The main challenges in designing downlink coordinated multicast beamforming in massive multiple-input multiple output (MIMO) cellular networks are the complex computational solutions and significant fronthaul overhead...
详细信息
At present, there is a huge trend in the industry for power converters that have the ability to reach a high efficiency at high switching frequencies. General power converters have switching losses which act as a barr...
At present, there is a huge trend in the industry for power converters that have the ability to reach a high efficiency at high switching frequencies. General power converters have switching losses which act as a barrier in terms of efficiency while designing high-frequency converters. As a solution for this, soft switching techniques come into play by integrating a resonant circuit into the traditional power converter. These resonant converters give the ability of the power converters to operate in high frequencies without switching losses. Also in electric vehicles as well as wireless charging, the switching frequency can be increased to a higher level without the switching losses using these converters. Furthermore, it can be used in bi-directional converters with MOSFETs like SiC and GaN. An analytical review of the advanced resonant converter design topologies is discussed in this paper based on three main designs for auxiliary-based circuits. Finally, a comparative analysis is conducted using the various converter parameters. This analysis will be a huge benefit for the power electronics circuit design area to select the most suitable type of circuit for the application.
Accurate estimation of lithium-ion battery state of health (SOH) is essential for safe and reliable operation, especially for e-mobility applications due to dynamic variation of operating parameters. Existing SOH esti...
Accurate estimation of lithium-ion battery state of health (SOH) is essential for safe and reliable operation, especially for e-mobility applications due to dynamic variation of operating parameters. Existing SOH estimation methods primarily depend on changes in battery terminal voltage and capacity fade. Experimental and practical data demonstrated that the battery capacity fade is significantly influenced by battery operating temperature and discharge current profiles. Therefore, existing methods fail to provide accurate SOH information in practical scenarios despite excellent performance in a laboratory environment. A solution to that, an adaptive and robust SOH estimation method based on normalized change of battery state of temperature is proposed in this paper for practical applications. The proposed SOH estimation method accommodates the influence of the rate of change of battery temperature due to battery aging, making the method highly adaptive to the change in operating parameters and the rise in battery internal resistances due to aging. The proposed method is then validated using experimental data which are collected on one 21700 NMC lithium-ion battery cell under a wide range of operating conditions.
Fast charging of lithium-ion battery (LIB) packs at low temperatures can have several effects on the performance and overall health of the battery. Repetitive fast charging at low temperatures accelerates internal res...
Fast charging of lithium-ion battery (LIB) packs at low temperatures can have several effects on the performance and overall health of the battery. Repetitive fast charging at low temperatures accelerates internal resistance growth, leading to inefficient charging. Slow and inefficient chemical reactions at low temperatures result in slower charging rates and increased heat generation. Furthermore, repeated fast charging at subzero temperatures accelerates degradation processes due to increased wear on the battery, significantly reducing the cycle life of the battery. This research paper presents a series of experimental studies conducted on a 21700 Lithium-Nickel-Manganese-Cobalt-Oxide (NMC) LIB cell to investigate the temperature gradient and its impact on battery performance at a wide range of ambient temperatures (-5°C to 25°C) and charging rate (1C to 2 C). The findings highlight the highest rate of change of surface temperature and differential temperature (15°C) with a charging rate of 2 C at ambient temperature of -5°C. Moreover, a reduction in battery discharge performance is observed during low-temperature charging compared to charging at 25°C with the same charging rate. These findings are crucial for the development of health-conscious fast charging algorithms, improved thermal management techniques, and the establishment of a thermal safety framework.
Rapid printed circuit board (PCB) development for modular battery management systems (BMS) can provide a flexible, scalable, and customizable solution for all lithium-ion battery-powered devices as well as in research...
Rapid printed circuit board (PCB) development for modular battery management systems (BMS) can provide a flexible, scalable, and customizable solution for all lithium-ion battery-powered devices as well as in research and development. Therefore, to satisfy the increasing demand and stringent requirements of high performance, low cost, and smaller footprints, a single-stage PCB development technique using a CO 2 laser, galvo scanner, and two-axis computer numerical control (CNC) router is proposed in this paper. The system can selectively cut/erode between the copper or substrate layers with only a single active laser by controlling the intensity of the laser beam. The proposed method circumvents the fabrication of a negative and the chemical processing steps of PCB manufacturing, resulting in higher production speed at a reduced cost. The technique is also environmentally friendly and suitable for small-batch production to support small industries and research. The superiority of the proposed method is also demonstrated by a comparison between the proposed and a high-density fiber laser-based method.
