to check and validate the suggested approach for producing effective inspection plans for CMMs is presented. A decrease of nearly 50% was observed in the travel path of the probe, whereas the CMM measurement time was reduced by almost 25% during the actual component measurement. The proposed method yielded the optimum part setup and the most appropriate measuring sequence for the part considered.

Research on investment portfolio model based on neural network and genetic algorithm in big data era

With the maturity of neural network theory, it provides new ideas and methods for the prediction and analysis of stock market investment. The purpose of this paper is to improve the accuracy of stock market investment prediction, we build neural network model and genetic algorithm model, study the law of stock market operation, and improve the effectiveness of neural network and genetic algorithm. Through the empirical research, it is found that the neural network model can make up for the shortcomings of the traditional algorithm through the optimization of genetic algorithm.

Research on prediction model of thermal and moisture comfort of underwear based on principal component analysis and Genetic Algorithm–Back Propagation neural network

Abstract In order to improve the efficiency and accuracy of thermal and moisture comfort prediction of underwear, a new prediction model is designed by using principal component analysis method to reduce the dimension of related variables and eliminate the multi-collinearity relationship between variables, and then inputting the converted variables into genetic algorithm (GA) and BP neural network. In order to avoid the problems of slow convergence speed and easy falling into local minimum of Back Propagation (BP) neural network, this paper adopted GA to optimize the weights and thresholds of BP neural network, and utilized MATLAB software to program, and established the prediction models of BP neural network and GA–BP neural network. To verify the superiority of the model, the predicted result of GA–BP, PCA–BP and BP are compared with GA–BP neural network. The results show that PCA could improve the accuracy and adaptability of GA–BP neural network for thermal and moisture comfort prediction. PCA–GA–BP model is obviously superior to GA–BP, PCA–BP, BP, SVM and K-means prediction models, which could accurately predict thermal and moisture comfort of underwear. The model has better accuracy prediction and simpler structure.

Análisis del estado de arte de Técnicas de Soft Computing aplicadas a problemas de planificación de red en 5G.

En el presente artículo se realizó una revisión del estado del arte de la aplicación de las técnicas de Soft Computing en la resolución de problemas de planificación de redes 5G, para lo cual se clasificó las diferentes técnicas de soft computing existentes (redes neuronales, lógica difusa, algoritmos evolutivos) y de los trabajos e investigaciones realizados sobre el tema según sus autores, los modelos planteados y los métodos de solución. Adicionalmente se describieron las investigaciones más relevantes en donde se especifican técnicas para dar solución a los problemas de arquitecturas y funcionalidades cruciales en el desarrollo de esta tecnología, entre los cuales se resalta: encontrar una posición óptima para una Estación Base (BS) en un área de interés determinada, operar en las bandas de frecuencias múltiples deseadas mientras se mantiene una alta ganancia, limitar el consumo de energía en las infraestructuras de red 5G y tratar de incrementar la calidad del servicio al disminuir la probabilidad de bloqueo de llamadas. Finalmente se concluyó que las técnicas de soft computing más aplicadas a la solución de problemas de planificación de las redes 5G son lógica difusa, para limitar el consumo de energía en las infraestructuras de red 5G, además de las redes neuronales artificiales y algoritmos genéticos para la admisión de llamadas en redes 5G con la finalidad de incrementar la calidad del servicio al disminuir las interferencias.

PM2.5 Forecasting Based on Artificial Neural Network and Genetic Algorithm

PM2.5 Forecasting Based on Artificial Neural Network and Genetic Algorithm

Stability for feedback loops containing complex algorithms

With the advent of neural networks, fuzzy logic, genetic algorithms, and other soft computing methodologies, many researchers have demonstrated successful designs for intelligent control strategies that appear to outperform traditional linear feedback controls; however, industry has mostly ignored the new technology due to the lack of stability guarantees (required in most formal engineering risk-management processes). In this paper, we offer a Lyapunov-stability framework where one can place any arbitrary computational algorithm and still get guarantees of uniformly ultimately bounded (UUB) signals. We would expect an intelligent algorithm to be well-designed such that the proposed framework would not come into play unless unanticipated disturbances affect the system. But even if the intelligent algorithm was poorly designed, the resulting performance (inside our framework) would just look similar to that of a typical nonlinear neural-adaptive control. In our strategy, the intelligent algorithm trains a cerebellar model articulation controller (CMAC) neural network with arbitrarily bounded weights in order to achieve good performance, while a second CMAC trains in parallel using direct-adaptive-control laws in order to provide stability (even in the case of the first CMAC weights reaching their imposed bound). We test the strategy using a previously proposed ad hoc CMAC weight smoothing strategy, serving as the intelligent algorithm, with simulations and experiment controlling a two-link flexible-joint robot.

