Search

In situ and accurate oxygen concentration detection in pharmaceutical glass vials is of great significance in the development of the pharmaceutical industry. This article reveals a universal data distribution rule in which each sampling point follows an approximately normal distribution with its own mean ($\mu $) and standard deviation ($\sigma $) in a wavelength modulation-based tunable diode laser absorption spectroscopy (TDLAS/WMS) system. Based on this investigation, we propose a novel concept, harmonic amplitude pass probability (HAPP), where noisy data points in the demodulated second harmonic signal can be cleverly transformed into informative probability labels. That is, the awkward interference components in the harmonic data are implicitly washed out. Innovatively, the HAPP acts as a smart bridge between the raw harmonic data and Bayesian network. The HAPP-based Bayesian classifier achieves an average detection accuracy of 97.3% when fed with actual data captured from a penicillin encapsulation line, which presents a data-mining methodology for online package leak detection from a probability statistics perspective. [ABSTRACT FROM AUTHOR]

Cumulative damage processes for discrete degrading systems caused by stochastic shocks have received increasing attention. However, the degradation data is often subject to incomplete observation or censoring due to imperfect monitoring instruments or inspection strategies. In such situations, the arrival time of stochastic shocks cannot be accurately recorded. This makes a challenging task for parameter estimation and remaining useful lifetime (RUL) prognosis based on the cumulative damage process. Motivated by this fact, this article proposed a new Bayesian data augmentation approach to overcome the obstacles caused by incomplete observation. First, the proposal distribution for the occurrence number of shocks in the cumulative damage process subject to interval censoring is derived. Second, an effective data augmentation for the missing shock time is proposed by integrating rejection sampling and the order statistics technique. On this basis, multiple imputation is extended to discrete non-Gaussian stochastic processes subject to interval censoring. Finally, an RUL prognosis method of a system subject to the cumulative damage process is derived based on the Monte Carlo method. And a real case study on the head wear problem is applied to illustrate the superiority of the proposed approach, and the results show that the proposed approach can improve the estimation accuracy of RUL compared with the existing method. [ABSTRACT FROM AUTHOR]

To scientifically evaluate the damage effectiveness of a projectile explosion on the equivalent target, this article studies a warhead fragment distribution parameters’ measurement method based on a six-screen-sensors’ intersection test system; establishes a parameter calculation model of the warhead fragment hitting the equivalent target, the time and space constraints’ discriminant model of six infrared detection screens, and the time-value-matching model of multiple warhead fragments; and gives the calculation of the specific fragment distribution position and the time-matching algorithm of each infrared detection screen. According to the obtained warhead fragment parameters, a calculation method of the damage probability of equivalent target is constructed. Through experiments and analysis, it is verified that the established warhead fragment parameter measurement method and the damage probability model of the equivalent target are scientific. [ABSTRACT FROM AUTHOR]

The development of artificial intelligence has brought new opportunities and challenges in the field of mechanical fault diagnosis. Especially, data-driven intelligent fault diagnosis (DIFD) methods are favored by researchers. However, commonly used DIFD methods can only recognize the fault of a known class instead of an unknown class, which results in the DIFD’s inability to self-learning. To handle this problem, novel probability confidence convolutional neural network (PCCNN) is proposed in this article. The probability of belonging to each known class calculated by the PCCNN and the confidence of each known class is used to recognize the known and unknown classes. At the same time, the ability to recognize a new class is achieved by updating the architecture and parameters of the PCCNN model with a self-learning method. The bearing experiment dataset, the gearbox experiment dataset, and the rotating machinery failure data are used in the case studies. The study results show that the proposed method achieves an average accuracy of 97.42% and 96.87% for recognizing unknown and known classes, respectively, exhibiting at least 5.18% better performance than the 12 comparison methods. All the code and experimental data are available on the website. [ABSTRACT FROM AUTHOR]

A general probabilistic model of the measurement process is discussed, showing how it can address both static and dynamic measurements. Emphasis is placed on the latter, and two main kinds of dynamic measurements are considered: direct, where the goal is to monitor a phenomenon over time, and indirect, where an alternative representation is sought, such as, typically, the spectrum. It is shown that both cases can be treated in the same framework, through a proper interpretation of the variables involved. Then, direct measurements are investigated in greater detail, and the application of this approach is discussed, in the case of a Markovian model with a uniform prior for the measurand, and an illustrative numerical example is presented. These results may pave the way to a better harmonization of measurements, grounded on a sound scientific foundation. [ABSTRACT FROM AUTHOR]

A deep revision of the guide to the expression of uncertainty in measurement (GUM) is proposed, that will maintain the same original goals and principles but differ from the current version by providing greater support for users to implement those principles. This new approach may overcome some of the setbacks that have been encountered over the years in the application of the GUM, thus making the practice of evaluating measurement uncertainty more widespread. [ABSTRACT FROM AUTHOR]

