Abdalraheem Ijjeh

Abstract A novel full wavefield processing method by using fully convolutional neural networks is presented. The full wavefield of propagating Lamb waves in the fibre-reinforced composite plate was simulated by the parallel spectral element method. It resembles a full wavefield measurements acquired on a surface of the plate by the scanning laser Doppler vibrometer. The aim of the proposed technique is an identification of delamination location, size and shape. It is achieved by pixel-wise image segmentation by using the end-to-end approach. It is possible because of the large dataset of Lamb wave propagation patterns resulting from interaction with delaminations of random location, size and shape. It is demonstrated that the proposed method, tested on numerical data, is performing better than conventional adaptive wavenumber filtering method which was developed in previous work. Moreover, it enables better automation of delamination identification so that the damage map can be created without user intervention. The method was also tested on experimental data acquired on the surface of the specimen in which delamination was artificially created by a Teflon insert. The obtained results with the deep learning approach show its capability to predict the delamination in the numerically generated dataset with high accuracy compared to the conventional damage detection approach. Furthermore, the deep learning model shows the ability to generalize to a further experiential set.

The dataset contains 475 simulated cases of full wavefield of Lamb waves propagation in a plate made of carbon fibre reinforced plastic (CFRP). The simulated 475 cases represent delaminations with different locations, shapes, and sizes. The following random factors were simulated in each case: delamination geometrical size (ellipse minor and major axis randomly selected from the interval [10 mm, 40 mm], delamination angle (randomly selected from the interval [0◦ , 180◦]), coordinates of the centre of delamination (randomly selected from the interval [0 mm, 250 mm − δ] and [250 mm + δ, 500 mm], where δ = 10 mm). The guided waves were excited at the centre of the plate by applying equivalent piezoelectric forces. The excitation was in the form of toneburst sine signal modulated by the Hann window. The carrier frequency is assumed 50 kHz, and the modulation frequency is 10 kHz.<br> In each delamination case, 512 frames were generated to visualise the propagation of Lamb waves and...

A Wireless Sensor Network (WSN) contains a large number of sensor nodes equipped with limited energy supplies. In most applications, sensor nodes are deployed in a random fashion. Therefore, battery replacement or charging is considered not practical. As a result, routing protocols must be energy-efficient to prolong the network’s lifetime. In this paper, we propose a new Dynamic Re-clustering LEACH-Based protocol (DR-LEACH) which aims to reduce the energy consumption and extending the network’s lifetime. The idea is to balance energy consumption of Cluster Heads (CHs) by generating clusters with almost equal number of nodes during each round of the network life time. To perform this, we first calculate the optimal number of CHs in each round, and based on that we calculate the optimal size of each cluster. Results show that the proposed protocol improves network lifetime and reduces overall energy consumption compared to LEACH and BCDCP protocols.

A novel full wavefield processing method by using fully convolutional neural networks is presented. The full wavefield of propagating Lamb waves in the fibre-reinforced composite plate was simulated by the parallel spectral element method. It resembles a full wavefield measurements acquired on a surface of the plate by the scanning laser Doppler vibrometer. The aim of the proposed technique is an identification of delamination location, size and shape. It is achieved by pixel-wise image segmentation by using the end-to-end approach. It is possible because of the large dataset of Lamb wave propagation patterns resulting from interaction with delaminations of random location, size and shape. It is demonstrated that the proposed method, tested on numerical data, is performing better than conventional adaptive wavenumber filtering method which was developed in previous work. Moreover, it enables better automation of delamination identification so that the damage map can be created without user intervention. The method was also tested on experimental data acquired on the surface of the specimen in which delamination was artificially created by a Teflon insert. The obtained results with the deep learning approach show its capability to predict the delamination in the numerically generated dataset with high accuracy compared to the conventional damage detection approach. Furthermore, the deep learning model shows the ability to generalize to a further experiential set.

Due to the advances in wireless communications and electronics technology, Wireless Sensor Networks (WSNs) are used in many applications such as civil and military application. A WSN is composed on many nodes each of which is basically equipped with a sensing device to collect data from the environment, a processing unit to do some operations on data, a transceiver to send and receive collected, and an energy source to provide the required energy to operate (usually a battery). In most applications sensor nodes are randomly deployed in the field. Therefore, battery replacement or charging is considered not practical. As a result, routing protocols must be energy-efficient to prolong the network lifetime. Researchers have been working to develop routing techniques that enhances the WSN lifetime among which is the hierarchical routing. In this paper, we present a recent survey of hierarchical routing protocols which are based on LEACH protocol. Specifically, we will show the network life time and energy consumption for each protocol. Furthermore, a comparison of these protocols in terms of advantages (improvements over LEACH), disadvantages, assumptions, and the Cluster Head selection criteria are provided. 1

A Wireless Sensor Network (WSN) contains a large number of sensor nodes equipped with limited energy supplies. In most applications, sensor nodes are deployed in a random fashion. Therefore, battery replacement or charging is considered not practical. As a result, routing protocols must be energy-efficient to prolong the network's lifetime. In this paper, we propose a new Dynamic Re-clustering LEACH-Based protocol (DR-LEACH) which aims to reduce the energy consumption and extending the network's lifetime. The idea is to balance energy consumption of Cluster Heads (CHs) by generating clusters with almost equal number of nodes during each round of the network life time. To perform this, we first calculate the optimal number of CHs in each round, and based on that we calculate the optimal size of each cluster. Results show that the proposed protocol improves network lifetime and reduces overall energy consumption compared to LEACH and BCDCP protocols. KEYWORDS Wireless sensors network (WSN), dynamic clustering, optimal cluster size, energy balancing, network lifetime, residual energy 1.INTRODUCTION Wireless Sensors Networks (WSNs) are widely considered as one of the interesting and rapidly developing fields. They have attracted great attention because of the diverse applications they support in both civilian and military sectors [1]. Typically, a WSN consists of a large number of low-cost, low-power, and multifunctional wireless sensor nodes with sensing, wireless communication and computation capabilities. In many applications, the sensor nodes are randomly deployed. Accordingly, the sensor nodes must organize themselves into a wireless network and cooperate to perform the required task. In addition, WSNs are usually battery powered which means it is very difficult to replace or recharge the batteries as soon as the nodes are deployed [2] [3]. Based on that, many techniques were proposed to achieve longer lifetime and efficient energy consumption. Clustering is one of the effective techniques used to save energy in WSNs [4].Clustering means organizing sensor nodes into different groups called clusters. In each cluster, sensor nodes can be either a Cluster Head (CH) or an ordinary member node. A CH is the group leader in each cluster. It collects sensed data from member nodes, aggregates, and transmits the aggregated data to the next CH or to the Base Station [5]. The role of an ordinary member node is to sense data from the environment in which they are deployed and send it to the corresponding CH.