A method of sensor fault detection and identification

A method of Bayesian belief network (BBN)-based sensor fault detection and identification is presented. It is applicable to processes operating in transient or at steady-state. A single-sensor BBN model with adaptable nodes is used to handle cases in which process is in transient. The single-sensor BBN model is used as a building block to develop a multi-stage BBN model for all sensors in the process under consideration. In the context of BBN, conditional probability data represents correlation between process measurable variables. For a multi-stage BBN model, the conditional probability data should be available at each time instant during transient periods. This requires generating and processing a massive data bank that reduces computational efficiency. This paper presents a method that reduces the size of the required conditional probability data to one set. The method improves the computational efficiency without sacrificing detection and identification effectiveness. It is applicable to model- and data-driven techniques of generating conditional probability data. Therefore, there is no limitation on the source of process information. Through real-time operation and simulation of two processes, the application and performance of the proposed BBN method are shown. Detection and identification of different sensor fault types (bias, drift and noise) are presented. For one process, a first-principles model is used to generate the conditional probability data, while for the other, real-time process data (measurements) are used.     

Uncertainty-aware soft sensor using Bayesian recurrent neural networks

Data-driven soft sensors have been widely used to measure key variables for industrial processes. Soft sensors using deep learning models have attracted considerable attention and shown superior predictive performance. However, if a soft sensor encounters an unexpected situation in inferring data or if noisy input data is used, the estimated value derived by a standard soft sensor using deep learning may at best be untrustworthy. This problem can be mitigated by expressing a degree of uncertainty about the trustworthiness of the estimated value produced by the soft sensor. To address this issue of uncertainty, we propose using an uncertainty-aware soft sensor that uses Bayesian recurrent neural networks (RNNs). The proposed soft sensor uses a RNN model as a backbone and is then trained using Bayesian techniques. The experimental results demonstrated that such an uncertainty-aware soft sensor increases the reliability of predictive uncertainty. In comparisons with a standard soft sensor, it shows a capability to use uncertainties for interval prediction without compromising predictive performance.     

基于自组织免疫网络的传感器故障检测模型

提出了一种基于自组织免疫网络的传感器故障检测模型。该模型将自组织学习的思想引入到传感器免疫网络的建模中,通过学习向量量化确定免疫网络的连接权值,并对其结构和特点进行了分析,给出了相应的诊断算法。仿真结果表明,所提出的方法对故障传感器具有较高的检测灵敏度,并且对噪声具有一定的容忍能力,对于航空发动机传感器的监测具有一定的应用价值,并可方便地推广到其他类似的工业应用领域。     

An empirical comparison of combinations of evolutionary algorithms and neural networks for classification problems.

There are numerous combinations of neural networks (NNs) and evolutionary algorithms (EAs) used in classification problems. EAs have been used to train the networks, design their architecture, and select feature subsets. However, most of these combinations have been tested on only a few data sets and many comparisons are done inappropriately measuring the performance on training data or without using proper statistical tests to support the conclusions. This paper presents an empirical evaluation of eight combinations of EAs and NNs on 15 public-domain and artificial data sets. Our objective is to identify the methods that consistently produce accurate classifiers that generalize well. In most cases, the combinations of EAs and NNs perform equally well on the data sets we tried and were not more accurate than hand-designed neural networks trained with simple backpropagation.     

Sensor-centric energy-constrained reliable query routing for wireless sensor networks

Standard wireless sensor network models emphasize energy efficiency and distributed decision-making by considering untethered and unattended sensors. To this we add two constraints—the possibility of sensor failure and the fact that each sensor must tradeoff its own resource consumption with overall network objectives. In this paper, we develop an analytical model of energy-constrained, reliable, data-centric information routing in sensor networks under all the above constraints. Unlike existing techniques, we use game theory to model intelligent sensors thereby making our approach sensor-centric. Sensors behave as rational players in an N-player routing game, where they tradeoff individual communication and other costs with network wide benefits. The outcome of the sensor behavior is a sequence of communication link establishments, resulting in routing paths from reporting to querying sensors. We show that the optimal routing architecture is the Nash equilibrium of the N-player routing game and that computing the optimal paths (which maximizes payoffs of the individual sensors) is NP-Hard with and without data-aggregation. We develop a game-theoretic metric called path weakness to measure the qualitative performance of different routing mechanisms. This sensor-centric concept which is based on the contribution of individual sensors to the overall routing objective is used to define the quality of routing (QoR) paths. Analytical results on computing paths of bounded weakness are derived and game-theoretic heuristics for finding approximately optimal paths are presented. Simulation results are used to compare the QoR of different routing paths derived using various energy-constrained routing algorithms.     

