Multi-source Uncertainties Research Articles

Multi-attribute group decision-making considering opinion dynamics

Multi-attribute group decision-making (MAGDM) is a complicated cognitive process that involves evaluation of opinion expression, information fusion, and analysis of multi-source uncertainties. In the incipient stage of MAGDM, decision makers (DMs) express their opinions on alternatives for each attribute and tend to interact with others. Despite the opinions of DMs are dynamically evolved, the traditional information fusion techniques are always relying on the data acquired at a static decision-making time point, which leads to the loss of information. In this study, we introduce a novel MAGDM method considering opinion dynamics, which employs the 2-tuple linguistic model for the representation of linguistic judgements and the technique for order preference by similarity to an ideal solution (TOPSIS) as the decision-making framework, to reduce the loss of information from the dimension of opinion formation. Moreover, a modified opinion dynamics model is as well as developed by extending the hypothesis of bounded confidence, where the opinion similarity, the credibility of DMs, and the human bounded rationality are collectively regarded as influential factors within the process of opinion evolution. Subsequently, three simulations are carried out to verify the feasibility of the extended bounded confidence model. And finally, a case study of supplier selection, as a typical MAGDM problem, is implemented and a comparison analysis is conducted to demonstrate the rationality of the proposed method.

A support vector regression (SVR)-based method for dynamic load identification using heterogeneous responses under interval uncertainties

A support vector regression (SVR)-based method is presented in this paper, which aims to reconstruct the uncertain dynamic load using heterogeneous responses. Essentially, the technology of load identification can be regarded as an issue of nonlinear regression. Considering the incomplete limitation of the single response signal, the complicated relationship between the dynamic load and heterogeneous responses for a specific structure may be determined through some support vectors in simulated training samples based on the SVR algorithm. In view of multi-source uncertainties, an interval dimension-by-dimension (D–D) method is investigated, which can approximate the load changing concerning the each-dimensional uncertain parameter by Legendre orthogonal polynomials. Then, the dynamic load boundaries can be calculated through the maxima/minima of the approximation polynomial in the time history. The validity and specificity of the proposed methodology are clarified by 3 numerical examples. In addition, the results indicate that the proposed method can be utilized to identify the interval of dynamic load with outstanding accuracy and efficiency.

An asymptotic stochastic response surface approach to reliability assessment under multi-source heterogeneous uncertainties

An asymptotic stochastic response surface method for structural reliability assessment under multi-source heterogeneous uncertainties is proposed in this study. The fatigue reliability of an aeroengine Curvic coupling under the uncertainties from material, contact surface state, and part dimension deviation is used to motivate the methodology development. A randomized block design of experiments for numerical models are employed to obtain the fatigue strain responses under different variations and combinations of the uncertain variables. A stochastic response surface model is constructed, and the analytical forms of the mean and bound predictions of the fatigue strain are derived. By further coupling the fatigue strain into the low-cycle fatigue model, the fatigue life under multi-source uncertainties is obtained. The stochastic moment approximation is used to derive the asymptotic approximation of the fatigue life distribution, allowing for efficient reliability and risk-informed lifetime prediction. The variance of the fatigue life contributed by individual uncertain sources is naturally resolved in the method, providing a quantitative measure for uncertainty control and management. The adequacy of the stochastic response surface model and the accuracy of the asymptotic reliability assessment result are verified using statistical testing and the Monte Carlo method, respectively.

A Continual Learning Framework for Uncertainty-Aware Interactive Image Segmentation

Deep learning models have achieved state-of-the-art performance in semantic image segmentation, but the results provided by fully automatic algorithms are not always guaranteed satisfactory to users. Interactive segmentation offers a solution by accepting user annotations on selective areas of the images to refine the segmentation results. However, most existing models only focus on correcting the current image's misclassified pixels, with no knowledge carried over to other images. In this work, we formulate interactive image segmentation as a continual learning problem and propose a framework to effectively learn from user annotations, aiming to improve the segmentation on both the current image and unseen images in future tasks while avoiding deteriorated performance on previously-seen images. It employs a probabilistic mask to control the neural network's kernel activation and extract the most suitable features for segmenting images in each task. We also apply a task-aware embedding to automatically infer the optimal kernel activation for initial segmentation and subsequent refinement. Interactions with users are guided through multi-source uncertainty estimation so that users can focus on the most important areas to minimize the overall manual annotation effort. Experiments are performed on both medical and natural image datasets to illustrate the proposed framework's effectiveness on basic segmentation performance, forward knowledge transfer, and backward knowledge transfer.

