Search

Over the years, probabilistic nature of renewable energy sources (RES) and its influence on power system adequacy have been well studied. However, rather less attention has been paid to the impact of RES unit itself’s and its power conversion system’s (PCS’) reliability, as well as their various connection topologies. This paper devises a comprehensive sensitivity study on how each of these elements can affect overall generating system reliability. given the plethora of RES configurations and components, it is of import to identify the most vulnerable element in RES. In this work, component importance is extended, for the first time, to generating capacity adequacy assessment (HLI). Measurement index is the centerpiece in reliability importance. New indices have to be introduced to facilitate the study. While the physical meaning of previously developed indices is lost, in this study indices are proposed based on traditional importance measures, of which the physical meaning are strictly retained and consistent with the definitions. With the proposed assessment technique, components in various RES configuration can be ranked according to their reliability importance. It is found in the numerical study that different importance measures (such as risk-achievement based measures and risk-reduction based measures) can result in different rankings. Studies on contributing factors of the reliability importance are also performed. As more and more RES gaining foothold in generating systems, the proposed technique assist to achieve targeted reliability level of the system, by easily identifying and prioritizing reliability improvement tasks among various units/components in the increasing complex system. Energy Market Authority (EMA) National Research Foundation (NRF) Published version This work was supported in part by the Energy Market Authority and National Research Foundation Singapore underGrant NRF2017EWT-EP003-038, and in part by the National Key Research and Development Program of China underGrant 2019YFB1505400.

Under the framework of DiffServ, Dovrolis presented the proportional differentiation model (PDM) [IEEE Network, October 1999] and found the waiting time priority (WTP) scheme [Proc. ACM SIGCOMM, September 1999] to be a suitable scheduling algorithm to achieve proportional delay differentiation. In this paper, we propose scaled time priority (STP), which is an efficient approximation to waiting time priority (WTP). In particular, STP is able to provide near proportional delay at a complexity of O(1), which is lower than WTP's O(N) complexity, where N is the number of service classes in the system. Simulation results show that STP is able to emulate the performance of WTP.

In this paper, we propose the use of a control theoretical approach in the design of fair queueing algorithms based on the scalable core (SCORE)/dynamic packet state (DPS) architecture. In particular, we make use of the linear quadratic control method from optimal control theory to design stable and robust systems that are simple to implement. Simulation results show that this approach gives improved flow isolation between transmission control protocol and user datagram protocol flows when compared with other SCORE/DPS fair queueing algorithms, such as core-stateless fair queueing.

We focus on efficiently achieving proportional delay differentiation (PDD), an instance of the proportional differentiation model (PDM) first proposed under the DiffServ framework (see Dovrolis, C. and Ramanathan, P., IEEE Network, Oct. 1999). Waiting time priority (WTP) has been found to be a suitable algorithm to achieve PDD. Using WTP (see Dovrolis et al., Proc. ACM SIGCOMM, 1999) as a reference, we show that our proposed scaled time priority (STP) algorithm is able to provide near proportional delay at a complexity of O(1), which is lower than the O(N) complexity of WTP, where N is the number of service classes in the system. Simulation results also show that STP is able to emulate the performance of WTP.

In this paper, we propose a generalised minimum queuing delay (GMQD) service discipline for high speed networks, mainly asynchronous transfer mode (ATM) networks. This proposed scheme is similar to service disciplines based on fair queuing, but instead of using only a single service rate for each session for its entire connection lifetime, multiple service rates are used. The service rate of any session at any point in time is computed efficiently based on the number of bits backlogged in the queues of the session and another imaginary reference session at that point in time. The main advantage of this scheme is that the queuing delays suffered by all the sessions connected to a single output node are minimised, leading to a smaller delay variation. In addition, this smaller delay variation also implies a smaller variance in the maximum queue length, thereby, reducing the possibility of buffer overflow.

The problem of scheduling on the job shop floor is widely encountered in the manufacturing industry. However, generating good job shop schedules in a reasonable time remains a challenging problem, because of its inherent computational complexity. Dynamic job shop scheduling (DJSS) has been proven to be an intractable analytical problem and is largely solved by combinatorial analysis in traditional research. The NP-complete nature of the problem forces many assumptions to be made, which limit the use of analytical methods in practical problem solving. In this paper, the authors propose a new scheme named minimum queuing delay (MQD) for scheduling in a dynamic job shop derived from fair queuing service disciplines for asynchronous transfer mode (ATM) networks. This scheme combines the job process-time and work-content of jobs in the current queues into the dispatching methodology. The main advantage of this scheme is that the queuing delays suffered by all the jobs at every machine are minimized, leading to a minimized maximum completion time (make-span). A smaller queuing delay variation is also achieved using this algorithm. Results from experiments conducted on a testbed with several embedded controllers are presented.