Standalone noise and anomaly detection in wireless sensor networks: A novel time-series and adaptive Bayesian-network-based approach

Wireless sensor networks (WSNs) consist of small sensors with limited computational and communication capabilities. Reading data in WSN is not always reliable due to open environmental factors such as noise, weakly received signal strength, and intrusion attacks. The process of detecting highly noisy data is called anomaly or outlier detection. The challenging aspect of noise detection in WSN is related to the limited computational and communication capabilities of sensors. The purpose of this research is to design a local time-series-based data noise and anomaly detection approach for WSN. The proposed local outlier detection algorithm (LODA) is a decentralized noise detection algorithm that runs on each sensor node individually with three important features: reduction mechanism that eliminates the noneffective features, determination of the memory size of data histogram to accomplish the effective available memory, and classification for predicting noisy data. An adaptive Bayesian network is used as the classification algorithm for prediction and identification of outliers in each sensor node locally. Results of our approach are compared with four well-known algorithms using benchmark real-life datasets, which demonstrate that LODA can achieve higher (up to 89%) accuracy in the prediction of outliers in real sensory data.     

Scalable human-robot interactions in active sensor networks

Decentralized sensor networks promise virtually unlimited scalability and can tolerate individual component failures. An experimental active sensor network that leverages environment-centric modes of human-robot interaction can keep up with a network's arbitrary growth. Spatially distributed sensors provide better coverage, faster response to dynamically changing environments, better survivability, and robustness to failure. Taking an extra step to a decentralized system provides further benefits of scalability, modularity, and performance. Our active sensor network is a collection of sensing platforms connected into a network. Some or all of the network components have actuators that we can control, making them, in this sense, active. A mobile robot with onboard sensors and a communication facility is an example of an active component. We investigate the scalability of an important aspect of an ASN: interaction with human operations.     

Upper bound for variational free energy of Bayesian networks

In recent years, variational Bayesian learning has been used as an approximation of Bayesian learning. In spite of the computational tractability and good generalization in many applications, its statistical properties have yet to be clarified. In this paper, we focus on variational Bayesian learning of Bayesian networks which are widely used in information processing and uncertain artificial intelligence. We derive upper bounds for asymptotic variational free energy or stochastic complexities of bipartite Bayesian networks with discrete hidden variables. Our result theoretically supports the effectiveness of variational Bayesian learning as an approximation of Bayesian learning.     

Analyzing the techniques that improve fault tolerance of aggregation trees in sensor networks

Sensor networks are finding significant applications in large scale distributed systems. One of the basic operations in sensor networks is in-network aggregation. Among the various approaches to in-network aggregation, such as gossip and tree, including the hash-based techniques, the tree-based approaches have better performance and energy-saving characteristics. However, sensor networks are highly prone to failures. Numerous techniques suggested in the literature to counteract the effect of failures have not been carefully analyzed. In this paper, we focus on the performance of these tree-based aggregation techniques in the presence of failures. First, we identify a fault model that captures the important failure traits of the system. Then, we analyze the correctness of simple tree aggregation with our fault model. We then use the same fault model to analyze the techniques that utilize redundant trees to improve the variance. The impact of techniques for maintaining the correctness under faults, such as rebuilding or locally fixing the tree, is then studied under the same fault model. We also do the cost-benefit analysis of using the hash-based schemes which are based on FM sketches. We conclude that these fault tolerance techniques for tree aggregation do not necessarily result in substantial improvement in fault tolerance.     

基于标识的无线传感器网络密钥预分配方案

密钥分配是无线传感器网络中极具挑战性的安全问题之一.为了实现无线传感器网络中的安全通讯,需要对传感器结点间传递的信息进行加密.由于受每个传感器结点自身资源的限制,传统网络中使用的密钥分配策略,并不适用于无线传感器网络.提出了基于结点ID的密钥预分配方案,实现了非对称和对称密钥体制、分布式和集中式密钥管理的结合,有效地提高了传感器网络其安全性和连通性.     