Multi-objectives Optimization Design of A-type Frame in an Electric Mining Dump Truck Considering Multi-Source Uncertainties Based on the Interval Method

In order to improve fuzzy fatigue reliability and make weight reduction of A-type frame in an electric mining dump truck considering multi-source uncertainties from design, manufacturing and use stage, a multi-objectives optimization method based on the interval theory was proposed in this paper. The thickness of four steel plates in welded A-type frame was regarded as certain design variables, while the elastic modulus and density of material was considered as uncertain variables as well as the load at front traction joint. The relationship between optimization objectives and variables was constructed by the response surface method, and was transformed by the interval method. Then, the optimization problem was solved by the non-dominated sorting genetic algorithm, which was found that the fuzzy fatigue reliability reached up to 98.45 % at the expense of some weight.

A dynamic force reconstruction method based on modified Kalman filter using acceleration responses under multi-source uncertain samples

A novel method for dynamic force reconstruction is developed in this paper to identify the time history of uncertain force for the linear time-invariant system, which combines the modified Kalman filter and non-probabilistic uncertainty analysis. For the force reconstruction issue, some acceleration responses are regarded as the observation information. The optimal estimation of the system state vector and unknown input force vector are deduced recursively on the basis of minimum variance unbiased estimate, along with the covariance matrices updating of force reconstruction error and state estimation error under the framework of modified Kalman filter. Considering multi-source uncertainties in inherent characteristics and external environment, the multidimensional interval model, which is transformed by the principal component analysis (PCA) method, is involved to envelope experiential samples of uncertain parameters. To enhance the accuracy and efficiency of uncertainty propagation, the Chebyshev orthogonal polynomial is adopted to approximate the unknown force and to obtain the force interval boundaries. Eventually, the validity and feasibility of the proposed methodology are clarified by three numerical examples: a spring-oscillator, a plane truss and an equivalent rudder structure. The results indicate that the proposed method can be utilized to reconstruct the uncertain force interval with the interference of system uncertainties and measurement noise.

Real-Time Autonomous Residential Demand Response Management Based on Twin Delayed Deep Deterministic Policy Gradient Learning

With the roll-out of smart meters and the increasing prevalence of distributed energy resources (DERs) at the residential level, end-users rely on home energy management systems (HEMSs) that can harness real-time data and employ artificial intelligence techniques to optimally manage the operation of different DERs, which are targeted toward minimizing the end-user’s energy bill. In this respect, the performance of the conventional model-based demand response (DR) management approach may deteriorate due to the inaccuracy of the employed DER operating models and the probabilistic modeling of uncertain parameters. To overcome the above drawbacks, this paper develops a novel real-time DR management strategy for a residential household based on the twin delayed deep deterministic policy gradient (TD3) learning approach. This approach is model-free, and thus does not rely on knowledge of the distribution of uncertainties or the operating models and parameters of the DERs. It also enables learning of neural-network-based and fine-grained DR management policies in a multi-dimensional action space by exploiting high-dimensional sensory data that encapsulate the uncertainties associated with the renewable generation, appliances’ operating states, utility prices, and outdoor temperature. The proposed method is applied to the energy management problem for a household with a portfolio of the most prominent types of DERs. Case studies involving a real-world scenario are used to validate the superior performance of the proposed method in reducing the household’s energy costs while coping with the multi-source uncertainties through comprehensive comparisons with the state-of-the-art deep reinforcement learning (DRL) methods.

