Search results

The purpose of this paper is to propose a novel target automatic recognition method for unmanned aerial vehicle (UAV), which is based on backpropagation – artificial neural network (BP-ANN) algorithm, with the objective of optimizing the structure of backpropagation network, to increase the efficiency and decrease the recognition time. A hardware-in-the-loop system for UAV target automatic recognition is also developed.

The hybrid model of BP-ANN structure is established for aircraft automatic target recognition. This proposed method identifies controller parameters and reduces the computational complexity. Approaching speed of the network is faster and recognition accuracy is higher. This kind of network combines or better fuses the advantages of backpropagation artificial neural algorithm and Hu moment. with advantages of two networks and improves the speed and accuracy of identification. Finally, a hardware-in-the-loop system for UAV target automatic recognition is also developed.

The double hidden level backpropagation artificial neural can easily increase the speed of recognition process and get a good performance for recognition accuracy.

The proposed backpropagation artificial neural algorithm can be ANN easily applied to practice and can help the design of the aircraft automatic target recognition system. The standard backpropagation algorithm has some obvious drawbacks, namely, converging slowly and falling into the local minimum point easily. In this paper, an improved algorithm based on the standard backpropagation algorithm is constructed to make the aircraft target recognition more practicable.

A double hidden levels backpropagation artificial neural algorithm is presented for automatic target recognition system of UAV.

Combining the parallel and neural paradigms seems, at first glance, to be a natural process, since it is a methodology derived from the part played by the biological and mathematical behavior of a neuron. It is proposed that any neural algorithm is inherently a parallel application. The structure of a neural algorithm and the function of a neuron suggest the choice of the systolic approach. However, interest should be restricted only to those well‐known neural models such as the Hopfield and back‐propagation neural networks. It is also shown that the systolic approach is best suited to the parallelization of the patterns training phase of the neural algorithms in terms of mapping the two structures (systolic and neural networks).

The purpose of this paper is twofold: the selection of project portfolios through hybrid artificial neural network algorithms, feature selection based on grey relational analysis, decision tree and regression; and the identification of the features affecting project portfolio selection using the artificial neural network algorithm, decision tree and regression. The authors also aim to classify the available options using the decision tree algorithm.

In order to achieve the research goals, a project-oriented organization was selected and studied. In all, 49 project management indicators were chosen from A Guide to the Project Management Body of Knowledge (PMBOK Guide), and the most important indicators were identified using a feature selection algorithm and decision tree. After the extraction of rules, decision rule-based multi-criteria decision making matrices were produced. Each matrix was ranked through grey relational analysis, similarity to ideal solution method and multi-criteria optimization. Finally, a model for choosing the best ranking method was designed and implemented using the genetic algorithm. To analyze the responses, stability of the classes was investigated.

The results showed that projects ranked based on neural network weights by the grey relational analysis method prove to be better options for the selection of a project portfolio. The process of identification of the features affecting project portfolio selection resulted in the following factors: scope management, project charter, project management plan, stakeholders and risk.

This study presents the most effective features affecting project portfolio selection which is highly impressive in organizational decision making and must be considered seriously. Deploying sensitivity analysis, which is an innovation in such studies, played a constructive role in examining the accuracy and reliability of the proposed models, and it can be firmly argued that the results have had an important role in validating the findings of this study.

In the new era of highly developed Internet information, the prediction of the development trend of network public opinion has a very important reference significance for monitoring and control of public opinion by relevant government departments.

Aiming at the complex and nonlinear characteristics of the network public opinion, considering the accuracy and stability of the applicable model, a network public opinion prediction model based on the bald eagle algorithm optimized radial basis function neural network (BES-RBF) is proposed. Empirical research is conducted with Baidu indexes such as “COVID-19”, “Winter Olympic Games”, “The 100th Anniversary of the Founding of the Party” and “Aerospace” as samples of network public opinion.

The experimental results show that the model proposed in this paper can better describe the development trend of different network public opinion information, has good stability in predictive performance and can provide a good decision-making reference for government public opinion control departments.

A method for optimizing the central value, weight, width and other parameters of the radial basis function neural network with the bald eagle algorithm is given, and it is applied to network public opinion trend prediction. The example verifies that the prediction algorithm has higher accuracy and better stability.

To improve the quality of the additive manufacturing (AM) products, it is necessary to estimate surface roughness distribution in advance. Although surface roughness estimation has been previously studied, factors leading to the creation of a rough surface and a comprehensive test for model validation have not been adequately investigated. Therefore, this paper aims to establish a robust model using empirical data based on optimized artificial neural networks (ANNs) to estimate the surface roughness distribution in fused deposition modelling parts. Accordingly, process parameters such as time, cost and quality should be optimized in the process planning stage.

Process parameters were selected via a literature review of surface roughness estimation modelling by analytical and empirical methods, and then a specific test part was fabricated to provide a complete evaluation of the proposed model. The ANN structure was optimized by trial and error method and evolutionary algorithms. A novel methodology based on the combination of the intelligent algorithms including the ANN, linked to the particle swarm optimization (PSO) and imperialist competitive algorithm (ICA), was developed. The PSOICA algorithm was implemented to increase the capability of the ANN to perform much faster and converge more precisely to favorable results. The performances of the ANN models were compared to the most well-known analytical models at build angle intervals of equal size. The most effective process variable was found by sensitivity analysis. The validity of proposed model was studied comprehensively where different truncheon parts and medical case studies including molar tooth, skull, femur and a custom-made hip stem were built.