Event-based motion control for mobile-sensor networks

In many sensor networks, considerably more units are available than necessary for simple coverage of the space. Augmenting sensor networks with motion can exploit this surplus to enhance sensing while also improving the network's lifetime and reliability. Sensor mobility allows better coverage in areas where events occur frequently. Another use of mobility comes about if the specific area of interest (within a larger area) is unknown during deployment. We've developed distributed algorithms for mobile-sensor networks to physically react to changes or events in their environment or in the network itself. Distribution supports scalability and robustness during sensing and communication failures. We present two classes of motion-control algorithms that let sensors converge on arbitrary event distributions. These algorithms trade off the amount of required computation and memory with the accuracy of the sensor positions. We also present three algorithms that let sensor networks maintain coverage of their environment. These algorithms work alongside either type of motion-control algorithm such that the sensors can follow the control law unless they must stop to ensure coverage.     

Influence of the optimization methods on neural state estimation quality of the drive system with elasticity

The paper deals with the implementation of optimized neural networks (NNs) for state variable estimation of the drive system with an elastic joint. The signals estimated by NNs are used in the control structure with a state-space controller and additional feedbacks from the shaft torque and the load speed. High estimation quality is very important for the correct operation of a closed-loop system. The precision of state variables estimation depends on the generalization properties of NNs. A short review of optimization methods of the NN is presented. Two techniques typical for regularization and pruning methods are described and tested in detail: the Bayesian regularization and the Optimal Brain Damage methods. Simulation results show good precision of both optimized neural estimators for a wide range of changes of the load speed and the load torque, not only for nominal but also changed parameters of the drive system. The simulation results are verified in a laboratory setup.     

A novel energy-efficient clustering protocol with area coverage awareness for wireless sensor networks

Coverage is a key metric in evaluating the monitoring capacity and quality of services in wireless sensor networks. The energy consumption of self-contained sensors is also a challenging problem for energy-efficient use while still achieving better coverage performance. Although techniques have been developed to mitigate the problem of area coverage, particularly together with efficient clustering methods, none focuses intensively on the sensor activation stage, which is used to maintain coverage while optimizing energy usage. In this research, we thus propose a cover set to find the minimum set of sensors that completely cover the sensing ranges within an interest area as a criterion for sensor activation. Our main goal is to select an optimal number of active sensors considering residual energy and the cover set and to keep alive the important sensors for the sensing coverage task as long as possible. Additionally, this research proposes an area coverage-aware clustering protocol (ACACP) with energy consumption optimization with respect to the activation sensor, network clustering, and multi-hop communication to improve overall network lifetime while preserving coverage. Throughout the intensive simulation, given a diversity of deployments with scalability concern, the results demonstrate the effectiveness of ACACP when compared with other competitive approaches such as ECDC and DECAR, including state-of-the-art clustering protocols such as LEACH, in terms of coverage ratio and overall network lifetime.     

Handling large-scale node failures in mobile sensor/robot networks

In Wireless Sensor Networks (WSNs), maintaining connectivity with the sink node is a crucial issue to collect data from sensors without any interruption. While sensors are typically deployed in abundance to tolerate possible node failures, a large number of simultaneous node failures within the same region may result in partitioning the network which may disrupt the network operation significantly. Given that WSNs are deployed in inhospitable environments, such node failures are very likely due to storms, fires, floods, etc. The self-recovery of the network from these large-scale node failures is challenging since the nodes will not have any information about the location and span of the damage. In this paper, we first present a distributed partition detection algorithm which quickly makes the sensors aware of the partitioning in the network. This process is led by the sensors whose upstream nodes fail due to damages. Upon partition detection, sensors federate the partitions and restore data communication by utilizing the former routing information stored at each sensor to the sink node and exploiting sensor mobility. Specifically, the locations of failed sensors on former routes are used to assess the span of the damage and some of the sensors are relocated to such locations to re-establish the routes with the sink node. Relocation on such former routes is performed in such a way that the movement overhead on sensors is also minimized. Our proposed approach solely depends on the local information to ensure autonomicity, timeliness and scalability. The effectiveness of the proposed federation approach is validated through realistic simulation experiments and has been shown to provide the mentioned features.     