Uncertainty assessment of a 3D geological model by integrating data errors, spatial variations and cognition bias

A 3D geological structural model is an approximation of an actual geological phenomenon. Various uncertainty factors in modeling reduce the accuracy of the model; hence, it is necessary to assess the uncertainty of the model. To ensure the credibility of an uncertainty assessment, the comprehensive impacts of multi-source uncertainties should be considered. We propose a method to assess the comprehensive uncertainty of a 3D geological model affected by data errors, spatial variations and cognition bias. Based on Bayesian inference, the proposed method utilizes the established model and geostatistics algorithm to construct a likelihood function of modeler’s empirical knowledge. The uncertainties of data error and spatial variation are integrated into the probability distribution of geological interface with Bayesian Maximum Entropy (BME) method and updated with the likelihood function. According to the contact relationships of the strata, the comprehensive uncertainty of the geological structural model is calculated using the probability distribution of each geological interface. Using this approach, we analyze the comprehensive uncertainty of a 3D geological model of the Huangtupo slope in Badong, Hubei, China. The change in the uncertainty of the model during the integration process and the structure of the spatial distribution of the uncertainty in the geological model are visualized. The application shows the ability of this approach to assess the comprehensive uncertainty of 3D geological models.

Kinematic Reliability Analysis of Planar Metamorphic Mechanism with Multi-source Uncertainties

Kinematic Reliability Analysis of Planar Metamorphic Mechanism with Multi-source Uncertainties

A novel framework for reliability assessment of payload fairing separation considering multi-source uncertainties and multiple failure modes

Payload fairing separation is a critical process for various aerospace structures. To further guarantee the safety of separation, multi-source uncertainties in the processes of manufacturing, fabrication and assembly need to be considered in the stage of structural design. Reliability assessment is a powerful tool to deal with multi-source uncertainties, however, the high computational cost of structural analyses is the main challenge of the numerical simulation of payload fairing separation. In this paper, a novel framework for reliability assessment of payload fairing separation is proposed. In general, since not all potential failure modes will be happened for a certain design, sampling method is utilized to identify the active failure mode firstly. Then, an efficient global sensitivity analysis method is used for the dimension reduction to simply the high dimensional problem. And then, the MPP-based simplified bivariate dimension reduction method (MPP-based SBDRM) is proposed to improve the accuracy of MPP-based DRM for calculating the failure probability. In the proposed method, the augmented step size adjustment (ASSA) method is employed to ensure the robustness and efficiency of the most probable point (MPP) searching process. To further improve the efficiency, an adaptive surrogate model is constructed to replace the time-consuming finite element analysis (FEA). Two mathematical benchmarks are used to validate the performance of the proposed method. Finally, a typical practical payload fairing demonstrates that the proposed framework is able to assess the reliability of fairing separation in an efficient and accurate manner.

A Reliability based Multidisciplinary Design Optimization Method with Multi-Source Uncertainties

The complexity of engineering systems is increasing greatly, and more coupled disciplines and multi-source uncertainties are involved in the design and development of complex engineered systems (CESs). Actually, uncertainties are ubiquitous in design of CESs, which can be mainly classified into alteatory uncertainty (AU) and epistemic uncertainty (EU). To gain high reliability and safety of CESs, the Reliability-based MDO (RBMDO) which considers uncertainties of design variables and parameters has become a hot research topic. In this paper, the quantification of aleatory and epistemic uncertainties based on the probability theory and convex set-theory, multidisciplinary comprehensive reliability evaluation index, adaptive collaborative optimization strategy based on the intelligent optimization algorithm, sequential multidisciplinary reliability analysis method based on the concurrent subspace optimization (CSSO) and performance measure approach (PMA), and hierarchical and hybrid sequential optimization and reliability assessment (HSORA) RBMDO strategy under multi-source uncertainties have been researched. All of the researches can expand and improve the RBMDO theoretical system, and also provide the effective method for the design and optimization of CESs under multi-source uncertainties.

Calculation of hybrid reliability of turbine disk based on self-evolutionary game model with few shot learning

A turbine disk design based on uncertainty quantifies the risks and improves the structural reliability. An optimized design issue related to turbine disk fatigue life reliability with subjective and objective multi-source uncertainties is a three-layer nested analysis process, where each iterative step requires a double-layer nested analysis containing probability and improbability. A self-evolution game model is proposed to solve the efficiency problem of the optimized design issue under a specific accuracy requirement. In the initial state, only a few sample points are needed to start the game. In the process of its evolution, the most valuable points are searched automatically, and performance functions are identified only at the points where the game diverges. The maximum failure probability is obtained after game consistency is achieved. Accuracy and efficiency are verified with typical numerical examples. The method is adopted to analyze the hybrid low cycle fatigue reliability of a real turbine disk. It is found that, owing to the elimination of an unnecessary artificial distribution hypothesis, the reliability is lower under the same safety life, and the results are more in line with engineering practice.