Effective requirements specification for embedded software systems relies heavily on the collaboration between stakeholders to articulate accurate functional and non-functional requirements. At present, requirement en...
Effective requirements specification for embedded software systems relies heavily on the collaboration between stakeholders to articulate accurate functional and non-functional requirements. At present, requirement engineers manually coordinate the sophisticated task of correct stakeholders selection for requirements and compilation of feedback in the embedded software domain. Because embedded software has an extensive number of stakeholders, the absence of an efficient collaboration platform causes a prolonged project completion time, elevated maintenance costs, or project failure. In this preliminary research paper, a stakeholder collaboration platform is proposed using an auto-encoder-based recommender system and SysML modeling language for embedded software systems. In the proposed framework, forums are the collaboration space for stakeholders to contribute to requirements analysis and specifications and are generated from the requirements diagram of the SysML modeling language. Owners of the requirements will be directly assigned to the forum from the SysML requirement profile information. To ensure adequate stakeholder engagement in the requirements specification process, the Collaborative Denoising Auto-Encoder (CDAE) recommender system is used for advanced auto-recommendations of requirement forums to stakeholders. This approach facilitates feedback collection and analysis to refine requirements specifications. The automatic forum creation and the advanced recommendation process of this framework will add no overhead to the requirements engineers or analysts; at the same time, a centralized collaboration platform for the stakeholders will save requirements analysis time and avoid future conflicts.
Several advancements were made in wireless charging for e-transportation however, the knowledge acquired in the context of wireless charging of electric cars are not sufficient to apply to electric bicycle (e-bikes). ...
Several advancements were made in wireless charging for e-transportation however, the knowledge acquired in the context of wireless charging of electric cars are not sufficient to apply to electric bicycle (e-bikes). Therefore, this paper reviewed state-of-the-art wireless charging technologies with a special emphasis on e-bike charging. A comparative analysis of different kinds of wireless charging techniques is also presented to provide a clear understanding and guidelines to choose the best suitable technology for implementation in e-bikes. In addition, current issues, challenges, and future research scopes are identified where necessary modifications need to be done before implementing the technology for e-bikes.
Driven by the increase in the availability of private information in digital format combined with the ever increasing interest in the use of this information for data mining and knowledge discovery, new regulations an...
Driven by the increase in the availability of private information in digital format combined with the ever increasing interest in the use of this information for data mining and knowledge discovery, new regulations and standards are being created to protect private user information and achieve meaningful consent in the digital realm. Classical privacy enforcing solutions such as access-control-lists and non-standardized data sharing agreements are not capable of effectively covering the requirements set out by these new privacy regulations. At this same time, blockchain smart contracts have shown the possibility of representing various object types with user defined interaction rules, all of which can be unambiguously interpreted by machine. In this paper, we represent the private records of an individual as blockchain assets and demonstrate how this allows for a user to control access to their private records thus clarifying the user consent process and providing better regulatory compliance. We also present the design of a hardware wallet protocol which stores asset data and only reveals this data if blockchain transaction approval can be shown.
Privacy by Design (PbD) is a set of guiding principles for providing stronger guarantees of privacy protection in systems that handle sensitive information. In this paper, we present the design and development of PbD ...
Privacy by Design (PbD) is a set of guiding principles for providing stronger guarantees of privacy protection in systems that handle sensitive information. In this paper, we present the design and development of PbD constructs with a proof of concept prototype in Python, where data variables are tagged with information describing their privacy requirements while common methods are augmented with privacy requirement checking code. With these additions, we demonstrate how the task of writing Privacy by Design compliant code is greatly simplified.
This paper studies the design of wireless federated learning (FL) for simultaneously training multiple machine learning models. We consider round robin device-model assignment and downlink beamforming for concurrent m...
This paper studies the design of wireless federated learning (FL) for simultaneously training multiple machine learning models. We consider round robin device-model assignment and downlink beamforming for concurrent multiple model updates. After formulating the joint downlink-uplink transmission process, we derive the per-model global update expression over communication rounds, capturing the effect of beamforming and noisy reception. To maximize the multi-model training convergence rate, we derive an upper bound on the optimality gap of the global model update and use it to formulate a multi-group multicast beamforming problem. We show that this problem can be converted to minimizing the sum of inverse received signal-to-interference-plus-noise ratios, which can be solved efficiently by projected gradient descent. Simulation shows that our proposed multi-model FL solution outperforms other alternatives, including conventional single-model sequential training and multi-model zero-forcing beamforming.
暂无评论