Artificial Intelligence Enhanced Reliability Assessment Methodology With Small Samples.

Due to the high price of the product and the limitation of laboratory conditions, reliability tests often get a small number of failed samples. If the data are not handled properly, the reliability evaluation results will incur grave errors. In order to solve this problem, this work proposes an artificial intelligence (AI) enhanced reliability assessment methodology by combining Bayesian neural networks (BNNs) and differential evolution (DE) algorithms. First, a single hidden layer BNN model is constructed by fusing small samples and prior information to obtain the 95% confidence interval (CI) of the posterior distribution. Then, the DE algorithm is used to iteratively generate optimal virtual samples based on the 95% CI and small samples trends. A reliability assessment model is reconstructed based on double hidden layers BNN model by combining virtual samples and test samples in the last stage. In order to verify the effectiveness of the proposed method, an accelerated life test (ALT) of the subsurface electronic control unit (S-ECU) was carried out. The verification test results show that the proposed method can accurately evaluate the reliability life of a product. And compared with the two existing methods, the results show that this method can effectively improve the accuracy of the reliability assessment of a test product.

Risk assessment of marine oil spills using dynamic Bayesian network analyses.

Oil spills are serious threats to the marine ecosystem. Especially when an oil spill is faced with extreme weather, the consequences might be more severe. Until now, no such researches focus on the risk of these extreme scenarios. This paper proposes a novel dynamic assessment method to quantify the risk of oil spills in extreme winds based on dynamic Bayesian networks (DBNs). The physical models of advection, spreading, evaporation, and dispersion are transformed into DBNs, and the vulnerability model is established according to coastline types and socio-economic resources. By integrating all the sub-models, the overall DBN to quantify the dynamic risk of oil spills occurring in extreme winds is obtained. The proposed method is demonstrated by the Laizhou Bay. The developed model is validated by a three-axiom-based approach. Temporal and spatial dynamics of risk caused by oil spills in potential locations could be calculated. Based on the developed DBN, the risk of the Laizhou Bay coast caused by oil spills in annual extreme wind speeds corresponding to different mean recurrence intervals is studied. In addition, the effects of the occurrence time of annual extreme winds are also researched.

Compressor fault diagnosis system based on PCA-PSO-LSSVM algorithm.

On the basis of the principal components analysis-particle swarm optimization-least squares support vector machine (PCA-PSO-LSSVM) algorithm, a fault diagnosis system is proposed for the compressor system. The relationship between the working principle of a compressor system, the fault phenomenon, and the root cause is analyzed. A fault diagnosis model is established based on the LSSVM optimized using PSO, the compressor fault diagnosis test experimental platform is used to obtain the fault signal of various fault occurrence states, and the PCA algorithm is employed to extract the characteristic data in the fault signal as input to the fault diagnosis model. The back-propagation neural network, the LSSVM algorithm, and the PSO-LSSVM algorithm are analyzed and compared with the proposed fault diagnosis model. Results show that the PCA-PSO-LSSVM fault diagnosis model has a maximum fault recognition efficiency that is 10.4% higher than the other three models, the test sample classification time is reduced by 0.025 s, the PCA algorithm can effectively reduce the input dimension, and the PSO-LSSVM fault diagnosis model based on the PCA algorithm for extracting features has a high recognition rate and accuracy. Therefore, the proposed fault diagnosis system can effectively identify the compressor fault and improve the efficiency of the compressor.

A probabilistic model of decision making regarding the use of chemical dispersants to combat oil spills in the German Bight.

Oil spills are one of the major threats to the marine environment in the German Bight (North Sea). In case of an accident, application of chemical dispersants would be one response option among others. Dispersion breaks oil slicks into small droplets which get then mixed into the water column. Removal of the oil from the water surface may reduce contamination of the coast. However, the window of opportunity for effective dispersant application is short and there are concerns about potential effects to the marine life. We propose a Bayesian network (BN) as an interactive and intuitive tool for responders to justify decisions on using chemical dispersants and possibly the provision of appropriate assets. The BN combines detailed sub-BNs for different criteria that govern the decision process. Expected drift trajectories are estimated based on comprehensive numerical ensemble simulations of hypothetical oil spills. Ecological impacts are represented prototypically, focusing on vulnerability of seabird concentrations to pollution in coastal areas. Dispersant effectiveness is estimated considering oil properties and weather conditions. Decision making is supposed to be based on expected satisfaction. The definition of what is considered satisfactory is of central importance for the whole analysis.

Implications of using chemical dispersants to combat oil spills in the German Bight - Depiction by means of a Bayesian network.

Application of chemical dispersants is one option for combatting oil spills, dispersing oil into the water column and thereby reducing potential pollution to coastal areas. Efficiency of dispersant application depends on oil characteristics, sea and weather conditions. Potential environmental impacts must also be taken into account. Referring to the German Bight region (North Sea), we show how probabilistic Bayesian network (BN) technology can integrate all these aspects to support contingency planning. Expected effects of chemical dispersion on oil spill drift paths are quantified based on comprehensive numerical ensemble simulations. Ecological impacts are represented just in simplified terms focusing on nearshore seabird distributions. The intuitive and interactive BN summarizes expected benefits from chemical dispersion depending on where and under which weather conditions a hypothetical pollution occurs.

Reliability evaluation of auxiliary feedwater system by mapping GO-FLOW models into Bayesian networks.