Performance margin-based reliability analysis for aircraft lock mechanism considering multi-source uncertainties and wear

The failures of the lock mechanism of the aircraft landing gear door can impede the retraction and extension processes of the landing gear and cause severe accidents; thus, the aircraft lock mechanism is always required with extremely high reliability. During manufacturing and usage stages, uncertainties are ubiquitous and performance degradation is inevitable, and they influence the reliability of the aircraft lock mechanism. In this paper, based on the reliability science principles, a performance margin-based reliability analysis considering performance degradation caused by wear and multi-source uncertainties is developed for an aircraft lock mechanism. Firstly, the performance margin model with the degradation of the aircraft lock mechanism is constructed based on the functional principles and the influence mechanism of wear. Then, we analyze the essence of each uncertainty source including the uncertainties in manufacturing imperfections, material properties, operational and environmental stresses, and performance thresholds, and quantify each uncertainty with a probability distribution. Finally, the reliability model is established. A numerical study of an aircraft lock mechanism is conducted and the reliability sensitivity analysis is implemented. The results show that the proposed method can provide guidance to the design and manufacturing processes of the aircraft lock mechanism to meet reliability requirements.

Barrier Lyapunov Functions-Based Adaptive Fault Tolerant Control for Flexible Hypersonic Flight Vehicles With Full State Constraints

One of the key problems for space vehicles is how to deal with the contradiction between the control constraints and the disturbances from multiple resources. This paper focuses on the adaptive full state constrained controller design problem for the flexible air-breathing hypersonic vehicles with multisource uncertainties. In simultaneous presence of the aerodynamical uncertainties, the modeling errors, the external disturbances, and the contingent actuator failures, an adaptive fault-tolerant control scheme is put forward in the context of dynamic surface control. Then, the barrier Lyapunov functions are utilized to guarantee that the constraints on the velocity, the flight path angle, the altitude, and the pitch rate are never violated. In virtue of the bound estimation approach, the multisource disturbances are effectively dealt with. Finally, a number of illustrative examples are provided to demonstrate the effectiveness of the proposed methodology.

Probabilistic modeling of uncertainties in fatigue reliability analysis of turbine bladed disks

Turbine bladed disks normally operate under complex loadings coupling with uncertainties originate from multiple sources, including material variability, load variation and geometrical uncertainty. The influence of these uncertainties on mechanical response of engineering components are critical for their fatigue assessment and reliability evaluation. In this work, a general framework for fatigue reliability analysis is developed by coupling the Latin hypercube sampling with FE analysis to describe the combined effects of multi-source uncertainties. Fatigue reliability analysis of a full-scale bladed disk under multi-source uncertainties was performed as well as sensitivity analysis for fatigue design. In order to describe the manufacturing errors or tolerances, random dimensions are inputted. Comparing the predicted fatigue lifetime distributions with/without geometrical uncertainty, it shows that geometrical uncertainty matters in structural fatigue reliability. Particularly, sensitivity analysis indicates that the geometrical uncertainty exerts more critical influences on the fatigue lifetime and reliability of the turbine bladed disk than others. The sensitivity factors of three typical dimensions emerges the influence of designed sizes and dimensional tolerances on the failure probability, which provides a reference for engineering design.

Nonlinear disturbance observer based adaptive super-twisting sliding mode control for generic hypersonic vehicles with coupled multisource disturbances

In practical, there often exist structural uncertainties, measurement errors, and the aerodynamic perturbations in the flight control systems of the hypersonic vehicles (HSVs). The control performance degradation and loss of stability may be induced if the coupled multisource uncertainties are ignored. To address this problem, this paper develops a novel multivariable robust adaptive sliding mode control structure. Benefiting from the second-order characteristics of the super-twisting controller, the uncertainties from the unknown nonlinearities and the lumped disturbances can be surmounted respectively. Meanwhile, the finite time convergence can be obtained and the continuity of the control signals can be guaranteed. Moreover, to cope with the unknown upper bounds of lumped uncertainties, a novel update law is developed by analyzing the connection between the control gains and the boundaries. Moreover, for the purpose of compensating the mismatched uncertainties in HSVs, a robust disturbance observer is introduced into the sliding surface such that the sliding-mode dynamics are stable with L2−L∞ performance. The closed-loop control system is proved to be globally ultimately stable. Simulation results indicate that the proposed control scheme works well compared with the existing methods.

