Polymeric materials are widely used in photovoltaic (PV) systems, making it essential to understand their service life to ensure reliable PV performance. The primary failure mechanism of polymeric materials in PV systems is photodegradation caused by ultraviolet (UV) radiation. Degradation modeling provides a framework for predicting service life, with a key step being the development of predictive models for degradation paths. This paper presents statistical and machine learning approaches for predicting the outdoor degradation of polymeric components in PV systems. We describe the study design and data collection process for developing predictive models based on indoor laboratory testing data, which are then extended to outdoor field conditions with time-varying environmental variables, with prediction uncertainty quantified through simulation. Deep learning (DL) methods are also explored, and results are compared across modeling approaches. The parametric statistical model demonstrates good fit and predictive performance across datasets and shows greater robustness by incorporating physical and chemical knowledge. The DL model provides flexibility in capturing complex covariate relationships and often yields accurate predictions, though it is less robust across datasets. The paper concludes with remarks on key findings and their implications for PV reliability.
Simulation-based calibration aims to infer unknown parameters of complex simulation models by aligning model outputs with real-world observations. When simulation runs are computationally expensive, statistical emulators trained on simulation data are used to efficiently approximate the model. An intelligent, adaptive selection of simulation inputs for building the emulator can substantially improve the efficiency of the calibration process. This task is particularly challenging for stochastic simulations with noisy outputs, since both selecting new input locations (exploration) and allocating repeated runs at existing inputs (replication) are essential for efficiently learning the input-output relationship. In this paper, we introduce an active learning framework that adaptively balances exploration and replication for data-efficient calibration. Our uncertainty-aware acquisition criterion targets learning the posterior density of the unknown simulation parameters, and we derive two corresponding forms of the acquisition function for exploration and replication. Building on these, we propose a strategy that, at each stage of the sequential design, chooses between exploration and replication to most effectively reduce the uncertainty in the estimate of the posterior density of the simulation parameters. Experiments on synthetic benchmarks and a real epidemiological model demonstrate that our approach significantly improves learning of the posterior distribution of the simulation parameters while reducing the number of required simulations, making it well-suited for expensive stochastic simulation settings.
Except a few, the majority of the literature on monitoring ordinal data consider independent and identically distributed processes, where samples of data are collected sequentially in time. However, stationary ordinal processes exhibiting serial dependence are also common in many real-world process monitoring applications. This study proposes three classes of novel control charts for monitoring serially dependent stationary ordinal processes. Instead of sample statistics, individual observations are utilized. Exponentially weighted moving-average smoothing of a sequence of estimates is used for estimating the probability mass (cumulative distribution) function of the process. Defined real-valued functions of the probability mass (cumulative distribution) estimates are then used as the statistic plotted on the control charts. The methods are designed to be sensitive to a shift in the marginal distribution. Average run length performance of the control charts are computed under a comprehensive set of data-generating process models, which are inspired by real-world examples and exhibit quite different serial dependence structures. The performances of the proposed control charts are evaluated and compared to provide recommendations for implementations. The results show that the class of demerit-type charts generally perform better than the others. To illustrate the application and interpretation of the proposed methods, a real-world data example on monitoring of heating, ventilation, and air conditioning systems in passenger rail coaches is discussed.
For the case of sensor health monitoring, keeping the correlation level of two sensors measuring related quantities under surveillance is a promising idea. The corresponding raw data streams behave often unsteady. But having a stable correlation level is a typical in-control pattern, whereas mean and variance may vary from time window to time window. However, there are nearly none control charts for monitoring correlation. Here, we consider EWMA control charts for the linear correlation coefficient & rhov;. Despite it is known for a long time, the usage of the explicit (assuming normally distributed data) distribution of the estimator of & rhov; while setting up a control chart seems to be non-existent. Here, we build an EWMA chart utilizing this estimator, namely the Pearson correlation, and calculate the most popular performance measure, the zero-state average run length (ARL), by means of various numerical methods. Less surprisingly, the two standard methods work poorly for certain chart designs. We solve these problems by utilizing piece-wise collocation. Moreover, we examine further configuration details and provide some guidelines. Two applications illustrate the usefulness of monitoring the & rhov; level.