Cough Expired Volume and Cough Peak Flow Rate Estimation Based on GA-BP Method

Cough is a respiratory protective behavior for clearing the secretion. The cough process can be characterized by three features which are cough peak flow rate, peak velocity time, and cough expired volume. The cough expired volume (CEV) and the cough peak flow rate (CPFR) are important for medical diagnosis and cough effectiveness assessment. In this study, the CEV and CPFR values of 700 healthy participants were measured and collected by using a portable pulmonary function device. The gender, age, height, weight, and smoking status information of the 700 participants were also collected. Meanwhile, the integration of backpropagation neural network and genetic algorithm (GA-BP) method was developed to estimate CEV and CPFR values. The results showed that the estimation accuracy of GA-BP method exceeds 90%, which indicates that the GA-BP method could be effectively used for CEV and CPFR value estimation. Furthermore, the method proposed in this paper could be useful for medical diagnosis and medical device development.

Numerical investigation on the heat transfer performance and optimisation of a finned heat pipe using artificial neural networks and genetic algorithm

In this work, a lumped parameter model has been developed to study the thermal performance of a two-phase closed thermosyphon. The working fluids considered for the numerical study are water and R134a. The dimensions of the thermosyphon are, length 500 mm, ID (Inner diameter) 20 mm, and OD (outer diameter) 26 mm. The thermosyphon consists of three sections, namely, source (evaporator), adiabatic and sink (condenser), and their lengths are 200, 100 and 200 mm, respectively. The cross-sectional dimensions of fins are 5 mm × 1 mm and the length is 200 mm. Fins are placed internally at the condenser section. The mass of fluid to be charged into the thermosyphon with respect to evaporator volume, is known as the filling ratio of a thermosyphon, has been varied in this study. In the numerical model the fill ratio is varied between 20% and 80%, and fins are varied between 0 and 8. Finally, based on the parametric studies with the model an optimisation studies conducted with a combined Artificial Neural Network (ANN) and Genetic Algorithm (GA) approach to obtain the optimum conditions namely fill ratio and number of fins, at each heat input, at which the overall thermal resistance is minimum.

Research on fault tolerant control system based on optimized neural network algorithm

Due to the strict personnel control measures in COVID-19 epidemic, the control system cannot be maintained and managed manually. This puts forward higher requirements for the accuracy of its fault-tolerant performance. The control system plays an increasingly important role in the rapid development of industrial production. When the sensor in the system fails, the system will become unstable. Therefore, it is necessary to accurately and quickly diagnose the faults of the system sensors and maintain the system in time. This paper takes the control system as the object to carry out the fault diagnosis and fault-tolerant control research of its sensors. A network model of wavelet neural network is proposed, and an improved genetic algorithm is used to optimize the weights and thresholds of the neural network model to avoid the deficiencies of traditional neural network algorithms. For the depth sensor of a certain system, an online fault diagnosis scheme based on RBF (Radial Basis Function) neural network and genetic algorithm optimized neural network was designed. The disturbance fault, “stuck” fault, drift fault and oscillation fault of the depth sensor are simulated. Simulation experiments show that both online fault diagnosis schemes can accurately identify sensor faults and the genetic algorithm optimized neural network is superior to RBF neural network in both recognition accuracy and training time under the influence of COVID-19.

Design and analysis of novel bio inspired BELBIC and PSOBELBIC controlled semi active suspension

Passenger comfort, quality of ride and handling brings a lot of attention and concern of the automotive design engineers. Researchers are working and have introduced approaches like artificial neural network, genetic algorithm, fuzzy and hybrid control techniques have been proposed, modelled and analysed to check their impact on the above mentioned parameters. A new controller based on learning mode of human brain known as brain emotional learning based intelligent controller (BELBIC) and particle swarm optimised brain emotional learning based intelligent controller (PSOBELBIC) optimised by particle swarm optimisation (PSO) is introduced here. Six degrees of freedom (DOF) mathematical model of quarter car along with passenger and four degree of freedom half car is modelled and simulated. The PSOBELBIC gives the best performance with overall improvement of about 26% and 32.54% for speed bumps and 52.79%, 34.58% against random road.