DSP based inverter control for alternate energy systems

This paper presents a DSP based algorithm to control inverters used in interfacing alternate energy systems with the electric utility. Since a constant and ripple free dc bus voltage is not ensured at the output of alternate energy sources, the main aim of the proposed algorithm is to make the output of the inverter immune to the fluctuations in the dc input voltage. In this paper a modified space vector pulse width modulation (SVPWM) technique is proposed which will maintain the quality of the ac output of the inverter, regardless of the ripple present at the inverter input. The principle is explained qualitatively and extensive experiments have been carried out to verify and validate the proposed algorithm. A 16-bit fixed-point C2000 family DSP from Texas Instruments was used as the controller to implement the proposed control algorithm.     

A Hybrid Attention-Based Residual Unet for Semantic Segmentation of Brain Tumor

Segmenting brain tumors in Magnetic Resonance Imaging (MRI) volumes is challenging due to their diffuse and irregular shapes. Recently, 2D and 3D deep neural networks have become famous for medical image segmentation because of the availability of labelled datasets. However, 3D networks can be computationally expensive and require significant training resources. This research proposes a 3D deep learning model for brain tumor segmentation that uses lightweight feature extraction modules to improve performance without compromising contextual information or accuracy. The proposed model, called Hybrid Attention-Based Residual Unet (HA-RUnet), is based on the Unet architecture and utilizes residual blocks to extract low- and high-level features from MRI volumes. Attention and Squeeze-Excitation (SE) modules are also integrated at different levels to learn attention-aware features adaptively within local and global receptive fields. The proposed model was trained on the BraTS-2020 dataset and achieved a dice score of 0.867, 0.813, and 0.787, as well as a sensitivity of 0.93, 0.88, and 0.83 for Whole Tumor, Tumor Core, and Enhancing Tumor, on test dataset respectively. Experimental results show that the proposed HA-RUnet model outperforms the ResUnet and AResUnet base models while having a smaller number of parameters than other state-of-the-art models. Overall, the proposed HA-RUnet model can improve brain tumor segmentation accuracy and facilitate appropriate diagnosis and treatment planning for medical practitioners.     

The challenges of land development for housing provision in New Zealand

The land development process in New Zealand is criticised for causing delays in the delivery of adequate housing. These delays upset the demand and supply equilibrium, leading to housing shortages, expensive builds and rentals. This study investigates the challenges in land development, to ascertain the factors that are limiting its efficiency as a major catalyst to housing provision. An understanding of the complexities of the development process could enable the suggestion of feasible solutions for achieving housing goals. A two-stage process was adopted to achieve this study objective. In stage one, a critical review of relevant literature helped to identify 48 measurement items. Those items were included in a questionnaire survey in the second stage, to gather data from stakeholders involved in land development process in New Zealand. Using relative importance index (RII) method, nine significant challenges were identified, which were then categorised and discussed in accordance with the construction stakeholder groups that are responsible for creating those challenges. The nine major challenges are: delay in reviews and approval of documents; scope change; lengthy consent application processes; late response to queries by regulatory authorities; poor interaction between regulatory authorities; poor coordination within regulatory authorities; poor planning and scheduling; design errors and slow progress during design development. Findings of this study highlight the need for the development of proper workplan for consent processing, reasonable factoring of the risks associated with scope changes in the land development process, and the enhancement of project management skills of land development contractors.

     

A DSP-based active power filter for low voltage distribution systems

The use of nonlinear loads, which inject undesired harmonic currents into low voltage distribution systems, is increasing rapidly. Active power filters are being considered as a potential candidate for solving harmonic problems in order to meet harmonic standards and guidelines. A new digital signal processor (DSP)-based control method for a single-phase active power filter (APF) is presented in this paper. Compared to conventional analog-based methods, the DSP-based solution provides a flexible and cheaper method to control the APF. The proposed scheme employs a carrier-based control that requires less feedback information compared to other reported solutions. Only one current sensor is used to sense the nonlinear load current and two voltage sensors to sense the input supply voltage and the dc bus voltage. The proposed method provides both harmonic elimination and power factor correction. The PSpice simulation and experiments using the DSP-based prototype are made to verify the feasibility of the method.     

Adaptive biomimetic facades: Enhancing energy efficiency of highly glazed buildings

In this paper, we investigate the design of an adaptive biomimetic facade as a practical solution for enhancing energy efficiency of highly glazed buildings in the hot and humid regions. We present an adaptive facade that reduces solar heat gain and hence the energy consumption of the building, with minimal reduction in visual comfort (i.e., indoor lighting levels and visibility to the outer environment) of the users. The basic module of the facade consists of four shading devices that can be folded along both horizontal and vertical axes. The design enables shading under both high and low sun angles, without blocking visibility to the outdoor environment. To develop the facade, we explore and mimic the physical, physiological and adaptation properties of an Oxalis oregana—a leaf that has the natural ability to track sun path and change its angle/position accordingly. As a case study for the proposed facade, we take an existing 20-story office building in the hot and humid climate of Lahore, Pakistan. Our numerical results indicate that after retrofitting of the designed facade, the building's existing energy load decreases by 32%. Moreover, 50% of the interior space (as opposed to 55% before the retrofitting) still has lighting level within the recommended range of 500–750 lux. The investigation demonstrates that the proposed biomimetic facade can significantly reduce the energy consumption, with minimal reduction in visual comfort, of highly glazed buildings.     

S1P Lyase siRNA Dampens Malignancy of DLD-1 Colorectal Cancer Cells

Sphingosine-1-phosphate lyase 1 (S1P lyase or SGPL1) is an essential sphingosine-1-phosphate-degrading enzyme. Its manipulation favors onset and progression of colorectal cancer and others in vivo. Thus, SGPL1 is an important modulator of cancer initiation. However, in established cancer, the impact of retrospective SGPL1 modulation is elusive. Herein, we analyzed how SGPL1 siRNA affects malignancy of the human colorectal cancer cells DLD-1 and found that in parallel to the reduction of SGPL1 expression levels, migration, invasion, and differentiation status changed. Diminished SGPL1 expression was accompanied with reduced cell migration and cell invasion in scratch assays and transwell assays, whereas metabolic activity and proliferation was not altered. Decreased migration was attended by increased cell–cell-adhesion through upregulation of E-cadherin and formation of cadherin-actin complexes. Spreading cell islets showed lower vimentin abundance in border cells. Furthermore, SGPL1 siRNA treatment induced expression of epithelial cell differentiation markers, such as intestinal alkaline phosphatase and cytokeratin 20. Hence, interference with SGPL1 expression augmented a partial redifferentiation of colorectal cancer cells toward normal colon epithelial cells. Our investigation showed that SGPL1 siRNA influenced tumorigenic activity of established colorectal cancer cells. We therefore suggest SGPL1 as a target for lowering malignant potential of already existing cancer.     

Design and Analysis of High-Speed Bottleneck-Free One-Dimensional OCDMA Based Gigabit Symmetric PON

Wireless Personal Communications - Wireless sensor network (WSN) is an emerging research field in recent years. The advancement in sensory device and communication technologies has enabled the...