Method for coordination of competing processes or for control of the transport of mobile units within a network
First Claim
1. A method for coordination of competing processes or for control of a transport of mobile units within a network which has nodes and edges, with the network being subdivided into core areas n, each of which has an associated edge area so that the core area and the associated edge area each define a cohesive subnetwork of nodes and edges, with the edges having a limited buffer capacity for units to be received and with the nodes or edges being equipped with data acquisition elements, and with the nodes being equipped with a limited dispatch capability for units to be controlled and being equipped with control units for which a (switching) state of dispatch of mobile units, a (switching) state of non-dispatch and, between them, a changeover switching state are provided, wherein:
- a the network is controlled on a decentralized and self-organizing basis in control units of the node points or locally limited subnetworks, with the control units of adjacent node points or subnetworks being connected to one another for data interchange,b1 data from prediction models of local process procedures at the respective node, data from prediction models of local process procedures at adjacent nodes, or data from both prediction models of local process procedures at the respective node, and data from prediction models of local process procedures at adjacent nodes, orb2 data from data acquisition elements of the respective node or of the edges connected to it, data from data acquisition elements of adjacent nodes or of the edges connected to them, data from both data acquisition elements of the respective node or of the edges connected to it and data from data acquisition elements of adjacent nodes or of the edges connected to them, or a combination of data according to b1 and b2c are used for local simulation and optimization of switching sequences of the control unit in order to determine the performance of the nodes or subnetworks taking into account the buffer capacity of the edges on the basis of models for short-term predictions with fixed assumed switching states of adjacent nodes, withc1 a plurality of high-performance switching strategies being produced for the subnetworks by a combination of high-priority switching sequences for the relevant individual nodes, andc2 a test of the high-performance controllers being carried out in the subnetworks, and, subsequently,c3 that control strategy which has the best performance being selected in the respective core area of the subnetwork and being converted to corresponding switching states for the relevant nodes.
1 Assignment
0 Petitions
Accused Products
Abstract
A method for coordination of concurrent processes or for control of the transport of mobile units within a network, wherein a) the control of the network occurs in a decentralized and self-organizing manner in the controllers of node points or local defined sub-networks, whereby the control units of adjacent node points or sub-networks are connected to each other for data exchange of, b1) data from prediction models for local process sequences at each node and/or data from prediction models for the local process sequences of adjacent nodes and/or b2) data from data recording elements of each node or the boundaries associated therewith and/or data from data recording elements of adjacent nodes or the boundaries associated therewith, c) local simulation and optimization of switching the controller to establish the performance of the nodes or sub-networks with regard to the buffer capacity of the boundaries based on models for short-term predictions with fixed switch states for adjacent nodes, are applied.
31 Citations
24 Claims
-
1. A method for coordination of competing processes or for control of a transport of mobile units within a network which has nodes and edges, with the network being subdivided into core areas n, each of which has an associated edge area so that the core area and the associated edge area each define a cohesive subnetwork of nodes and edges, with the edges having a limited buffer capacity for units to be received and with the nodes or edges being equipped with data acquisition elements, and with the nodes being equipped with a limited dispatch capability for units to be controlled and being equipped with control units for which a (switching) state of dispatch of mobile units, a (switching) state of non-dispatch and, between them, a changeover switching state are provided, wherein:
-
a the network is controlled on a decentralized and self-organizing basis in control units of the node points or locally limited subnetworks, with the control units of adjacent node points or subnetworks being connected to one another for data interchange, b1 data from prediction models of local process procedures at the respective node, data from prediction models of local process procedures at adjacent nodes, or data from both prediction models of local process procedures at the respective node, and data from prediction models of local process procedures at adjacent nodes, or b2 data from data acquisition elements of the respective node or of the edges connected to it, data from data acquisition elements of adjacent nodes or of the edges connected to them, data from both data acquisition elements of the respective node or of the edges connected to it and data from data acquisition elements of adjacent nodes or of the edges connected to them, or a combination of data according to b1 and b2 c are used for local simulation and optimization of switching sequences of the control unit in order to determine the performance of the nodes or subnetworks taking into account the buffer capacity of the edges on the basis of models for short-term predictions with fixed assumed switching states of adjacent nodes, with c1 a plurality of high-performance switching strategies being produced for the subnetworks by a combination of high-priority switching sequences for the relevant individual nodes, and c2 a test of the high-performance controllers being carried out in the subnetworks, and, subsequently, c3 that control strategy which has the best performance being selected in the respective core area of the subnetwork and being converted to corresponding switching states for the relevant nodes. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24)
-
Specification