Generative knowledge-based transfer learning for few-shot health condition estimation

In current studies of the belief rule base (BRB) model, the attributes are assumed to be fully reliable and the observation data are directly used as input. However, in engineering practice, the observation data may be affected by some disturbance factors, including the quality of the sensors and noise in the environment. Then, the reliability of observation data may be affected and the modeling accuracy of the BRB is therefore influenced. As such, a new BRB model with attribute reliability (BRB-r) is proposed in this paper. In particular, a calculation method of attribute reliability is given based on the statistical method. Moreover, to integrate the attribute reliability into the BRB-r, a new calculation method of matching degree is developed. The model's overall reliability denotes its ability to provide the correct result. When the attributes are unreliable, the overall reliability of the BRB-r is degraded. Thus, a calculation method for the overall reliability of the BRB-r is developed to support decision-making in engineering practice. A case study of the safety assessment of a diesel engine is conducted to demonstrate the efficiency of the proposed BRB-r model.

A new hierarchical belief-rule-based method for reliability evaluation of wireless sensor network

With the growing demand for high safety in industrial system, fault diagnosis has attracted more and more attention. Currently, belief rule base (BRB) has shown an excellent performance in modeling complex system, where the expert knowledge is used effectively. Existing BRB models are assumed that the inputs of the attributes are independent and the attribute correlation is not taken into account. However, in some engineering system, there is an obvious correlation among these attributes. The correlated attributes may produce redundant information which limits the abilities of attributes to express the accurate information of system. In this paper, a new BRB model with considering attribute correlation (BRB-c) is proposed. Moreover, a decoupling matrix is introduced to eliminate the redundant information from the attributes. The initial parameters of the decoupling matrix are given according to the expert knowledge. And then, when the inputs of the attributes are available, the parameters in the decoupling matrix are trained by an optimization model. The projection covariance matrix adaption evolution strategy is chosen as an optimization algorithm. A practical case study about fault diagnosis of oil pipeline is conducted and the results show that the BRB-c model can diagnose the leak size and leak time of oil pipeline accurately, which can demonstrate that the proposed model can be widely applied in engineering for fault diagnosis.

In engineering practice, to ensure the safety of complex engineering systems, the life prediction has attracted an increasing attention. In previous studies, few effective methods can be used to establish the life prediction model of a type of systems that have the following features: (1) the storage lives are much longer than the working lives, (2) lots of quantitative testing data are not available, (3) some expert knowledge can be used, and (4) the system’s performance degradation cannot be observed directly. As such, a life prediction method is constructed on the basis of belief rule base (BRB) and the accelerated degradation test, where two BRBs are involved, expressed by BRB_1 and BRB_2 respectively. Moreover, BRB_1 is employed to describe the relevance between the external stress and the hidden working state. BRB_2 is applied to establish the forecasting model of hidden working states. Once the predicted hidden working state achieves the boundary line, the corresponding working time is the life of the equipment and some measures need to be taken to maintain or change the equipment. Some uncertain parameters exist in the prediction model which can be determined by experts and it can lead to the inaccuracy of the model. Thus, an algorithm is proposed for updating parameters based on the maximum likelihood (ML). An experiment of the life prediction for aerospace relay is used to prove the reasonability of the life prediction model.KeywordsBelief rule base (BRB)Life predictionAccelerated degradation test

The combinatorial explosion problem is a great challenge for belief rule base (BRB) when a complex system has overnumbered attributes and/or referenced values for the attributes. This is because the BRB is conventionally constructed under the conjunctive assumption, conjunctive BRB, which requires covering each possible combination of all referenced values for all attributes. To solve this challenge, this study proposes a generic modeling, inferencing, and optimization approach for BRB under the disjunctive assumption, disjunctive BRB, that can significantly reduce its size. First, a disjunctive BRB is defined based on the mathematical description of the BRB space. The minimum size requirement for a disjunctive BRB is also discussed in comparison to a conjunctive one. Building on this, the generic disjunctive BRB modeling and inferencing procedures are proposed. Furthermore, an improved optimization model with further relaxed restrictions is constructed, and an optimization algorithm is developed in which only the new rule is optimized and its referenced values range is determined by the optimal solution in the former round optimization. The new optimization algorithm is more efficient with fewer variables and a more concise solution space. The results of three case studies confirm that by integrating both experts' knowledge and historic data, the modeling and inferencing processes can be well understood. Moreover, optimization can further improve the modeling accuracy while it facilitates downsizing the BRB in comparison with previous studies and other approaches.