This paper presents several improvements in surface roughness distribution modelling including a more suitable method for process parameter selection according to the design criteria and improvements in the overall surface roughness of parts as compared to analytical methods. The optimized ANN based on the proposed advanced algorithm (PSOICA) represents precise estimation and faster convergence. The validity assessment confirms that the proposed methodology performs better in varied conditions and complex shapes.

This research fills an important gap in surface roughness distribution estimation modelling by using a test part designed for that purpose and optimized ANN models which uses purely empirical data. The novel PSOICA combination enhances the ability of the ANN to perform more accurately and quickly. The advantage in using actual surface roughness values is that all factors resulting in the creation of a rough surface are included, which is impossible if other methods are used.

Compares the predictive performance of artificial neural networks to hedonic pricing models, a more traditional valuation tool. The results document similar predictive performance evidenced from both techniques, which contradicts some of the earlier studies which support a position of artificial neural network superiority. Demonstrates that at least 18 per cent of the “normal” property predictions and over 70 per cent of the “outlier” property predictions contained valuation errors greater than 15 per cent of the actual sales price. The combination of these substantial errors and the model‐optimization costs incurred motivate a message of caution before artificial neural networks are adopted by the real estate valuation and/or lending industries.

The main purpose of this controller is to carryout irrigation by the farmers with renewable energy resources.

The proposed design includes the Deep learning based intelligent stand-alone energy management system used for irrigation purpose. The deep algorithm applied here is Radial basis function neural network which tracks the maximum power, maintains the battery as well as load system.

The Radial Basis Function Neural Network algorithm is used for carrying out the training process. In comparison with other conventional algorithms, this algorithm outperforms by higher efficiency and lower tracking time without oscillation.

It is little complex to implement the hardware setup of neural network in terms of training process but the work is under progress.

The practical hardware implementation is under progress.

If controller are implemented in a real-time environment, definitely it helps the human-less farming and irrigation process.

If this system is implemented in real-time environment, every farmer gets benefitted.

To research the feasibility in using artificial neural networks (ANN) and response surfaces (RS) techniques for reliability analysis of concrete structures.

The evaluation of the failure probability and safety levels of structural systems is of extreme importance in structural design, mainly when the variables are eminently random. It is necessary to quantify and compare the importance of each one of these variables in the structural safety. RS and the ANN techniques have emerged attempting to solve complex and more elaborated problems. In this work, these two techniques are presented, and comparisons are carried out using the well‐known first‐order reliability method (FORM), with non‐linear limit state functions. The reliability analysis of reinforced concrete structure problems is specially considered taking into account the spatial variability of the material properties using random fields and the inherent non‐linearity.

It was observed that direct Monte Carlo simulation technique has a low performance in complex problems. FORM, RS and neural networks techniques are suitable alternatives, despite the loss of accuracy due to approximations characterizing these methods.

The examples tested are limited to moderated large non‐linear reinforced concrete finite element models. Conclusions are drawn based on the examples.

Some remarks are outlined regarding the fact that RS and ANN techniques have presented equivalent precision levels. It is observed that in problems where the computational cost of structural evaluations (computing failure probability and safety levels) is high, these two techniques could improve the performance of the structural reliability analysis through simulation techniques.

This paper is important in the field of reliability analysis of concrete structures specially when neural networks or RS techniques are used.

The purpose of this paper is to define a new method (grey relational analysis (GRA)) for extracting pattern samples of dissolved gases in power transformer oil, then a hybrid algorithm of the back‐propagation (BP) network and fuzzy genetic algorithm‐artificial neural network (FGA‐ANN) is used to power transformer fault diagnosis based on extracted pattern samples.

The existing manners (e.g. international electro technical commission triple‐ratio method), in practice, have certain faultiness due to the ambiguity of the inference and insufficient standard for judgment. So GRA method is chosen to solve a problem of optimal pattern samples data, then a hybrid algorithm of the BP network and FGA‐ANN is developed to optimize initial weights and to enable fast convergence of the BP network, and lastly, this algorithm is applied to the classification of dissolved gas analysis (DGA) data and power transformer fault diagnosis.

If possible, the results should be accompanied by significance. For comparative studies, the proposed scheme does not require the three ratio code and high diagnosis accuracy is obtained. In addition, useful information is provided for future fault trends and multiple faults analysis.

Accessibility and availability of data are the main limitations which model will be applied.

This paper provides useful advice for power transformer fault diagnosis method based on DGA data.

The new method of optimal choice of options of pattern samples due to GRA. The paper is aimed at optimized samples data classified and abandons the traditional ratio method.

The purpose of this paper is to describe a novel diagnosis approach, using neural networks (NNs), which can be used to identify faulty nodes in distributed and multiprocessor systems.

Based on a literature‐based study focusing on research methodology and theoretical frameworks, the conduct of an ethnographic case study is described in detail. A discussion of the reporting and analysis of the data is also included.

This work shows that NNs can be used to implement a more efficient and adaptable approach for diagnosing faulty nodes in distributed systems. Simulations results indicate that the perceptron‐based diagnosis is a viable addition to present diagnosis problems.

This paper presents a solution for the asymmetric comparison model. For a more generalized approach that can be used for other comparison or invalidation models this approach requires a multilayer neural network.

The extensive simulations conducted clearly showed that the perceptron‐based diagnosis algorithm correctly identified all the millions of faulty situations tested. In addition, the perceptron‐based diagnosis requires an off‐line learning phase which does not have an impact on the diagnosis latency. This means that a fault set can be easily and rapidly identified. Simulations results showed that only few milliseconds are required to diagnose a system, hence, one can start talking about “real‐time” diagnosis.

The paper is first work that uses NNs to solve the system‐level diagnosis problem.