Presence Of Missing Measurements Research Articles

An adaptive IMM filter for jump Markov systems with inaccurate noise covariances in the presence of missing measurements

This paper derives a Kullback-Leibler average (KLA) based adaptive interacting multiple model (IMM) filter for jump Markov systems with inaccurate noise covariances in the presence of missing measurements. Firstly, a switching error model (SEM) is constructed to model the probability density function of measurement likelihood. The SEM can automatically choose appropriate error model for the real measurement or pure measurement noise by estimating a binary indicator. Secondly, by assigning conjugate priors on inaccurate noise covariances and binary indicator, the state, noise covariances and binary indicator are jointly estimated based on variational Bayesian (VB) technique. Finally, by using the KLA fusion scheme to fuse the conditioned estimates from every mode in the mixing and output phase, an adaptive IMM filter is derived. Simulation results demonstrate that the proposed filter achieves better performance than existing typical algorithms.

Tensor-structured decomposition improves systems serology analysis.

Systems serology provides a broad view of humoral immunity by profiling both the antigen‐binding and Fc properties of antibodies. These studies contain structured biophysical profiling across disease‐relevant antigen targets, alongside additional measurements made for single antigens or in an antigen‐generic manner. Identifying patterns in these measurements helps guide vaccine and therapeutic antibody development, improve our understanding of diseases, and discover conserved regulatory mechanisms. Here, we report that coupled matrix–tensor factorization (CMTF) can reduce these data into consistent patterns by recognizing the intrinsic structure of these data. We use measurements from two previous studies of HIV‐ and SARS‐CoV‐2‐infected subjects as examples. CMTF outperforms standard methods like principal components analysis in the extent of data reduction while maintaining equivalent prediction of immune functional responses and disease status. Under CMTF, model interpretation improves through effective data reduction, separation of the Fc and antigen‐binding effects, and recognition of consistent patterns across individual measurements. Data reduction also helps make prediction models more replicable. Therefore, we propose that CMTF is an effective general strategy for data exploration in systems serology.

Robust fault detection approach for wind farms considering missing data tolerance and recovery

The advancement in sensing technologies and infrastructure allows real-time condition monitoring on wind turbines (WTs), which helps improve the power generation efficiency, lower the maintenance costs of wind farms (WFs). Practically, the real-time measurements could be unavailable at the Supervisory Control and Data Acquisition end due to unintended events such as sensor faults and communication loss, which significantly depreciates the condition monitoring and fault detection performance. Aiming to mitigate the missing data impact on data-driven WF applications, this study develops a robust anomaly detection approach for WT fault detection using a denoising variational autoencoder. In presence of missing measurements, the proposed approach can not only sustain high fault detection performance but also recover the missing data as an auxiliary function. The proposed approach is tested on a realistic offshore WF and compared with other autoencoder variants and traditional anomaly detection methods. The testing results verify the outstanding robustness of the proposed approach against missing data events and demonstrate its great potential in missing data recovery.

Correlation Clustering Imputation for Diagnosing Attacks and Faults With Missing Power Grid Data

While the quality of the synchronized measurements is of paramount importance for real-time monitoring and protection of the power grids, collected measurements often contain missing values. This paper proposes a scheme for diagnosing attacks and faults in the presence of missing measurements in power grid data. The proposed scheme contains four modules for clustering, missing data imputation, decision-making, and optimization. This paper develops a novel technique for missing data imputation based on the correlation-connected clusters that consider local correlation among the measurements in estimating missing data, handle high-dimensional data, and tolerate high missing ratios. The optimization module ties the imputation process to diagnostic performance. The proposed novel imputation technique is compared with other state-of-the-art techniques within the diagnostic scheme. The achieved results show that the proposed technique significantly outperforms other competitors.

Event‐triggered state estimation for stochastic hybrid systems with missing measurements

This study is concerned with the event-triggered state estimation problem for a class of stochastic hybrid systems with missing measurements in a networked environment. Two independent Markov chains are introduced to, respectively, characterise the stochastic measurement missing and the possible modal (or mode) transition of the system. In consideration of the constrained bandwidth and limited power resources of networked systems, a closed-loop event-triggered mechanism based on the measurement innovation is designed to trigger data transmission only when trigger conditions are satisfied. To keep the exponentially increasing number of full hypothesis sequences in optimal estimation to bounded computational complexity, the interacting multiple model framework is extended to tackle event-triggered sampling with the statistical information implicit in event-triggered conditions sufficiently explored and the possible measurement missing taken into account. A Monte Carlo simulation involving tracking a two-dimensional manoeuvring target with two operational modes is provided to demonstrate the effectiveness and efficiency of the proposed event-triggered hybrid state estimation in the presence of missing measurements.

