Jaidilson Jó da Silva

System identification and energy-aware processor utilization control

Minimization of power consumption is essential for battery-powered computing systems. This paper proposes an innovative solution for the processor power management problem. The proposed technique does not require accurate power consumption and performance model of the processor. The proposed solution exploits feedback control principles and the controller design is optimized for a given workload. Experimental results obtained from a case study are presented demonstrating, usefulness and the feasibility of the proposed solution.

Implementation of a Decentralized PID Control System on an Experimental Platform Using LabVIEW

In this paper, the implementation of a decentralized control system is done on an experimental platform, which is dedicated to the study of fouling in industrial tubes. This plant is modelled as a multivariable system of order 2 × 2, whose the inputs are the voltage and current applied respectively in the plant actuators, and the outputs are flow and pressure measurements obtained by sensors in the main tube. The decentralized PID control system is performed by a supervisory system developed in LabVIEW software to verify the impact of fouling on the current behavior of the experimental platform, and thus enable performing predictive maintenance.

Computer systems power model estimation

The estimation of power models for computer systems is important for the development process of dynamic power managers. In this work, a power model is estimated based on a constrained linear least-squares technique.

Identification and control for processor power management

Minimization of power consumption is essential for battery-powered computing systems. In this paper, System identification and Feedback control are applied to the processor power management problem when the workload is known a priori.

Hammer impact test applied for fouling detection in pipelines

The fouling detection in duct systems is a persistent problem and remains a relevant demand for the chemical, oil, food and pharmaceutical industries. The fouling is a deposition process of heavy organic and other dissolved solid materials in the transported fluids or suspensions onto inner wall surfaces in fluid transport systems. This work presents research results of hammer test for vibration analysis using non-invasive tests to fouling detection in pipelines. Data were taken from the vibration in presence of an inner pipe fouling layer using an accelerometer and a microphone for detection. With the analysis of the obtained results is possible to detect the fouling process, in the monitored pipelines, with and without load conditions, using a microphone and an accelerometer as sensors.

Estimating computer systems power consumption for time scaled experiments

Power policies are widely applied at computer systems such as desktops, notebooks, and smartphones. For example, the timeout policy to turn off the display. A power management policy must be tested by measuring the power consumption of an experimental platform running workloads. Such tests must be fast enough to feedback the development process with valuable information about the power policy performance. In this work, we have presented a methodology for estimating computer systems power consumption for time scaled experiments. Using the proposed methodology, we have reduced the overall time for developing a power management policy.

Dynamic timeout power policy for network interfaces

Network interfaces are present in a variety of battery powered computer systems, such as notebooks, ultrabooks, tablets and mobile phones. Considering that the battery life is still a constraint, there are opportunities for power management by reducing the power consumption of idle network interfaces. In this work, a dynamic timeout power policy for network interfaces is presented, incorporating workload estimation and a problem optimization at runtime, thus providing power savings and suitable performance.

Development of a Remote Laboratory for Control-engineering Education based on an Industrial Fluid Transport Platform.

This paper presents the development of a remote laboratory based on an industrial fluid transport platform. The goal is to improve the control-engineering education using new technologies, saving equipment and personnel for the institution and time and money for the remote students. The pilot plant was initially developed for the study of fouling detection and adapted in this work for the development of a laboratory, in which students and researchers can, over the Internet, perform experiments without any limitation of time and location. The LabVIEW software was used to implement the Human-Machine Interface (HMI) through a didactic interaction and the developed remote laboratory has been tested to be used in different disciplines.

Battery lifetime estimation by means of an analytical continuous-time model

Analytical battery models such as Rakhmatov-Vrudhula are tipically represented in discrete-time, and then implemented by software. However, by means of Model-based design and automatic code generation, it is possible to generate embedded code directly from a continuous time representation. Therefore, in this work we present an analytical battery model in continuous-time by means of the Laplace transform, which is a suitable representation for integrated development environments for model-based design. We show that the battery lifetime estimation error for our continuous-time representation is less than 1% compared to the discrete-time Rakhmatov-Vrudhula model.

Classification methods for detecting idle intervals of network interfaces

Network interfaces are components largely employed in battery powered mobile devices. Considering the need to save energy on these devices, power management policies were developed taking advantage of idle components, including network interfaces. To increase the efficiency of power management policies for network interfaces, the detection of idle periods with a high hit rate is relevant. In this work, a state-of-the-art method for detecting idle intervals of network interfaces is analysed and compared with a widely used method for classification problems.