Search

Enhancers are short DNA regulatory elements which play a vital role in gene expression. Due to their important roles in genomics, several computational models have been proposed in the literature for identification of enhancers and their strengths using traditional machine learning algorithms, however, the proposed models are unable to identify enhancers and their strength with reasonable accuracy because of high non-linearity in DNA sequences. This article proposes a two-level intelligent model based on Deep Neural Network (DNN) along with multiple feature extraction methods. Firstly, the proposed model represents the given DNA sequences into feature vectors using Pseudo K-tuple Nucleotide Composition (PseKNC) and FastText methods. Secondly, the features vectors are fused to make a heterogeneous features vector that considered the local and global correlation amongst the given sequences along with internal structure information. Finally, the heterogeneous feature vector is given to a DNN model to make final predictions. The proposed iEnhancer-DHF is developed using two-layer approach. The first layer predicts whether the given DNA samples are enhancers or non-enhancers whereas the second layer identifies either the enhancers are strong enhancers or weak enhancers. The outcome of the proposed model was rigorously assessed using both training and independent datasets via 10-fold cross validation method. The validation outcome demonstrated that the iEnhancer-DHF model yielded accuracies 86.07% and 69.60% at first layer and second layer respectively utilizing the training dataset. Similarly, the model yielded accuracies 83.21% and 67.54% at first layer and at second layer respectively by using the independent dataset. Additionally, the outcomes of the proposed model was initially compared with widely applied classifiers such as Support Vector Machine, Random Forest and K-nearest Neighbor and subsequently the performance is compared with the existing models using both the training and independent datasets. The comparison results exhibited that the iEnhancer-DHF model performed superior than the recently published models.

Generative Adversarial Networks (GANs) have gained notoriety by generating highly realistic images. The present work explores GAN for simulating High Energy Physics detectors, interpreting detector output as three-dimensional images. The demands and requirements of a scientific simulation are quite stringent, as compared to the domain of visual images. Image characteristics such as pixel intensity and sparsity, for example, have very different distributions. Moreover, detector simulation requires conditioning on physics inputs, and domain knowledge becomes essential. We, therefore, adjust the pre-processing and incorporate physics-based constraints in the loss function. We also introduce a multi-step training process based on transfer learning by breaking up the task complexity. Validation of the results primarily consists of a detailed comparison to full Monte Carlo in terms of several physics quantities where a high level of agreement is found (ranging from a few percent up to 10% across a large particle energy range). In addition, we assess the performance by physics unrelated metrics, thereby proving further the variability and pertinence through diverse standpoints. We have demonstrated that an image generation technique from vision can successfully simulate highly complex physics processes while achieving a speedup of more than three orders of magnitude in comparison to the standard Monte Carlo.

With recent advancement in computational biology, high throughput next generation sequencing technology has become a de facto standard technology for genes expression studies including DNAs, RNAs and proteins. As a promising technology, it has significant impact on medical sciences and genomic research. However, it generates several millions of short DNA and RNA sequences with several petabytes size in single run. In addition, the raw sequencing datasets such as RNAs are increasing exponentially leading to a big data analytics issue in computational biology. Due to the explosive growth of RNA sequences, the timely classification of RNAs sequence into piRNAs and non-piRNAs have become a challenging issue for traditional technology and conventional machine learning algorithms. Parallel and distributed computing models along with deep neural network have become a major computing platform for big data analytics now required in the field of computational biology. This paper presents a computational model based on parallel deep neural network for timely classification of large number of RNAs sequence into piRNAs and non-piRNAs, taking advantages of parallel and distributed computing platform. The performance of the proposed model was extensively evaluated using two-fold performance metrics. In the first fold, the performance of the proposed model was assessed using accuracy-based metrics such as accuracy, specificity, sensitivity and Matthews’s correlation coefficient. In the second fold, computational-based metrics such as computation times, speedup and scalability were observed. Moreover, initially the performance of the proposed model was assessed using real benchmark dataset and subsequently the performance was assessed using replicated benchmark dataset. The evaluation results in both cases showed that the proposed model improved computation speedup in order of magnitude in comparison with sequential approach without affected accuracy level.

