Figure - uploaded by Blazhe Gjorgiev
Content may be subject to copyright.
Contexts in source publication
Context 1
... fault tree method is applied for each load demand node in the UK gas network, thus the failure probability of gas supply at each node is calculated and the obtained results are presented in Table 1. In all of the fault tree analyses, the maximum number of basic events in a minimal cut set is truncated at 10, while the minimum failure probability of minimal cut set is 1E-15. ...Context 2
... all of the fault tree analyses, the maximum number of basic events in a minimal cut set is truncated at 10, while the minimum failure probability of minimal cut set is 1E-15. The first column from Table 1 represents the nodes with gas demands, while the second column represents the average gas demand per node in í µí± 3 ℎ ⁄ . Node 41 is the node with the highest demand of gas, with average requirement of 1.77E+06 (í µí± 3 ℎ ⁄ ). ...Context 3
... total gas network failure probability is 3.00E-5 and is calculated with Equation 2 based on the individual demand node failure probability. Table 1 shows that in general the nodes with highest gas demands have the largest weighted probability of failure. ...Context 4
... curtailments are specified for each gas demand node and for the entire network. The first column from Table 1 represents the component/basic event name, such that the letters denote the component type (i.e. L stands for pipeline, B stands for compressor bypass, C stand for compressor, TER stands for gas terminal and TES stands for gas storage), while the numerical digits represents the nodes where the respective components are connected. ...Citations
... One of the reasons for the development of the method presented in this paper and utilised in the SecureGas project was to improve the accuracy of GTS modelling and estimates of the unsupplied gas volume in comparison with such MF-based methods used in some preceding projects [11]. Several methods trying to incorporate more physical aspects of GTS operation and utilise hydraulic steady-state simulations (SSSs) with probabilistic risk and resilience analysis have been put forward [17,18]. In comparison with such solutions, the developed method provides opportunities for a more detailed coordination of demand curtailment for different consumer types or better representation of pressure regulators (PRs) and compressor stations (CSs), automatically considering and implementing more of their limitations and control mechanisms. ...
... As a result, for such consumer nodes a rise of demand curtailment corresponds to an increasing equivalent value of the penalty coefficient. The option for such a detailed representation of individual demand nodes is rarely considered for analysis of GN disturbances or resilience [18], even if some level of consumer supply priorities are applied [17]. It must be stated that the use of all consumer subgroups is not mandatory, and the framework can be easily adapted by changing the input data. ...
... This exhaustible nature of UGSs is also often overlooked in other studies (i.e. [10,18], and [25]). Afterwards, a graph analysis block (GAB) is used to exclude the edges or nodes that are lost directly due to the disruption from the working GTS model, to determine if the remaining GN parts could supply any consumers, and to analyse the demand and supply situation of the remaining GN parts. ...
The current study presents a novel method for mitigating the consequences of disturbances in gas transmission systems (GTSs). The method presented was developed during the SecureGas project for identification of critical infrastructure and risk and resilience analysis. The method determines the optimal response strategies consisting of detailed coordination of demand curtailment for consumer nodes and the use of available reserve gas sources. Key characteristics of the method are an ability to adapt the modelling complexity for a particular task or a network part and several computational time economy measures making the method practical for large GTSs. The method can simultaneously consider variations of demand rates, consumer group compositions, source capacities, and average temperature and density of the gas in user-defined network parts during one or multiple concurrent disruptions in a GTS with different element restoration times. Preliminary analysis of a GTS model, which dynamically excludes network parts that are impossible to supply, implementation of a tailored genetic algorithm and supplemental sub-algorithms are used to save computation time while providing an optimal solution. A robust hydraulic steady-state solver is utilised for verification of solution feasibility in combination with dynamic changes to a GTS model implementing an adaptive network-wise control of multi-directional compressor stations and pressure regulators during different disturbance states of a GTS. A case study using a model of a real GTS demonstrated the effects of consumer categorisation and different modelling complexities, possible to consider with the developed method, on the estimated consequences of several types of disturbances and actions necessary to mitigate them by creating optimal response strategies.
This chapter summarizes the entire DLOC development described earlier in this book and presents recent work and emerging perspectives of disruptions handling and control. Five main areas of emerging applications are: (1) Evolutionary DLOC; (2) Collaborative analytics and intelligence; (3) Disruption handling by robots; (4) Quality assurance in supply networks; (5) Supply networks security. We also discuss future directions of extending the current research.