LPV model identification with an unknown scheduling variable in the presence of missing observations – a robust global approach

This study focuses on identifying the linear parameter varying (LPV) system with an unknown scheduling variable in the presence of missing measurements and the system output data contaminated with outliers. The parameter interpolated LPV autoregressive exogenous (ARX) model with an unknown scheduling variable is considered and the scheduling variable dynamic is described by a non-linear state-space model. The outliers treatment and unknown scheduling variable estimation with missing observations are both taken into consideration. The robust LPV model is established based on the Student's t -distribution in order to handle the outliers and the particle smoother is adopted to estimate the true scheduling variable from incomplete data set. The formulations of the proposed algorithm are finally derived in the expectation-maximisation algorithm scheme and the formulas to estimate the unknown parameters of LPV ARX model and scheduling variable dynamic model are derived simultaneously. A numerical example and a chemical process are used to present the efficacy of the proposed approach.

Surrogate‐splits ensembles for real‐time voltage stability assessment in the presence of missing synchrophasor measurements

This paper proposes a new machine learning approach in the presence of missing measurements using surrogate-splits ensembles for real-time voltage stability assessment comprising of the classification of the operating state of the power system, and the prediction of the power system's margin to voltage collapse. The proposed approach applied the Boosting and Bagging machine learning methods in the design of the ensemble models required for the classification and regression tasks, respectively, based on the feature attributes obtained from synchrophasor measurements. The performance of the trained classifiers and regressors was evaluated for the case of missing phasor measurement unit (PMU) measurements. Robust classifier and regressor models immune to missing PMU measurements were afterwards developed using the surrogate-splits technique. The validity of the proposed approach was tested using the New England 39-bus test system. The experimental results obtained validate the effectiveness of the proposed method for various operating scenarios and contingencies even in the presence of missing PMU measurements.

Asynchronous H∞ control for unmanned aerial vehicles: switched polytopic system approach

This study is concerned with the H∞ control for the full-envelope unmanned aerial vehicles (UAVs) in the presence of missing measurements and external disturbances. With the dramatic parameter variations in large flight envelope and the locally overlapped switching laws in flight, the system dynamics is modeled as a locally overlapped switched polytopic system to reduce designing conservatism and solving complexity. Then, considering updating lags of controller's switching signals and the weighted coefficients of the polytopic subsystems induced by missing measurements, an asynchronous H∞ control method is proposed such that the system is stable and a desired disturbance attenuation level is satisfied. Furthermore, the sufficient existing conditions of the desired switched parameter-dependent H∞ controller are derived in the form of linear matrix inequality (LMIs) by combining the switched parameter-dependent Lyapunov function method and average dwell time method. Finally, a numerical example based on a highly maneuverable technology (HiMAT) vehicle is given to verify the validity of the proposed method.

Robust Sampled-DataH∞Control for Vibration Mitigation of Offshore Platforms with Missing Measurements

This paper presents a robust sampled-dataH∞control for vibration attenuation of offshore platforms with missing measurements subject to external wave force. It is well known that the phenomena of missing measurements are unavoidable due to various reasons, such as sensor aging and sensor temporal failure, which may degrade the control system performance or even cause instability. For active vibration control of offshore platforms with sampling measurements, to deal with the missing measurements, a robust sampled-dataH∞control method is proposed in this paper. The robust sampled-dataH∞state feedback controller is obtained in terms of the solvability of certain linear matrix inequalities (LMIs). Finally, simulations on an offshore platform are exploited to demonstrate the effectiveness of the proposed method. The simulation results show that the designed robust sampled-dataH∞control scheme is effective to attenuate the external wave force in the presence of missing measurements.

Structural Approach To Design Sensor Networks for Fault Diagnosis

Key faults significantly affect the normal operation of the process originating risk conditions. These failures should be identified even in the presence of missing measurements or outliers. In this work a new strategy to design sensor networks, which are able to resolve a set of key faults when sensors fail, is presented. The procedure deals with failure isolation using the Fault Resolution Degree concept. This is incorporated as a constraint of the minimum-cost design formulation, and the resulting optimization problem is solved using MILP codes. The strategy only uses low uncertainty data that are readily available at the process design stage. Application results of the methodology to case studies extracted from the literature are presented and compared with those provided by other existing techniques.

The impact of missing measurements on PCA and PLS prediction and monitoring applications

A significant issue in the practical application of PCA and PLS models for inferential sensors and process monitoring is the presence of sets of measurements (objects) that are incomplete. Since the missing measurements in the incomplete objects are usually correlated with some of the available measurements, an opportunity exists to use these objects if efficient algorithms and tools exist and if their performance and limitations are defined. This paper analyses the uncertainties in the predictions, latent variables, Hotelling T 2 and residual squared prediction error (SPE) that arise from the missing measurements. Intervals for these values are developed which give an indication of the degree of uncertainty introduced by the missing measurements. These intervals can be used to assess whether or not the inferential models or the process monitoring scheme perform well in the presence of missing measurements. They can also be used to determine which measurements to recover to get the greatest uncertainty reduction. The results are illustrated by application to industrial data from a Kamyr digester.