Use of cloud resources has increased with the increasing trend of organizations and governments towards cloud adaptation. This increase in cloud resource usage, leads to enormous amounts of energy consumption by cloud data center servers. Energy can be conserved in a cloud server by demand-based scaling of resources. But reactive scaling may lead to excessive scaling. That, in turn, results in enormous energy consumption by useless scale up and scale down. The scaling granularity can also result in excessive scaling of the resource. Without a proper mechanism for estimating cloud resource usage may lead to significant scaling overheads. To overcome, such inefficiencies, we present Cartesian genetic programming based neural network for resource estimation and a rule-based scaling system for IaaS cloud server. Our system consists of a resource monitor, a resource estimator and a scaling mechanism. The resource monitor takes resource utilizations and feeds to the estimator for efficient estimation of resources. The scaling system uses the resource estimator's output for scaling the resource with the granularity of a CPU core. The proposed method has been trained and tested with real traces of Bitbrains data center, producing promising results in real-time. It has shown better prediction accuracy and energy efficiency than predictive scaling systems from literature.

This article presents a novel strategy regarding the stabilization control problem for plants with unmatched uncertainties. The methodology is based on Adaptive Smooth Super Twisting Sliding Mode Control. At first, as an initial step, the plant with unmatched uncertainty is transformed into a plant with matched uncertainty. At the second step, the plant with matched uncertainty is decomposed into a unique framework containing the nominal part and some unknown terms (where these unknown terms are computed adaptively). The nominal system is stabilized by using Smooth Super Twisting Sliding Mode Control. The stabilizing controller for the plant with matched uncertainty is designed in a way; it contains some nominal control plus some compensator term. The stability of the said technique is presented impressively. The compensator controller and the adapted laws are derived in such a way that the time derivative of a Lyapunov function becomes strictly negative. The proposed method is tested on a fourth-order plant. The simulation results show the effectiveness and validity of the proposed controller.

Advance Digital Signal Processing (DSP) techniques are controlling a wide range of signals in Smart Grid (SG) and narrowing the communication gap between various SG entities. Therefore, DSP is considered the backbone of SG that will meet future needs. The DSP interactions within SG are vital for SG reliability and efficiency. The advanced techniques for DSP interactions within SG are playing a pivotal role in improving the Power Quality (PQ). PQ includes technical challenges, such as instrumentation, measurement, and DSP techniques to increase knowledge of disturbances suffered by SG. PQ disturbances can have significantly different parameters and SG must deal with wide ranges of voltage and current magnitude disturbances, frequency drifts, and duration of disturbances. Power electronic components, such as (a) AC/DC converters, (b) adjustable-speed drives, (c) inverters, and (d) light dimmer controls the flow of power by de-shaping, flattening, or chopping sinusoidal power system voltages and currents. The aforementioned distortion of waveforms can affect PQ and create problems for nearby loads. SG ensures to operate linear and non-linear loads: (a) without damaging them, and (b) without reducing the efficiency of the power system.

This article introduces a quantile penalized regression technique for variable selection and estimation of conditional quantiles of counts in sparse high-dimensional models. The direct estimation and variable selection of the quantile regression is not feasible due to the discreteness of the count data and non-differentiability of the objective function, therefore, some smoothness must be artificially imposed on the problem. To achieve the necessary smoothness, we use the Jittering process by adding a uniformly distributed noise to the response count variable. The proposed method is compared with the existing penalized regression methods in terms of prediction accuracy and variable selection. We compare the proposed approach in zero-inflated count data regression models and in the presence of outliers. The performance and implementation of the proposed method are illustrated by detailed simulation studies and real data applications.

Some areas in Pakistan are still without excess to electricity and have no utility grid structure. Recently they are using diesel generators with fuel to fulfill their requirements. To overcome these problems, renewable energy technologies are capable of providing the electrical demand in those areas which have a shortage of electricity. Native renewable energy resources (RES) in remote areas can be utilized to establish a hybrid energy system that meets the electrical supplies of the community. This paper study a hybrid renewable energy system (HRES) based on solar, wind, and diesel resources to meet different industrial needs of the selected area. Industrial Estate Gadoon (District Swabi), Khyber Pakhtunkhwa, Pakistan is selected as a project site where there is a continuous need for a reliable energy resource throughout the year. The HOMER optimization software is utilized to design and analyze the system with estimated load requirements and existing energy resources. Optimization of the hybrid renewable energy system (HRES) has been done by selecting the best components and cheap, efficient, reliable, and cost-effective alternative energy systems. The system proposed comes out to have a very low cost of energy (i.e., 0.0655 $ /kWh). Sensitivity analysis of the proposed system shows that with an increase of 10% in wind speed, COE, and the total operating cost are reduced. This HRES proposed in the study can prove its efficiency and reliability at any place with the same system conditions.