Load disaggregation techniques infer the operation of different power-consuming devices from a single measurement point that records the total power drawn over time. Thus, a device consuming power at the moment can be understood as information encoded in the power draw. However, similar power draws or similar combinations of power draws limit the ability to detect the currently active device set. We present an information coding perspective of load disaggregation to enable a better understanding of this process and support its future improvement. In typical cases of quantity and types of devices and their respective power consumption, not all possible device configurations can be mapped to distinguishable power values. We introduce the term proficiency to describe the suitability of a device set for load disaggregation. We provide the notion and calculation of entropy of the initial device states, mutual information of power values, and the resulting uncertainty coefficient or proficiency. We show that the proficiency is highly dependent on the device running probability, especially for devices with multiple states of power consumption. The application of the concept is demonstrated by artificial data as well as with actual power consumption data from real-world power draw data sets. [ABSTRACT FROM AUTHOR]

This paper proposes Kalman-filter drift removal (DR) and Heron-bilateration location estimation (LE) to significantly reduce the received signal strength index (RSSI) drift, localization error, computational complexity, and deployment cost of conventional radio frequency identification (RFID) indoor positioning systems without any sacrifice of localization granularity and accuracy. By means of only one portable RFID reader as the targeted device and only one pair of active RFID tags as the border-deployed landmarks, this paper develops a real-time portable RFID indoor positioning device and cost-effective scalable RFID indoor positioning infrastructure, based on Kalman-filter DR, Heron-bilateration LE, and four novel preprocessing/postprocessing techniques. Experimental results reveal that the proposed Kalman-filter DR method is faster and better to converge the distance measurement (DM) error than conventional probability/statistics in terms of various relative distances under certain RSSI drift effect condition, and the proposed Heron-bilateration LE method is also faster and better to converge the LE error than conventional proximity pattern matching and trilateration in terms of three or more landmarks under certain DM error condition. On the other hand, a portable RFID indoor positioning device is smoothly implemented on an Android smartphone platform attached with a portable Bluetooth-based RFID reader. [ABSTRACT FROM PUBLISHER]

The capability to locate and characterize precursor wiring faults, such as chafing or pinching potentially enables preventive maintenance well before hard failures occur, thus maximizing system functionality and safety while minimizing out-of-service time. Toward this goal, results are presented on the application of a deterministic Bayesian inference procedure well suited for detecting chafing and pinch faults through the use of a newly developed physics-based model for shielded-twisted-pair cable. This method is significantly faster than more traditional nondeterministic Bayesian methods, such as Markov chain Monte Carlo, and retains many of the desirable features inherent to the Bayesian approach. These include the ability to quantify estimation uncertainty and model evidence in probabilistic terms, which then enables the study and design of noise-tolerant fault detection algorithms capable of classifying different types of faults. The fault parameter estimation results from both laboratory and field measurements on a C17 jet engine are shown to demonstrate the achievable model fidelity and the overall viability of the approach. [ABSTRACT FROM PUBLISHER]

Exploiting spatial diversity by means of cooperative readpoints is a popular approach to the missing tag problem in radio frequency identification (RFID) systems. However, there is little theory behind this approach, and RFID system design is often limited to trial-and-error and extensive experiments. In this paper, we present a system model for cooperative readpoints based on a correlated binomial distribution. The model can be used to determine the required number of readpoints for a given requirement and provides an online estimate of the system performance at runtime. The analysis of a real-world dataset justifies our modeling assumptions and shows that the diversity concept is able to significantly enhance system performance. [ABSTRACT FROM PUBLISHER]

The GUM and its supplements one and two have come to be regarded as de facto standards for evaluating measurement uncertainty. The latter documents have stressed the benefits of deriving probability density functions (PDFs) for the output quantities instead of just evaluating the standard and expanded uncertainties associated with their best estimates. Those supplements, however, present a brute force numerical method for obtaining the output PDFs. In this paper, we rely instead on the use of the change-of-variables theorem as a way of obtaining analytical expressions for PDFs in the form of integrals that, at least in principle, can be numerically evaluated. But more importantly, the GUM supplements do not include situations in which the available information is more than enough, in the sense that by separately considering its various components one may arrive at concurrent PDFs for the quantity or quantities of interest. In some situations these PDFs will be largely similar, so that implementing some form of merging method would perhaps be reasonable. But it may also happen that significant discrepancies between those PDFs are apparent, in which case a modification of the measurement model might make all information appear to be consistent. Herein we review two procedures for merging concurrent and not too dissimilar PDFs. We also discuss some elements to be considered if extending the measurement model appears to be a better alternative. [ABSTRACT FROM PUBLISHER]

This paper presents an algorithmic method for measurements of relaxed statistical properties of point processes (PPs). In particular, this method makes it possible to determine the probability density function of the residual waiting time, which is the distance between an arbitrary origin and the first point of the PP posterior to this origin and also the probabilities of counting n points of the PP in an interval open by an instant independent of the PP. These quantities are rarely accessible from a theoretical calculation. However, in the cases where this is possible, various experiments show quite good agreement with the theory. [ABSTRACT FROM PUBLISHER]

