Narrow Results

Aerospace systems are highly complex and critical, with stringent reliability requirements to ensure safe operations. The maintenance and management of these systems are of paramount importance, as any fault or failure can lead to significant consequences. Traditional fault prediction methods rely on manual inspection and rule-based systems, which are inefficient, error-prone, and reactive. The primary objective is to develop an intelligent fault prediction and reliability optimization framework for aerospace systems using DL techniques. The data are gathered from various aerospace systems, including sensor data, maintenance logs, and failure reports, sourced from operational aircraft and aerospace machinery over several years. Data were preprocessed using data cleaning and normalization methods. The research proposed a novel intelligent beluga whale optimized Elman neural network (IBWO-ENN) to identify and predict faults in aerospace systems. IBWO is used for feature selection, and ENN is employed to predict faults in aerospace systems. The aerospace propulsion system demonstrates the efficacy of the suggested model, achieving significant improvements in fault prediction accuracy (98.39%), precision (98.54%), and F1-score (98.23%) and system reliability compared to traditional methods. The research contributes to the urgent demand for improved safety and operational efficiency in increasingly complex aircraft settings by presenting a unique DL-based framework for fault prediction and reliability improvement of aerospace systems.

The increasing reliance on New Energy Vehicles (NEVs) has highlighted the need for efficient fault prediction and maintenance strategies to ensure their optimal performance and longevity. This study introduces the Refined Puffin Algorithm-refined Scalable Random Forest Tree (RPA-SRFT), a novel hybrid model designed to predict faults and optimize maintenance decisions in NEVs. The dataset used fault prediction and optimization in NEV. It covers multiple vehicle models and includes real-time operational data for fault prediction. Data preprocessing includes cleaning and normalization to handle missing values, remove noise, and scale the data to ensure consistency across features. Feature extraction is performed using Principal Component Analysis (PCA), which reduces dimensionality while retaining key information necessary for accurate fault prediction. The Refined Puffin Algorithm (RPA) is applied for fine-tuning the hyperparameters of the Scalable Random Forest Tree (SRFT), ensuring the model is well-adapted to large-scale NEVs data. The implementation of the RPA-SRFT algorithm in Python resulted in a significant improvement in fault prediction, with recall (97.60%), F1-score (98%), accuracy (98.50%), and precision (98%) outperforming conventional models in terms of both predictive performance and scalability. By leveraging advanced machine learning technique and big data analysis model, it can proactively predict faults and optimize maintenance schedules, resulting in reduced downtime and lower maintenance costs.

The power communication network is the key to ensuring the safe operation of the distribution network, and how to quickly and accurately predict faults is a challenging task, especially considering the variable topology of the distribution network. To address this issue, common faults will be modeled and simulated to predict faults in power communication systems. In this study, considering the generalized Laplacian smoothing filters and the long sequence representation capability of the Transformer, an adaptive graph encoder based on historical performance graph embedding is proposed and used for fault prediction in power communication networks. The proposed method consists of two modules: (1) To better alleviate high-frequency noise in node features, a carefully designed Laplacian smoothing filter is first applied. (2) Adopting a transformer-based adaptive encoder to iteratively enhance filtering characteristics for better node embedding. The performance of the proposed method in fault prediction tasks is tested using a dataset collected from a real power communication network. The experimental results show that the proposed method consistently outperforms other fault prediction methods in terms of performance.

Today’s development environment has changed drastically; the development periods are shorter than ever and the number of team members has increased. Consequently, controlling the activities and predicting when a development will end are difficult tasks. To adapt to changes, we propose a generalized software reliability model (GSRM) based on a stochastic process to simulate developments, which include uncertainties and dynamics such as unpredictable changes in the requirements and the number of team members. We assess two actual datasets using our formulated equations, which are related to three types of development uncertainties by employing simple approximations in GSRM. The results show that developments can be evaluated quantitatively. Additionally, a comparison of GSRM with existing software reliability models confirms that the approximation by GSRM is more precise than those by existing models.

Software defect prediction suffers from the class-imbalance. Solving the class-imbalance is more important for improving the prediction performance. SMOTE is a useful over-sampling method which solves the class-imbalance. In this paper, we study about some problems that faced in software defect prediction using SMOTE algorithm. We perform experiments for investigating how they, the percentage of appended minority class and the number of nearest neighbors, influence the prediction performance, and compare the performance of classifiers. We use paired t-test to test the statistical significance of results. Also, we introduce the effectiveness and ineffectiveness of over-sampling, and evaluation criteria for evaluating if an over-sampling is effective or not. We use those concepts to evaluate the results in accordance with the evaluation criteria for the effectiveness of over-sampling.

The results show that they, the percentage of appended minority class and the number of nearest neighbors, influence the prediction performance, and show that the over-sampling by SMOTE is effective in several classifiers.

With the explosion of 5G network scale, the network structure becomes increasingly complex. During the operation of the network devices, the probability of anomalies or faults increases accordingly. Network faults may lead to the disappearance of important information and cause unpredictable losses. The prediction of network faults can enhance the quality of network services and reduce economic loss. In this paper, we propose the concept of 4D features and use the BERT algorithm to extract semantic features, the graph neural network algorithm to extract network topology information, and the Temporal Convolutional Network (TCN) algorithm to extract time series. Based on this, we propose Fault Prediction based on GraphSage and TCN (GTFP), an end-to-end solution of network fault alarm prediction, which is based on GraphSage and TCN (GTCN), a hybrid algorithm of a graph neural network and the TCN model. Our solution takes the historical alarm data as input. First, we filter out the alarm noises irrelevant to the faults through data cleaning. Then, we employ feature engineering to extract the valid alarm features, including the statistical features of the network alarm information, the semantic features of the alarm texts, the sequential features of the alarms and the network topology features of the nodes where the alarms are located. Finally, we use GTCN to predict future fault alarms based on the extracted features. Experiments on the alarm data of a real service system show that GTFP performs better than the state-of-the-art algorithms of fault alarm prediction.

The size and complexity of software products have expanded significantly, prompting researchers to focus on quality assessment by quantitatively assessing software product’s reliability growth. Several time dependent software reliability growth models (SRGMs) have been produced and are available in this regard. The authors evaluated errors of varied severity, but they believed that low severity faults would be repaired first, followed by high severity faults. Furthermore, the authors assumed in the change point model that the team changed their strategy, i.e., the fault detection rate changed after a particular time but making a plan and executing that strategy takes a long time, and as a result, the industry confronts a delay in software launching. As a result, two fault categories (low and high severity) are considered in this study, and two types of testing effort function (TEF) related to fault detection rate are presented to detect and fix them. Because parameters have such a large impact on model performance, their sensitivity is also investigated.

In this paper, a fault prediction method for oil well equipment based on the analysis of time series data obtained from multiple sensors is proposed. The proposed method is based on deep learning (DL). For this purpose, comparative analysis of single-layer long short-term memory (LSTM) with the convolutional neural network (CNN) and stacked LSTM methods is provided. To demonstrate the efficacy of the proposed method, some experiments are conducted on the real data set obtained from eight sensors installed in oil wells. In this paper, compared to the single-layer LSTM model, the CNN and stacked LSTM predicted the faulty time series with a minimal loss.