The socio-economic development of our rural areas is mainly affected by the shortage of electricity that is carried out in these remote areas. This situation can be readily mitigated by utilizing renewable energy resources present in these areas so that a continuous supply of energy is maintained. In this study, a combined hydro and Solar renewable energy system is proposed for a small village in Muzaffargarh District at Muzaffargarh-Garhmor road near the TP-Link canal, Pakistan. A combined micro-hydro and solar system is proposed that is capable of meeting the energy requirements in this specific village by utilizing the existing energy resources available. Moreover, the continuous supply of electrical energy is ensured by integrating this Hybrid Energy System (HES) with the local area grid station. The proposed system along with satisfactorily fulfilling the energy demands of this area ensured the cheap production of electricity at the rate of 0.0438 $/kWh. The system efficiency can be easily judged by the fact that the proposed system in this study is capable of generating electricity even cheaper than the energy supplied by the grid.

Datacenter networks play a vital role in accommodating massive data transmission, such as in Giga/Terabytes on multiple links. With this inevitable data transmission, some uncertainties like dynamic traffic behavior, topology asymmetry, and link failure are major issues that data center faces, resulting in poor performance. The load balancing scheme should be accurate, especially path sensing and timely responses to mitigate the data center network's fallouts. Different solutions have been proposed to overcome these issues. Some of the proposed techniques are unaware of path conditions. They reroute at a fixed data rate blindly. Others could sense the path condition but reroute and adjust the load, balancing very slowly, degrading the performance. A few research studies demand hardware changes, particularly for acquiring accurate congestion feedback from the end host and adopting load balancing. Thus they can not react timely to these uncertainties, which results in reasonable performance degradation, especially in the worst situation to detect some issues like random packets drop and throughput of the channel. In this paper, a proposed solution focuses on two things (1) accurate timely acknowledgment of packet delivery, (2) Current load on each link. In our proposed work, there is no need for hardware changes. In our proposed scheme, the accurate response of acknowledgment of each packet delivery is achieved. The proposed scheme has been implemented using the NS3 simulator. The results of our proposed scheme have been compared with the other existing state of the art solutions.

Wireless sensor networks have become so popular in many applications such as vehicle tracking and monitoring, environmental measurements and radiation analysis. These applications can be ready to go for further processing by connecting it to remote servers through protocols that outside world used such as internet. This brings IPv6 over low power wireless sensor network (6LowPAN) into very important role to develop a bridge between internet and WSN network. Though a reliable communication demands many parameters such as data rate, effective data transmission, data security as well as packet size etc. A gateway between 6lowPAN network and IPV6 is needed where frame size compression is required in order to increase payload of data frame on hardware platform.

Landslides are the calamities responsible for the losses of lives and property at large. Early warning systems to predict the upcoming landslide can be of great advantage in developing a guide to control and alleviate these disasters. In this research, we have used artificial neural network (ANN) and long short term neural network (LSTM) for landslide prediction in Pakistan. The data has been collected from various regions of Pakistan including Jammu and Kashmir, Chitral, Gilgit Baltistan and Hazara division. A total of 12 parameters: surface pressure, specific humidity, surface runoff, rainfall, soil moisture at root level, soil moisture at 10cm, average temperature, transpiration, wind speed, aspect, slope and elevation are extracted for each location. The developed models have achieved accuracy greater than 95%.