Bayesian network (BN) is a widely used formalism for representing uncertainty in probabilistic systems and it has become a popular tool in reliability engineering. The GO-FLOW method is a success-oriented system analysis technique and capable of evaluating system reliability and risk. To overcome the limitations of GO-FLOW method and add new method for BN model development, this paper presents a novel approach on constructing a BN from GO-FLOW model. GO-FLOW model involves with several discrete time points and some signals change at different time points. But it is a static system at one time point, which can be described with BN. Therefore, the developed BN with the proposed method in this paper is equivalent to GO-FLOW model at one time point. The equivalent BNs of the fourteen basic operators in the GO-FLOW methodology are developed. Then, the existing GO-FLOW models can be mapped into equivalent BNs on basis of the developed BNs of operators. A case of auxiliary feedwater system of a pressurized water reactor is used to illustrate the method. The results demonstrate that the GO-FLOW chart can be successfully mapped into equivalent BNs.

Real-time reliability evaluation methodology based on dynamic Bayesian networks: A case study of a subsea pipe ram BOP system.

A novel real-time reliability evaluation methodology is proposed by combining root cause diagnosis phase based on Bayesian networks (BNs) and reliability evaluation phase based on dynamic BNs (DBNs). The root cause diagnosis phase exactly locates the root cause of a complex mechatronic system failure in real time to increase diagnostic coverage and is performed through backward analysis of BNs. The reliability evaluation phase calculates the real-time reliability of the entire system by forward inference of DBNs. The application of the proposed methodology is demonstrated using a case of a subsea pipe ram blowout preventer system. The value and the variation trend of real-time system reliability when the faults of components occur are studied; the importance degree sequence of components at different times is also determined using mutual information and belief variance.

A Fault Diagnosis Methodology for Gear Pump Based on EEMD and Bayesian Network.

This paper proposes a fault diagnosis methodology for a gear pump based on the ensemble empirical mode decomposition (EEMD) method and the Bayesian network. Essentially, the presented scheme is a multi-source information fusion based methodology. Compared with the conventional fault diagnosis with only EEMD, the proposed method is able to take advantage of all useful information besides sensor signals. The presented diagnostic Bayesian network consists of a fault layer, a fault feature layer and a multi-source information layer. Vibration signals from sensor measurement are decomposed by the EEMD method and the energy of intrinsic mode functions (IMFs) are calculated as fault features. These features are added into the fault feature layer in the Bayesian network. The other sources of useful information are added to the information layer. The generalized three-layer Bayesian network can be developed by fully incorporating faults and fault symptoms as well as other useful information such as naked eye inspection and maintenance records. Therefore, diagnostic accuracy and capacity can be improved. The proposed methodology is applied to the fault diagnosis of a gear pump and the structure and parameters of the Bayesian network is established. Compared with artificial neural network and support vector machine classification algorithms, the proposed model has the best diagnostic performance when sensor data is used only. A case study has demonstrated that some information from human observation or system repair records is very helpful to the fault diagnosis. It is effective and efficient in diagnosing faults based on uncertain, incomplete information.

Application of Petri nets to performance evaluation of subsea blowout preventer system.

This paper presents an application of deterministic and stochastic Petri nets (DSPN) to evaluate the performance of subsea blowout preventer (BOP) system. The overall subsea BOP system is comprised of five mechanical subsystems and five electrical subsystems, which can be viewed as a series-parallel system. In regard to common cause failures, TimeNET 4.0 toolkit is utilized to develop and analyze the DSPN models. Availability and reliability of the subsea BOP system and its subsystems are obtained. Besides, the effects of failure rate and repair time of each component on system performance are researched.

Reliability analysis of the electrical control system of subsea blowout preventers using Markov models.

Reliability analysis of the electrical control system of a subsea blowout preventer (BOP) stack is carried out based on Markov method. For the subsea BOP electrical control system used in the current work, the 3-2-1-0 and 3-2-0 input voting schemes are available. The effects of the voting schemes on system performance are evaluated based on Markov models. In addition, the effects of failure rates of the modules and repair time on system reliability indices are also investigated.

Application of Bayesian networks in quantitative risk assessment of subsea blowout preventer operations.

This article proposes a methodology for the application of Bayesian networks in conducting quantitative risk assessment of operations in offshore oil and gas industry. The method involves translating a flow chart of operations into the Bayesian network directly. The proposed methodology consists of five steps. First, the flow chart is translated into a Bayesian network. Second, the influencing factors of the network nodes are classified. Third, the Bayesian network for each factor is established. Fourth, the entire Bayesian network model is established. Lastly, the Bayesian network model is analyzed. Subsequently, five categories of influencing factors, namely, human, hardware, software, mechanical, and hydraulic, are modeled and then added to the main Bayesian network. The methodology is demonstrated through the evaluation of a case study that shows the probability of failure on demand in closing subsea ram blowout preventer operations. The results show that mechanical and hydraulic factors have the most important effects on operation safety. Software and hardware factors have almost no influence, whereas human factors are in between. The results of the sensitivity analysis agree with the findings of the quantitative analysis. The three-axiom-based analysis partially validates the correctness and rationality of the proposed Bayesian network model.

Development of an automatic subsea blowout preventer stack control system using PLC based SCADA.

An extremely reliable remote control system for subsea blowout preventer stack is developed based on the off-the-shelf triple modular redundancy system. To meet a high reliability requirement, various redundancy techniques such as controller redundancy, bus redundancy and network redundancy are used to design the system hardware architecture. The control logic, human-machine interface graphical design and redundant databases are developed by using the off-the-shelf software. A series of experiments were performed in laboratory to test the subsea blowout preventer stack control system. The results showed that the tested subsea blowout preventer functions could be executed successfully. For the faults of programmable logic controllers, discrete input groups and analog input groups, the control system could give correct alarms in the human-machine interface.