No abstract

Reliable communication is the backbone of advanced metering infrastructure (AMI).Within the AMI, the neighbourhood area network (NAN) transports a multitude of traffic, each with unique requirements. In order to deliver an acceptable level of reliability and latency, the underlying network, such as the wireless mesh network (WMN), must provide or guarantee the quality-of-service (QoS) level required by the respective application traffic. Existing WMN routing protocols, such as optimised link state routing (OLSR), typically utilise a single metric and do not consider the requirements of individual traffic; hence, packets are delivered on a best-effort basis. This paper presents a QoS-aware WMN routing technique that employs multiple metrics in OLSR optimal path selection for AMI applications. The problems arising from this approach are non deterministic polynomial time (NP)-complete in nature, which were solved through the combined use of the analytical hierarchy process (AHP) algorithm and pruning techniques. For smart meters transmitting Internet Protocol (IP) packets of varying sizes at different intervals, the proposed technique considers the constraints of NAN and the applications' traffic characteristics. The technique was developed by combining multiple OLSR path selection metrics with the AHP algorithm in ns-2. Compared with the conventional link metric in OLSR, the results show improvements of about 23% and 45% in latency and Packet Delivery Ratio (PDR), respectively, in a 25-node grid NAN.

Multisource energy data, including from distributed energy resources and its multivariate nature, necessitate the integration of robust data predictive frameworks to minimise prediction error. This work presents a hybrid deep learning framework to accurately predict the energy consumption of different building types, both commercial and domestic, spanning different countries, including Canada and the UK. Specifically, we propose architectures comprising convolutional neural network (CNN), an autoencoder (AE) with bidirectional long short-term memory (LSTM), and bidirectional LSTM BLSTM). The CNN layer extracts important features from the dataset and the AE-BLSTM and LSTM layers are used for prediction. We use the individual household electric power consumption dataset from the University of California, Irvine to compare the skillfulness of the proposed framework to the state-of-the-art frameworks. Results show performance improvement in computation time of 56% and 75.2%, and mean squared error (MSE) of 80% and 98.7% in comparison with a CNN BLSTM-based framework (EECP-CBL) and vanilla LSTM, respectively. In addition, we use various datasets from Canada and the UK to further validate the generalisation ability of the proposed framework to underfitting and overfitting, which was tested on real consumers’ smart boxes. The results show that the framework generalises well to varying data and constraints, giving an average MSE of ∼0.09 across all datasets, demonstrating its robustness to different building types, locations, weather, and load distributions.

Abstract-Recent advances in ad hoc Wireless Mesh Networks (WMN) has posited it as a strong candidate in Smart Grid's Neighbourhood Area Network (NAN) for Advanced Metering Infrastructure (AMI). However, its abysmal capacity and poor multi-hoping performance in harsh dynamic environment will require an improvement to its protocol stacks in order for it to effectively support the variable requirements of application traffic in Smart Grid. This paper presents a classification of Smart Grid traffics and examines the performance of HWMP (which is the default routing protocol of the IEEE 802.11s standard) with the Optimised Link State Routing (OLSR) protocol in a NAN based ad hoc WMN. Results from simulations in ns-3 show that HWMP does not outperform OLSR. This indicates that cross layer modifications can be developed in OLSR protocol to address the routing challenges in a NAN based ad hoc WMN.

No abstract