When making maintenance decisions for diesel engines, the safety assessment plays a crucial role. Belief rule base (BRB) is a rule-based model embedded in expert knowledge with the advantage of being interpretable. However, due to the complexity of diesel engine systems and the impact of potential failures, their safety assessment faces the following challenges: (1) The interpretability of BRBs is often weakened during the building and optimization process. (2) Existing BRB models for safety assessment lack effective interpretability quantification methods. To address the above challenges, a diesel engine safety assessment method based on an interpretable BRB with the metric is proposed in this paper. First, an interpretability metric method for the interpretable BRB model for diesel engine safety assessment is proposed to quantify the interpretability of the entire modeling process. Second, a new statistical-based method for calculating the credibility of expert knowledge is proposed. Third, on the basis of these, three new interpretability maintenance strategies are designed to correct behaviors that destroy the interpretability of the model. The validity of the proposed model is verified through a case study of the safety assessment of a WD615 diesel engine.

Online updating belief rule based system for pipeline leak detection under expert intervention

Running gear, which is one of crucial element of high-speed train, is vital to security and dependability of the whole high-speed train. For the sake of guaranteeing normal running of running gear, it is very necessary to accurately estimate the health state of running gear. Consequently, this thesis presents a novel state evaluation model based on belief rule base (BRB). Considering the problem of combinatorial explosion in BRB. In the proposed evaluation model based on BRB and canonical correlation analysis (CCA), which is used to select core attributes by calculating the correlation between features. The results demonstrate that the model based on BRB and CCA can exactly estimate the health state of running gear.

Health state prediction is an effective way to improve the reliability for sensors. In the process of sensor degradation, it is difficult to obtain more effective monitoring data. And in the classification of health states, how to identify the adjacent state is also a problem. This paper proposed a health state prediction model based on belief rule base (BRB) and membership function (MF), which is called BRB-MF. In the model, BRB can make full use of expert knowledge and poor effective data. In the prediction results of BRB, it may be not completely logical or not entirely appropriate facing adjacent states of sensor. In order to solve the problem, MF is used to continue the analysis of the predicted results of BRB. In the BRB-MF model, the covariance matrix adaptation evolutionary strategies (CMA-ES) optimization algorithm is used to update the model parameters to make up for the uncertainty of expert knowledge. In the end, the brightness sensor of the rail vehicle LED lighting system is taken as a case study. The results show that the BRB-MF model can predict the health state of sensor with a high accuracy and a reasonable state.

Data-driven Extraction Method of Belief Rule for Reagent Addition in Antimony Rougher Flotation

Correlation-oriented complex system structural risk assessment using Copula and belief rule base

A new aeronautical relay health state assessment method based on generic belief rule base with attribute reliability

Research and application of park multi-source synergistic and complementary heating system and its coordinated control method

Pneumonia remains a leading cause of mortality among children under five years of age, accounting for approximately 1.6 million deaths annually. Early detection is critical for reducing child mortality; however, traditional diagnostic methods rely on trained personnel and medical resources, often scarce in low-resource settings. Machine learning has emerged as a promising approach to pneumonia detection, but its effectiveness is hindered by the limited availability of labeled data required for training robust models. This study reviews existing machine learning techniques for pneumonia classification, including convolutional neural networks (CNNs), transfer learning, and few-shot learning approaches. Furthermore, we propose an optimized few-shot learning model that integrates Siamese networks with transfer learning to improve pneumonia detection using minimal labeled data. The model leverages MobileNetV3 as a pre-trained feature extractor, producing high-quality embeddings that enhance similarity learning within the Siamese framework. Additionally, triplet loss is incorporated to ensure a more discriminative embedding space, facilitating robust classification. Experimental results indicate that our model outperforms conventional machine learning classifiers, achieving an accuracy of 92.04%, precision of 91.20%, recall of 90.32%, and F1-score of 90.09%. The integration of Siamese networks with triplet loss enhances generalization while mitigating overfitting, making it particularly suitable for resource-constrained environments. These findings highlight the effectiveness of few-shot learning in pneumonia detection, offering a computationally efficient and scalable solution that bridges critical gaps in medical imaging.

Safety assessment is one of important aspects in health management. In safety assessment for practical systems, three problems exist: lack of observation information, high system complexity and environment interference. Belief rule base with attribute reliability (BRB-r) is an expert system that provides a useful way for dealing with these three problems. In BRB-r, once the input information is unreliable, the reliability of belief rule is influenced, which further influences the accuracy of its output belief degree. On the other hand, when many system characteristics exist, the belief rule combination will explode in BRB-r, and the BRB-r based safety assessment model becomes too complicated to be applied. Thus, in this paper, to balance the complexity and accuracy of the safety assessment model, a new safety assessment model based on BRB-r with considering belief rule reliability is developed for the first time. In the developed model, a new calculation method of the belief rule reliability is proposed with considering both attribute reliability and global ignorance. Moreover, to reduce the influence of uncertainty of expert knowledge, an optimization model for the developed safety assessment model is constructed. A case study of safety assessment of liquefied natural gas (LNG) storage tank is conducted to illustrate the effectiveness of the new developed model.