Spark (mathematics)

About: Spark (mathematics) is a research topic. Over the lifetime, 7304 publications have been published within this topic receiving 63322 citations.

The STARK Framework for Spatio-Temporal Data Analytics on Spark.

Abstract: Big Data sets can contain all types of information: from server log files to tracking information of mobile users with their location at a point in time. Apache Spark has been widely accepted for Big Data analytics because of its very fast processing model. However, Spark has no native support for spatial or spatio-temporal data. Spatial filters or joins using, e.g., a contains predicate are not supported and would have to be implemented inefficiently by the users. Also, Spark cannot make use of, e.g., spatial distribution for optimal partitioning. Here we present our STARK framework that adds spatio-temporal support to Spark. It includes spatial partitioners, different modes for indexing, as well as filter, join, and clustering operators. In contrast to existing solutions, STARK integrates seamlessly into any (Scala) Spark program and provides more flexible and comprehensive operators. Furthermore, our experimental evaluation shows that our implementation outperforms existing solutions.

Spark Ignition Feedback Control by Means of Combustion Phase Indicators on Steady and Transient Operation

Abstract: In order to reduce fuel cost and CO2 emissions, modern spark ignition (SI) engines need to lower as much as possible fuel consumption. A crucial factor for efficiency improvement is represented by the combustion phase, which in an SI engine is controlled acting on the spark advance. This fundamental engine parameter is currently controlled in an open-loop by means of maps stored in the electronic control unit (ECU) memory: suchkind of control, however, does not allow running the engine always at its best performance, since optimal combustion phase depends on many variables, like ambient conditions, fuel quality, engine aging, and wear, etc. A better choice would be represented by a closed-loop spark timing control, which may be pursued by means of combustion phase indicators, i.e., parameters mostly derived from in-cylinder pressure analysisthat assume fixed reference values when the combustion phase is optimal. As documented in literature, the use of combustion phase indicators allows the determination of the best spark advance, apart from any variable or boundary condition. The implementation of a feedback spark timing control, based on the use of these combustion phase indicators, would ensure the minimum fuel consumption in every possible condition. Despite the presence of many literature references on the use combustion phase indicators, there is no evidence of any experimental comparison on the performance obtainable, in terms of both control accuracy and transient response, by the use of such indicators in a spark timing feedback control. The author, hence, carried out a proper experimental campaign comparing the performances of a proportional-integral spark timing control based on the use of five different in-cylinder pressure derived indicators. The experiments were carried out on a bench test, equipped with a series production four cylinder spark ignition engine and an eddy current dynamometer, using two data acquisition (DAQ) systems for data acquisition and spark timing control. Pressure sampling was performed by means of a flush mounted piezoelectric pressure transducer with the resolution of one crank angle degree. The feedback control was compared to the conventional map based control in terms of response time, control stability, and control accuracy in three different kinds of tests: steady-state, step response, and transient operation. All the combustion phase indicators proved to be suitable for proportionalintegral feedback spark advance control, allowing fast and reliable control even in transient operations. [DOI: 10.1115/1.4026966]

Comparative analysis for idle speed control: a crank-angle domain viewpoint

Abstract: Objectives for the idle speed control problem in an internal combustion, spark ignited engine are basically to achieve smooth engine operation at idle speeds, under significant accessory load disturbances. Previous work has focused on this problem for models developed in the time domain and, more recently, in the crank-angle domain, primarily for linear designs. In this paper, we present simulation results in a comparative analysis for linear and nonlinear control design techniques, for application to a simplified model of the GM V6 3800 IC engine. The control inputs are air and spark, and our focus will be on an engine model for control design that is based on the crank angle as the independent variable.

Electrical fuel control system for internal combustion engines

Abstract: An electrical fuel control system for internal combustion engines, wherein various data representing the operating conditions of an internal combustion engine are converted into digital signals which are operated on to generate such fuel controlling signals that suit the required characteristics of the engine, whereby the quantity of fuel injected as well as the spark timing are controlled to ensure the optimum amount of fuel injected and the optimum spark timing.