Safe cities initiative promises the eradication of crimes specially the volume crimes which constitute fighting, shooting, and vandalism. This paper focuses on the detection of real-world volume crimes in a video using spatio-temporal convolutional neural network. Firstly, a benchmark dataset has been annotated on the fine-grained level to ensure reliable training. Secondly, a comparison is made between two state of the art networks recently used for spatio temporal analysis, over the developed dataset. Also, detailed analysis has been performed for detection and classification over the two networks. From the comparison it is concluded that the performance of C3D model is better than ResNet3D for binary classification. Whereas, ResNet3D outperforms C3d incase of multi-class classification.

In the era of technological advancement, numerous energy sources have been discovered for facilitation of human life on earth across the globe. Major renewable sources for energy are solar, wind, hydro, ocean/tidal, geothermal, and biomass. Ocean energy is a form of hydro energy which is captured by wave or tidal current stream. Marine tidal stream is reflected as a reliable renewable power generation source owing to its predictability present in tidal current. Globally, tidal energy production for the coastal areas, is estimated around 1 terawatts (TW/h/) per annum. On the contrary, availability of useful energy for power generation by tidal resource vary between 120 and 400 GW, dependent on the location and the prospective potential for conversion. Currently, Pakistan is facing power crisis which needs to be catered by utilizing various energy resources. The total energy requirement in Pakistan is about 25000MW approximately. One of the potential renewable sources that can cater energy need is tidal power generation. Pakistan owing to its vast coastline of about 1040 km long have a great potential to invest in tidal energy generation. In the year 1988, the National Institute of Oceanography (NIO) executed a study resulting that the coastal belt of Pakistan can have sufficient tidal energy generation at 17 creeks locations at economical rates. Therefore, energy shortfall in Pakistan can be catered via tidal energy. Enabling Pakistan to enter in the pool of green tidal energy generation region globally. In this paper, we will discuss energy sources, recent technology in renewable, tidal energy generation, benefits, and challenges for renewable energy generation using tidal energy. Furthermore, its comparison with other energy generating sources in Pakistan. For instance, coal, hydel, solar, wind, biogas, solid waste and nuclear power generation. Hence energy potential from underwater tidal turbine can be exploited for benefit of Pakistan.

The utilization of sustainable power source is expanding everywhere throughout the world. This research talked about the feasibility of giving power by utilizing hybrid sustainable power system to a far off small village, the populace of 15 families with a normal of five to six members in each family. The region of study, Kandhkot located in the Sindh, Pakistan. In the current scenarios, day by day vitality utilization of 500 kWh and a peak maximum load of 104 kW were taken into account. This research paper shows an approach to study different situations of sustainable power source system based on the economy, production of power, design, control, biomass accessibility and emissions of pollution. The HOMER programming software is utilized to completely simulate various situations. The output of this analysis demonstrates that the value of maximum capacity per year increases, the total present expense and levelized cost of vitality diminishes. The analysis of sensitivity parameters demonstrated that this design is the only optimized practical system under given varieties of various parameters.

The complex power system accommodating massive penetrations of distributed renewable energy sources is known as Smart Grid (SG). The SG allows bidirectional energy and power flow due to the deployment of communication technologies and generates massive digital data that need to be efficiently processed. Therefore, to process the massive data, Digital Signal Processing (DSP) filters are employed to filter the required data and these filters play a key role in SG development and solve major complex challenges. This paper emphasis DSP filters deployed for enhanced operation of SG. A filter allows one band of amplifiers or transistors. Filters can isolate desired signals from undesired signals and block interfering signals. Further, the frequency response is the most important factor and is used to categorize filters. Moreover, our work presents a detailed study of various filters using pictorial and tabular descriptions. Finally, adaptive filters are thoroughly explained in this paper.

The evident increase in energy demand and the utilization of rapidly depleting fossil fuels are causing serious environmental impacts. To mitigate these effects Renewable energy production is the only way out and solar energy is promising to be a more and more reliable energy source than ever before. In solar technology, Floating solar PV (FSPV) is being proposed as a suitable alternative with several advantages other than the preservation of land and water resources. In this study, we analyzed the combined hydro and FSPV at Taunsa Barrage, Pakistan where a low head hydropower project is recently proposed. Our study shows that the combination of the proposed FSPV system could provide a considerable amount of electrical energy in conjunction with hydroelectric energy by utilizing the existing structure (transmission and distribution network) of the hydropower project. Moreover, we can produce about 10% more electrical energy through FSPV as compared to the land-based Solar system. FSPV in the study performed in this paper is capable of producing up to 94MW of Electrical Energy which proves its importance and utility in mitigating the energy deficiency in Pakistan.