Measurement plays a key role in science and technology in which it is a privileged way of gathering information from the real world and of validating scientific models. Is it possible to achieve a unified measurement theory for the different disciplines, ranging from physics to psychology, in which it is involved? A basic kernel for such a theory is surveyed and discussed. A probabilistic approach is followed in order to consider uncertainty as an inherent feature. [ABSTRACT FROM AUTHOR]

The concept of uncertainty in measurement stems from that of (probable) error and is intimately intertwined with it. Both concepts can be viewed as measures of the quality of a measurement or, better, of the corresponding estimate. There is an endless list of misunderstandings, false beliefs, and misinterpretations on this subject. People tend also to use the same word with different meanings. In this paper, while giving an overview of the evolution of the concept, from the initial unawareness of the need for a quality assessment (still present in many areas) to the present views, the author also tries to shed some light and some clarity on the most popular and debated misunderstandings, particularly about the concepts of error, true quantity value(s), and measured quantity value. The author is the convener of the Joint Committee for Guides in Metrology Working Group 1 (Guide to the expression of uncertainty in measurement). The opinion expressed in this paper does not necessarily represent the view of this working group. [ABSTRACT FROM AUTHOR]

This paper considers the well-known concept of uncertainty of a measurement result and discusses the mathematical approach adopted for its evaluation and expression by the present standard documents and, in particular, the Guide to the Expression of Uncertainty in Measurement. The practical limitations introduced by the assumptions of this mathematical approach are discussed, and the requirements of a new mathematical approach capable of overcoming the present limitations are drawn. A different mathematical approach, based on the theory of evidence and originally developed by the authors in recent years, is presented and discussed, with some new findings and observations. [ABSTRACT FROM AUTHOR]

Three methods for the probability distribution function F(x) estimation, based on the traditional Monte Carlo method, the Chebyshev inequality, and the theorems of Smimov and Kolmogorov, are developed and compared. The methods are based on the functional dependence of the sample size n on the estimated value of F(x) with an accuracy factor as a parameter. Expressions for this dependence are derived for all methods and compared. It is shown that the accuracy of the estimation can be expressed and controlled by two parameters: the confidence and the confidence limits factor. Based on the methods developed, two techniques for F(x) estimation are proposed and demonstrated: a technique with prespecified accuracy and a technique with controlled accuracy. The technique with controlled accuracy allows the processing of a random sample starting with a limited initial sample of observations and then proceeding with the processing on an observation-by-observation basis. This way, the time that is needed to sort the sample has been minimized, and the estimation of distribution functions in real-time has become possible. This technique is demonstrated by estimating the distribution functions of chaotic and Gaussian sequences that are applied in codedivision multiple-access (CDMA) systems. [ABSTRACT FROM AUTHOR]

The need for information security is self-evident. The pervasiveness of this critical topic requires primarily risk assessment and management through quantitative means. To do an assessment, repeated security probes, surveys, and input data measurements must be taken and verified toward the goal of risk mitigation. One can evaluate risk using a probabilistically accurate statistical estimation scheme in a quantitative security meter (SM) model that mimics the events of the breach of security. An empirical study is presented and verified by discrete-event and Monte Carlo simulations. The design improves as more data are collected and updated. Practical aspects of the SM are presented with a real- world example and a risk-management scenario. [ABSTRACT FROM AUTHOR]

The voltage waveform in an electric power system usually presents a fluctuation due to the intermittent operation of nonlinear devices. In some cases, the frequency and width of these voltage variations can produce a physiologically irritating phenomenon due to luminance fluctuations of the lighting (flicker effect). In addition to the visual inconvenience, some of the electrical equipment can be strongly disturbed. The International Electrotechnical Commission standardized the implementation of a flickermeter based on the simulation of the process of physiological visual perception to supply information about the human reaction to voltage amplitude modulation. A standard parameter (i.e., the flicker severity index Pst) is obtained from the cumulated probability function of the instantaneous perception. In this paper, a different measurement approach for the flicker effect evaluation is proposed. It is based on the calculation of the luminous variation perception function by processing digital samples of the acquired voltage. The choice of this technique is motivated by the aim of simplifying the flickermeter implementation and reducing the hardware requirements. A prototype of the instrument has been implemented, and the performance has been evaluated, by means of direct comparison with standard flickermeters. In this paper, the main results are also reported. [ABSTRACT FROM PUBLISHER]

The generalized lambda distribution family fits the probability distributions of a wide variety of data sets, including the most important distributions encountered in the measurement applications (normal, uniform, Student's t, U-shaped, exponential). This paper illustrates how the four parameters needed for such distribution can be exploited in the expression of measurement uncertainty and to extend the information related to a measurement. The obtained representation allows an immediate calculation of coverage intervals and is particularly useful to support the techniques commonly applied in the estimation of the combined uncertainty. Moreover, in order to include the classical measurement information, a novel parameterization of the distribution is proposed. [ABSTRACT FROM AUTHOR]

The 12 papers i nthis special section focus on measurement techniques and fundamental applications. [ABSTRACT FROM AUTHOR]