Search

Renewable energy becomes increasingly popular in the global electric energy grid, improving the accuracy of renewable energy forecasting is critical to power system planning, management, and operations. However, this is a challenging task due to the intermittent and chaotic nature of renewable energy data. To date, various methods have been developed, including physical models, statistical methods, artificial intelligence techniques, and their hybrids to improve the forecasting accuracy of renewable energy. Hence this research proposed to hybridize two strong deep learning algorithms where modeling of more complex functioning is allowed by the use of multiple layers of abstraction in order to come up with a powerful forecasting model to predict solar power generation over long term horizon. Finally, the Deep Neutral Network and Long-short Term memory Network (DNN-LSTM) method can generate predicted solar energy consumption in a fully connected hierarchy. The proposed DNN-LSTM model achieved Mean Square Error (MSE) of 0.00825 and MAE of 0.00100 respectively. This is by far the lowest value when compare against the existing model i.e MLSHM which has MSE of 0.05700 and MAPE of 0.00695, LSTM which has MSE of 0.0536 and MAE of 0.0037 and Gated Recurrent Unit (GRU) which has MSE of 0.03460 and MAE of 0.00243 respectively. Thus, the proposed DNN-LSTM have clearly enhanced the forecasting accuracy as against all the existing models that was used for the evaluation and achieved the lowest values in terms of validation of MSE and MAE.

Datasets are a major requirement in the development of breast cancer classification/detection models using machine learning algorithms. These models can provide an effective, accurate and less expensive diagnosis method and reduce life losses. However, using the same machine learning algorithms on different datasets yields different results. This research developed several machine learning models for breast cancer classification/detection using Random forest, support vector machine, K Nearest Neighbors, Gaussian Naïve Bayes, Perceptron and Logistic regression. Three widely used test data sets were used; Wisconsin Breast Cancer (WBC) Original, Wisconsin Diagnostic Breast Cancer (WDBC) and Wisconsin Prognostic Breast Cancer (WPBC). The results show that datasets affect the performance of machine learning classifiers. Also, the machine learning classifiers have different performances with a given breast cancer dataset

We have proposed a suited buffer size for Heterogeneous network traffics, with an average significant throughput and response delay.

This paper explains the verification of a hybrid model for prime decision-making in driving using mathematical analysis technique which defines the study problem. The hybrid model is the integration of Automaticity Recognition Prime Decision (RPD) training model and Situation Awareness (SA) model. Verification is a mechanism used to certify that the hybrid model simulation results, assumptions and specifications are correct and to also confirms that it is rightly built to achieve the design and implementation objectives. The verification of the model can be done in two phases thus; by automated and mathematical analysis. The first phase has been achieved. The mathematical analysis is describing the relevant system mathematically in the form of a partial differential equation (PDE) using the first order principle technique. It verifies the theoretical and structural correctness of the hybrid model. It also, determines the stability points that explains the presence or possibility of getting the equilibrium state of the system. This equilibrium occurs when for all states no changes occur. The presence of reasonable equilibrium is a sign of accuracy of the hybrid model. The results acquired evaluate the hybrid model validity and it also depicts the behavior of the driver on the road. Given that evaluation of the hybrid model for prime decision-making by verification using mathematical analysis is yet be achieved, this paper bridges the gap by providing the verification of the model by using the mathematical analysis to ensure that the hybrid model is rightly built, which is the novelty of this paper.

Choosing a suitable router buffer size is still a significant research problem. The small buffer used guarantees low packet delay but higher chances of packet loss. Similarly, larger buffer results in buffer bloat which causes a high delay in a network. Buffer bloat is a significant problem due to the high changing link characteristics of modern heterogeneous networks. The access links can have connections with speedy links (Gbps) with a small number of packet losses and also have a connection with a susceptible high packet loss with lowbandwidth links like wireless and last-mile connections. Hence, what may be thought to be a rational buffer size, might be flawed when link rates and delays fall below the minimum value. In this paper, extensive NS2 simulations of multi-flow traffic were run to obtain a better-suited buffer size that results in better average throughput and delays over the wired network without extra signaling and computation by the router. The results showed that having a buffer between five and ten packets gave the best average throughput with a low average delay at the congested router.

IEEE known as WiMAX supports wider coverage, higher bandwidth, and quality of service (QoS). 802.16e was introduced which made battery-life of Mobile Station (MS) a critical challenge Since MS are battery powered with an impose rechargeable life. An Efficient Battery Lifetime Aware Power Saving Scheme was proposed to minimize frequent transition of MS to listening-wake mode in order-to reduce power consumption. However, it increases average response delay due to a longer sleep interval used. Thus, an efficient sleep window based power saving scheme (ESPSS) is proposed to reduce the response delay. The ESPSS introduces an average based sleep window to minimize the longer sleep interval. The scheme also proposed a modified minimum and maximum sleep interval to reduce the response delay. The ESPSS was evaluated using a discrete event simulator. The results showed that ESPSS achieves a superior performance compared to the existing Scheme in terms of response delay while improving QoS.

This paper presents a virtualised network environment that serves as a stable and re-usable platform for the analysis of malware propagation. The platform, which has been developed using VMware virtualisation technology, enables the use of either a graphical user interface or scripts to create virtual networks, clone, restart and take snapshots of virtual machines, reset experiments, clean virtual machines and manage the entire infrastructure remotely. The virtualised environment uses open source routing software to support the deployment of intrusion detection systems and other malware attack sensors, and is therefore suitable for evaluating countermeasure systems before deployment on live networks. An empirical analysis of network worm propagation has been conducted using worm outbreak experiments on Class A size networks to demonstrate the capability of the developed platform.