Distributed fiber optic sensor placement
A procedure to solve the DFOS placement problem that uses a genetic algorithm to achieve a global optimization of sensor placement. First, our procedure according to aspects of the present disclosure defines a fitness function that counts the number of DFOS sensors used. Second, the procedure uses a valid DFOS placement assignment to model an individual in the genetic algorithm. Each individual consists of N genes, where N is the number of nodes in the given network infrastructure, e.g., N=|V|. Each gene has two genomes: (1) a list of 0s and/or 1s, in which is represent the network nodes that are equipped with DFOS sensors, and 0s represent the nodes that are not equipped with DFOS sensors; (2) a list of sensing fiber routes. An individual that has smallest number of is in their genes will be considered as the strongest individual. Thirdly, the procedure randomly generates a population of individuals. After a certain number of generations of population, the strongest individual in the last generation will be the global optima for the DFOS placement assignment.
1. A computer implemented method for determining distributed fiber optic sensing (DFOS) sensor placement in an optical network including a plurality of nodes interconnected via a plurality of optical links, the method comprising:
by the computer:
initialize an assignment list as empty and an uncovered list as containing all the optical links in graph G(V,E) where G is the graph, Vis a set of vertices, and E is a set of paired vertices;
if uncovered is not empty
1) Randomly select a node s from V to place a DFOS sensor fiber;
2) randomly select a neighbor node d of s that is within a sensing range limit R to place a DFOS sensor fiber terminator;
3) remove all optical links that are traveled through by a shortest path between s and d from uncovered;
2) Add s and the shortest path from s to d to assignment;
repeat 1-4 if uncovered is not empty, else return assignment; and
output an indicia of the assignment as a location in the optical network at which a DFOS sensor fiber is to be placed;
wherein the assignment list includes an identification of assignment of DFOS sensor fiber placement, sensing route, and sensing range of a sensor fiber; and
wherein the assignment list includes increased birefringence of the optical fiber is increased with respect to its average magnitude.
2. The computer implemented method of claim 1 FURTHER CHARACTERIZED BY a genetic algorithm for determining strongest individuals, wherein such strongest individuals each represent a location in the graph at which to place a DFOS sensor.
3. The computer implemented method of claim 2 FURTHER CHARACTERIZED BY determining strongest individuals over three generations, wherein the strongest individuals of the 3 rd generation represent locations in the graph at which to place a DFOS sensor.