Dissipativity-Based Disturbance Attenuation Control for T-S Fuzzy Markov Jumping Systems With Nonlinear Multisource Uncertainties and Partly Unknown Transition Probabilities.

This article is concerned with the dissipativity-based disturbance attenuation control for a class of Takagi-Sugeno (T-S) fuzzy Markov jump systems (FMJSs) suffering from nonlinear multisource disturbances. The considered system possesses nonlinear and stochastic jumping disturbances generated by multiple sources, constituting the main challenge to control design and dissipativity analysis. By proposing an adaptive fuzzy disturbance observer and a hybrid feedback controller, a novel fuzzy disturbance attenuation control structure has been constructed. In terms of strict linear matrix inequalities (LMIs), a new sufficient condition is established to guarantee the (Z, Y, X)-ε- dissipative and stochastic exponentially stability of the closed-loop FMJSs. Furthermore, for the concerned FMJSs with partly unknown transition probabilities, the sufficient conditions are also derived and the gains of controller or observer can be computed immediately. Finally, a numerical example is provided to verify the effectiveness of the proposed theory.

A radial basis function artificial neural network (RBF ANN) based method for uncertain distributed force reconstruction considering signal noises and material dispersion

A radial basis function artificial neural network (RBF ANN) based method for distributed dynamic force reconstruction is presented in this study considering multi-source uncertainties. Foremost, the distributed forces are approximated by truncated Legendre orthogonal polynomial in the time history. The RBF ANN implementation is introduced, whereby the normalized structural responses are deemed as input variables, and the normalized polynomial coefficients are regarded as output variables. To improve the generalization of the trained RBF ANN, the structural responses and polynomial coefficients with respect to different types of distributed dynamic forces at every instant are defined as the samples. Thus, the coefficients of the distributed force to be identified may be ascertained after inputting the corresponding responses. In terms of the multi-source uncertainties, the random white Gaussian noise is brought in to simulate the noise of measured signals polluted, and the interval vector is involved to characterize the uncertainties of material dispersion subsequently. In view of the issue of large interval parameters, the subinterval theory is established by a combination of the Taylor series expansion in each subinterval. Eventually, the validity and feasibility of the developed methodology are demonstrated by two numerical examples.

New hybrid reliability-based topology optimization method combining fuzzy and probabilistic models for handling epistemic and aleatory uncertainties

This study presents a hybrid reliability-based topology optimization (RBTO) method for handling epistemic and aleatory uncertainties. First, we establish a new triple-nested RBTO model based on fuzzy and probabilistic theory for describing the multi-source uncertainties. Subsequently, an efficient single-loop optimization method is proposed to degrade the triple-nested optimization problem into a deterministic optimization problem using the Karush–Kuhn–Tucker optimality condition. Furthermore, the sensitivities of the hybrid reliability constraint with respect to the random probabilistic variables, fuzzy variables, and deterministic design variables are derived using the adjoint variable method. Finally, a cantilever beam example, an L-shape beam design and a 3D example are tested to verify the validity of the proposed single-loop method.

A novel method of distributed dynamic load identification for aircraft structure considering multi-source uncertainties

A series of work for distributed dynamic load identification is investigated in this paper considering unknown-but-bounded uncertainties in the aircraft structure. To facilitate the analysis, the complicated rudder structure is simplified to a plate structure based on the robust equivalence principle of mechanical property under multi-cases of flight environments. Aiming at the plate structure, a time domain–based model for distributed dynamic load identification is established through the acceleration response measured by sensors. Among them, the spatial distributed load is approximated by Chebyshev orthogonal polynomials at each sampling time, and load boundaries can be calculated by the Taylor-expansion-based uncertain propagation analysis. As keys to improve the reliability of recognition results, the optimization process for sensor placement is constructed by the particle swarm optimization algorithm, taking the robustness evaluation index and sensor distribution index into consideration. The validity and the feasibility of the proposed methodology are demonstrated by several numerical examples, and the results reveal that designer can make a rational tradeoff choice among the cost of sensor placement and the performance of load identification in a systematic framework.