China
Africa
Explore chapters and articles related to this topic
Information Granularity, Information Granules, and Granular Computing
Published in Witold Pedrycz, Granular Computing, 2018
Granulation of time. Time is another important and omnipresent variable that is subjected to granulation. We use seconds, minutes, days, months, and years. Depending upon a specific problem we have in mind and who the user is, the size of information granules (time intervals) could vary quite significantly. To high-level management, time intervals of quarters of a year or a few years could be meaningful temporal information granules on the basis of which one develops any predictive model. For those in charge of everyday operation of a dispatching center, minutes and hours could form a viable scale of time granulation. Long-term planning is very much different from day-to-day operation. For the designer of high-speed integrated circuits and digital systems, the temporal information granules concern nanoseconds, microseconds, and perhaps microseconds. Granularity of information (in this case time) helps us focus on the most suitable level of detail.
Networked Control Systems for Manufacturing
Published in Richard Zurawski, Industrial Communication Technology Handbook, 2017
James R. Moyne, Dawn M. Tilbury, Dhananjay Anand
NCSs in current manufacturing plants generally operate at periodic intervals of 10 ms or larger [6]. In order to ensure the fidelity of sampled events, the authors in [85] suggest that clock synchronization between networked nodes be accurate within 1 ms. Clock synchronization accuracy required for a system is defined by either the maximum sampling rate within the system or the minimum time difference between monitored events. NCS sampling intervals are continually shrinking with tighter process control requirements. For example, the authors in [71,87] identify fault conditions such as electric arcing in the semiconductor fabrication process that occur over a time frame of 1–100 μs and state that these fast events may also have to be accurately reported and addressed by an NCS in the future [44]. In order to accurately record and respond to events occurring in the order of microseconds, clocks must presumably be synchronized to within a microsecond of accuracy.
Fuzzy Pattern Recognition
Published in Bogdan M. Wilamowski, J. David Irwin, Intelligent Systems, 2018
Information granules permeate numerous human endeavors [BP03a,BP03b,Z97]. No matter what problem is taken into consideration, we usually express it in a certain conceptual framework of basic entities, which we regard to be of relevance to the problem formulation and problem solving. This becomes a framework in which we formulate generic concepts adhering to some level of abstraction, carry out processing, and communicate the results to the external environment. Consider, for instance, image processing. In spite of the continuous progress in the area, a human being assumes a dominant and very much uncontested position when it comes to understanding and interpreting images. Surely, we do not focus our attention on individual pixels and process them as such but group them together into semantically meaningful constructs—familiar objects we deal with in everyday life. Such objects involve regions that consist of pixels or categories of pixels drawn together because of their proximity in the image, similar texture, color, etc. This remarkable and unchallenged ability of humans dwells on our effortless ability to construct information granules, manipulate them, and arrive at sound conclusions. As another example, consider a collection of time series. From our perspective we can describe them in a semi-qualitative manner by pointing at specific regions of such signals. Specialists can effortlessly interpret ECG signals. They distinguish some segments of such signals and interpret their combinations. Experts can interpret temporal readings of sensors and assess the status of the monitored system. Again, in all these situations, the individual samples of the signals are not the focal point of the analysis and the ensuing signal interpretation. We always granulate all phenomena (no matter if they are originally discrete or analog in their nature). Time is another important variable that is subjected to granulation. We use seconds, minutes, days, months, and years. Depending upon a specific problem we have in mind and who the user is, the size of information granules (time intervals) could vary quite dramatically. To the high level management time intervals of quarters of year or a few years could be meaningful temporal information granules on basis of which one develops any predictive model. For those in charge of everyday operation of a dispatching plant, minutes and hours could form a viable scale of time granulation. For the designer of high-speed integrated circuits and digital systems, the temporal information granules concern nanoseconds, microseconds, and perhaps microseconds. Even such commonly encountered and simple examples are convincing enough to lead us to ascertain that (a) information granules are the key components of knowledge representation and processing, (b) the level of granularity of information granules (their size, to be more descriptive) becomes crucial to the problem description and an overall strategy of problem solving, and (c) there is no universal level of granularity of information; the size of granules is problem-oriented and user dependent.
Hotspot auto-ignition induced detonation development: emphasis on energy density and chemical reactivity
Published in Combustion Theory and Modelling, 2022
Jiaying Pan, Lei Wang, Yu He, Haiqiao Wei, Gequn Shu, Tao Li
As the excitation time is at the microsecond, three orders lower than the ignition delay time at the millisecond. Meanwhile, the demarcation of the detonation development regime is much sensitive to the excitation time. Therefore, an adaptive time step algorithm is used to calculate the excitation time, with the temporal resolution reaching 110−12 s, which can achieve sufficient accuracy. Figure 7 further shows the uncertainties of excitation time of stoichiometric CH4/air mixture. It is observed that the detonation development regime in the detonation peninsula can match well with each other when using the above algorithm for excitation time calculation. Therefore, high-accuracy excitation time is necessary for constructing the detonation peninsula in the future.
A game-theoretic method for resilient control design in industrial multi-agent CPSs with Markovian and coupled dynamics
Published in International Journal of Control, 2021
Jiajun Shen, Xiangshen Ye, Dongqin Feng
In addition, the research approach of multi-agent control problems can be categorised according to centralised and decentralised information structure (LePape, 1990). In practical industrial multi-agent CPSs, bus and star topology are widely applied in most of fieldbus and industrial Ethernet technologies, such as Profibus, ProfiNet, EtherCAT, CANopen, Modbus, FF (Foundation Fieldbus), EPA (Ethernet for plant automation), etc. In addition, high real-time performance guarantees that each agent can collect the information/data/command messages from all the other agents, proceed/compute and make response to them promptly (in a quite short time period). Take EPA as an instance, due to its CSME (Communication Scheduling Management Entity) mechanism and PTP (Precise Time Protocol) realised in MAC layer, it can realise high-precision time synchronisation where skew is of nanosecond-level, and meanwhile the macro cycle time is of microsecond/millisecond-level (Lu et al., 2011; Thomesse, 2005). It indicates the reasonability and availability of applying CA setting and corresponding centralised information structure to our industrial multi-agent CPSs case.
Microchannel fabrication and metallurgical characterization on titanium by nanosecond fiber laser micromilling
Published in Materials and Manufacturing Processes, 2020
Laser micromachining system equipped with SPI make fiber laser source of 50W with nanosecond pulse waveform. The laser has pulse tune technology, which allows programmed waveform to deliver in CW and pulsed mode. Gaussian beam profile of laser beam helps to achieve high fluence with effective processing of metals and nonmetals. For high-quality machining, it is required to have laser pulses with shorter pulse duration. So the machining surface was found better in nanosecond as compared to milli- or microsecond pulses. It is equipped with suitable optics to focus laser beam on very small spot size for higher fluence and miniature product fabrication. The collimated beam expanded through beam expander and with focusing lens (F50), focused to small spot size of 7 µm. Laser cutting head has provision to supply assist gas coaxially through nozzle. There are three ports that can supply gas up to 15 bar at 120° angle. Dry compressed air is supplied from compressor and regulated by pressure control valve. Specification of laser source shown in Table 1.