Robust hierarchical clustering for novelty identification in sensor networks: With applications to industrial systems

The paper proposes a new, robust cluster-based classification technique for Novelty Identification in sensor networks that possess a high degree of correlation among data streams. During normal operation, a uniform cluster across objects (sensors) is generated that indicates the absence of novelties. Conversely, in presence of novelty, the associated sensor is clustered distinctly from the remaining sensors, thereby isolating the data stream which exhibits the novelty. It is shown how small perturbations (stemming from noise, for instance) can affect the performance of traditional clustering methods, and that the proposed variant exhibits a robustness to such influences. Moreover, the proposed method is compared with a recently reported technique, and shown that it performs 365% faster computationally. To provide an application case study, the technique is used to identify emerging fault modes in a sensor network on a sub-15MW industrial gas turbine in presence of other abrupt, but normal changes that visually might otherwise be interpreted as malfunctions.     

无线传感器网络中移动传感器的再定位技术

回顾了在传感器网络中引入移动传感器的过程。介绍了移动传感器再定位技术可解决传感器网络中的事件深入感知、传感器失效和非精确投放等问题。详细阐述了现有的移动传感器移动至感兴趣区域深入感知、扩大网络覆盖面积和修补网络覆盖洞三类技术。剖析了现有技术中存在的问题。总结分析了主要解决方法和模型,并对未来研究方向进行了展望。     

Fault tolerant aggregation in heterogeneous sensor networks

Fault tolerance and scalability are important considerations in the design of sensor network applications. Data aggregation is an essential operation in sensor networks. Multiple techniques have been proposed recently to tackle the issues of scalability and fault tolerance of aggregation in sensor networks. In this article, we analyze the impact of using a few of the more reliable, though expensive, nodes–such as the Intel XScale–called microservers, in addition to the standard motes, on the fault tolerance and scalability of the aggregation algorithms in sensor networks. In particular, we propose a simple model that captures the essence of tree aggregation in such heterogeneous sensor networks. We validate this theoretical model with simulation results. We also study the effective impact on the sustainable probability of failure, and perform cost-benefit analysis. We also show how hybrid aggregation can be utilized instead of tree, to improve the performance of aggregation in heterogeneous sensor networks. We show that our work can be applied for effectively optimizing the use of expensive hardware while designing fault-tolerant, distributed sensor networks.     

Identifying source(s) of out-of-control signals in multivariate manufacturing processes using selective neural network ensemble

In multivariate statistical process control (MSPC), most multivariate quality control charts are shown to be effective in detecting out-of-control signals based upon an overall statistic. But these charts do not relieve the need for pinpointing source(s) of the out-of-control signals. Neural networks (NNs) have excellent noise tolerance and high pattern identification capability in real time, which have been applied successfully in MSPC. This study proposed a selective NN ensemble approach DPSOEN, where several selected NNs are jointly used to classify source(s) of out-of-control signals in multivariate processes. The immediate location of the abnormal source(s) can greatly narrow down the set of possible assignable causes, facilitating rapid analysis and corrective action by quality operators. The performance of DPSOEN is analyzed in multivariate processes. It shows improved generalization performance that outperforms those of single NNs and Ensemble All approach. The investigation proposed a heuristic approach for applying the DPSOEN-based model as an effective and useful tool to identify abnormal source(s) in bivariate statistical process control (SPC) with potential application for MSPC in general.     

Fault-tolerant design of wireless sensor networks with directional antennas

A tree structure is often used in wireless sensor networks to deliver sensor data to a sink node. Such a tree can be built using directional antennas as they offer considerable advantage over the omni-directional ones. A tree is adequate for data gathering from all sensor nodes if no node fails. We study the problem of enhancing the fault tolerance of a data gathering tree by adding additional links so that failure of a sensor or a pair of adjacent sensors would not disconnect the tree. We prove that the least-cost tree augmentation problem is NP-complete and provide approximation algorithms one for single node failure and the other for a pair of adjacent node failure, with performance bounds of two and four respectively.