The capacitor voltage regulation is an important parameter in the stable and effective operation of Modular Multilevel Converter (MMC). This paper presents the different modulation techniques used in MMC. Phase Shift (PS) Pulse Width Modulation (PWM) have a natural capability of balancing submodule (SM) capacitor voltage. However, in Level Shifted (LS) modulation techniques lag due to this issue. Similarly, Nearest Level Control (NLC) technique have also the same issue of capacitor voltage imbalances. The possible remedies for each modulation technique to regulated capacitor voltage are implemented for synthesizing a well-shaped and less distorted output voltage waveform. The simulation is carried out in MATLAB/Simulink for five-level MMC.

Under-water wireless communication link has been a topic of higher interest due to an increase in various underwater operations. Such operations demand higher bandwidth with a fast velocity. This paper presents the experimental and FEM based simulations to investigate the radio waves based wireless link because of its higher magnitude in bandwidth and speed as compared to acoustic and optical technologies. Conduction current method is employed to transmit and receive the data in a conductive medium (Saltwater) with varying propagation heights. The investigations reveal that at the shallow water (less than the length of dipole, below sea level) produce significant surface waves (Airwaves) resulting in unidentified data reception. The height below sea level, more than the dipole length has resulted in the least BER as such it may be called an authentic data reception model.

In modern homes, energy consumption accounts for most of the economic aspects and environmental sustainability. Intelligent energy management and its control play an important role in energy supply and demand; and it will change behavior and environmental changes. For energy management and its control, a hybrid Internet of Things (IoT) and personal wireless network-based devices have been developed. In terms of the need-side-management approach, the use of energy can be intelligently controlled by the device for greater durability. In this study, electricity consumption and utilization are categorized accurately based on data collected from consumer behavior in energy consumption and utilization. First, the data cut through the device is used to identify and summarize the power consumption patterns hidden in the data. Second, the different linear mode algorithms extracted from the Schick-Lear Python library will be used for energy consumption and its intelligent power control. By analyzing different algorithms, the predictive score is found to be sufficiently efficient for the recurrence prediction, while the multi-step and lead-time technique proved to be suitable for multidimensional energy prediction. Results show that root squared mean error (RSME) performance of the predictive model increased by 35% in the lead time approach. Similarly, in per day approach it is 33% more efficient than the recursive model when residual energy forecasting is utilized.

This paper presents a PSCAD model review platform (PMView) developed in ERCOT to evaluate the performance of PSCAD models associated with power system dynamic devices, especially inverter-based resources (IBRs). Currently, PMView includes voltage and frequency response tests, instantaneous voltage playback, short-circuit ratio (SCR) limit tests, and a subsynchronous oscillation (SSO) testbed. PMView can be utilized to assess the PSCAD model performance under different disturbances. With a previously developed PSS/e model review platform (DMView), the PSCAD and PSS/e model performance can be compared to identify the validity of the developed models and system conditions where PSS/e models may not be adequate, for example in a weak grid condition. PMView also provides an instantaneous voltage playback module to facilitate the communication with IBR manufactures regarding the problematic PSCAD model response.

This paper reports on the use of an interdisciplinary project-based learning approach for undergraduate engineering education in machine/deep learning, and the internet of things (IoT). Machine learning has evolved from pattern recognition and is an important element of artificial intelligence. IoT has also seen rapid growth in multiple application domains including embedded systems, wireless sensor networks, control systems, automation, and sensors. A challenge for traditional electrical/computer engineering curriculum is to effectively educate students in these areas through hands-on activities and projects. There is a need to develop a project-based learning approach to involve undergraduate students in real-world problem solving to develop use cases of machine learning and IoT. This paper reports on the implementation of an interdisciplinary project-based learning approach followed in the undergraduate electrical/computer engineering curriculum. The students were involved in solving real-world problems through machine/deep learning. They also developed IoT applications in multiple domains to address the limitations of existing systems and to go through the engineering design process. The qualitative results indicate that the PBL approach was highly effective in improving their learning outcomes.