A substation is the most important component of transporting or distributing electricity in any region or industry. To control and monitor the substation, different automation structures are developed in our country a...
详细信息
A substation is the most important component of transporting or distributing electricity in any region or industry. To control and monitor the substation, different automation structures are developed in our country and around the world. The proposed system is to develop an IoT-based smart solution for monitoring and controlling the transformer in a substation. The control unit serves as the hub for all the system equipment and activities. Using an ultrasonic sensor and the DHT11, this smart solution can monitor transformer oil and temperatures. When the temperature of the transformer climbs over the specified value, the cooling fan turns on and provides sufficient air to lower the temperature. The voltage and current sensors' collected data determine whether the circuit is open or closed. This eliminates the expense at the substation by minimizing operating costs. As a result, both observational and operational effectiveness will undoubtedly improve.
Mobile edge computing (MEC) represents the promising technology that targets at facilitating different resources for processing and storing near the edge of mobile devices. Nevertheless, limited availability of resour...
详细信息
IoT industry has come to the fore in this modern techno-frenzy age. This has caused digital data explosion and has kept the data storage industry on its toes. The paper proposes a novel blockchain-based data storage m...
详细信息
ISBN:
(纸本)9781665428644
IoT industry has come to the fore in this modern techno-frenzy age. This has caused digital data explosion and has kept the data storage industry on its toes. The paper proposes a novel blockchain-based data storage model where users contribute the storage space of their personal electronic devices to meet the growing data storage needs. The paper dives into a decentralized system design, along with an equitable compensation model for all the storage space contributing users. The decentralized framework provides a common platform for interaction between the storage space contributors and buyers. The proposed solution provides an alternative to cloud storage which relies heavily on servers.
Diabetes mellitus is the century’s epidemic. It is a metabolic disease and causes high blood sugar. Diabetes prediction is usually performed through collecting a blood sample and testing with some conditions. It is d...
详细信息
At present, Chinese electrical fire prevention and control is still in the development stage. The existing electrical fire monitoring products have technical defects, the false alarm rate is high, and the monitoring e...
详细信息
The issue of data center power consumption is becoming increasingly serious. One of the most effective techniques for increasing resource utilization and lowering power consumption is virtual machine consolidation. Th...
详细信息
Fairness has emerged as a crucial topic in data mining and machine learning applications, driven by ethical and legal considerations. It is important to recognize that not all samples are treated unfairly, resulting i...
详细信息
ISBN:
(数字)9789819991198
ISBN:
(纸本)9789819991181;9789819991198
Fairness has emerged as a crucial topic in data mining and machine learning applications, driven by ethical and legal considerations. It is important to recognize that not all samples are treated unfairly, resulting in data heterogeneity in fairmachine learning. Existing fair models primarily focus on achieving fairness across all heterogeneous data, yet they often fall short in ensuring fairness within specific subgroups, such as fairly treated and unfairly treated data. This paper presents a novel problem of training a fair model on heterogeneous data, aiming to achieve fairness for both types of data, with a particular emphasis on the unfairly treated subset. To address this challenge, an effective approach is to recover the distribution of both fairly and unfairly treated data. In this study, we adopt the Structural Causal Model (SCM) to model the heterogeneous data as a mixture of causal structures. Leveraging the perspective of SCM, we propose a framework called FairdR, which utilizes the Hirschfeld-Gebelein-R ' enyi (HGR) correlation to accurately recover the distribution of both fairly and unfairly treated data. FairdR can serve as a pre-processing method for other fair machine learning models, providing protection for the unfairly treated members. Through empirical evaluation on synthetic and real-world datasets, we demonstrate that the presence of heterogeneous data can introduce unfairness in previous algorithms. However, FairdR successfully recovers the distribution of fairly and unfairly treated data, thus improving the fairness of downstream algorithms when dealing with heterogeneous data.
The FastICA algorithm is a novel iterative algorithm, also known as the fixed-point algorithm, proposed by Hyvrinen et al, at the University of Helsinki, Finland, unlike conventional neural network algorithms, FastICA...
The FastICA algorithm is a novel iterative algorithm, also known as the fixed-point algorithm, proposed by Hyvrinen et al, at the University of Helsinki, Finland, unlike conventional neural network algorithms, FastICA uses a batch processing algorithm and requires larger data per iteration. In terms of both distribution and parallelism, the algorithm is still an artificial neural network-based algorithm. fastICA algorithms can be classified into three types: steepness, likelihood maxima, and negative entropy maxima. In this thesis, based on the FastICA algorithm based on negative entropy maximum, the algorithm has faster convergence speed and stronger robustness by introducing Newton iteration algorithm.
An essential instrument for the operation of a power system is to monitor and analyze the data to find the fault and rectify it before the System collapses completely. This paper intents to utilize the idea to create ...
详细信息
An essential instrument for the operation of a power system is to monitor and analyze the data to find the fault and rectify it before the System collapses completely. This paper intents to utilize the idea to create a control system that will fulfill three objectives, monitoring of vital parameters controlling the power distribution, outage management by fault detection based on the variation of voltage, frequency, and current & protection of the circuit against any significant incidents by isolating the load from utility and flagging the information through feedback to the utility authority. The method used in this project can provide necessary safety from total system outages by adequately monitoring the instant data and historic data, managing the outage system by detecting faults, and cutting loads required to avoid a widespread blackout of a power system. Implementation of the proposed project can solve the problem of system blackout due to overload, under/over voltage, or under/over frequency. This developed system can supply necessary timestamped monitored data that can be accessed remotely and can also archive to create a proper load profile to ultimately help the modeling of Load Forecasting for a smooth and economic grid operation and can be used for developing the Smart Grid network.
In this paper, several optimization techniques including the Particle Swarm Optimization (PSO) technique, the Genetic Algorithm (GA), and the Adaptive Neuro-Fuzzy Inference System (ANFIS) are applied to determine the ...
详细信息
In this paper, several optimization techniques including the Particle Swarm Optimization (PSO) technique, the Genetic Algorithm (GA), and the Adaptive Neuro-Fuzzy Inference System (ANFIS) are applied to determine the most efficient output for load frequency control. These optimization techniques analyze the optimal level of system performance. The goal of this paper is to identify the most effective optimization technique for this sophisticated LFC system. In this research, three strategies (PSO, GA, ANFIS) are used in the LFC system to analyze frequency fluctuation and compare the load change rate. The model consists of the transfer function of the governor, turbine, rotating mass, and load. In this analysis, the ideal performance is examined across three separate case scenarios. The MATLAB/SIMULINK software simulates the performance analysis, which offers more realistic data and is generally preferred in this sort of optimization strategy work.
暂无评论