CN104836620A - Optical waveguide array-optical fiber array automatic butt-coupling parallel index optimization method - Google Patents

Abstract

The invention provides an optical waveguide array-optical fiber array automatic butt-coupling parallel index optimization method. The optical waveguide array-optical fiber array automatic butt-coupling parallel index optimization method comprises that mapping between an optical waveguide array-optical fiber array auto-aligning physical parameter and a parallel index optimization model is established, and a first objective function and a second objective function are obtained; a maximum evolutional generation is set, input single-mode fiber arrays and output single-mode fiber arrays enter specified spatial positions in a specified direction and attitude according to the regulation of each individual, optical power values of two or more than two waveguide channels are read, recorded and stored, and the Pareto sequence and the crowding distance are calculated; evaluation of each individual in one generation cluster is completed, a next generation cluster is generated by means of intersection, variation and lowliest place elimination, the evolutional generation is accumulated, whether the evolutional generation reaches the maximum evolutional generation or not is determined, if the evolutional generation reaches the maximum evolutional generation, the process is finished; if the evolutional generation does not reach the maximum evolutional generation, the intersection step is returned. According to the invention, the low-loss rapid butt-coupling automation degree and the working efficiency of an optical fiber array-waveguide device-optical fiber array system are effectively improved.

Description

The parallel index optimization method of fiber waveguide array-fiber array automatic butt coupling

Technical field

The invention belongs in optical communication field, the automatic butt coupling technology of fiber waveguide device and monomode fiber array and encapsulation technology field, relate to a kind of parallel index optimization method, particularly relate to the parallel index optimization method of a kind of fiber waveguide array-fiber array automatic butt coupling.

Background technology

In recent years, in the ultrahigh speed transmission experiment of optic communication link net and the fiber to the home technology of Access Network, have employed fiber waveguide device more and more.Lead-in light waveguide device in fiber optic network, must solve the link encapsulation of optical fiber and fiber waveguide.A key technology that wherein will solve is that fiber waveguide device docks with the low-loss of fiber array.International advanced index is that every damaged on end consumption is lower than 0.1dB.Reach this index, require on the one hand consistent as far as possible with optical fiber of the mode distributions of fiber waveguide, necessarily require the optical axis alignment precision controlling of fiber waveguide and optical fiber below 0.1 μm on the other hand.

Utilize high accuracy adjusting device, adopt manual operation to dock with the low-loss of fiber array to realize fiber waveguide device, technical requirement is very high, particularly after being coupled into 0.3dB, it is key problem in technology that micro various signal as operation criterion is accurately tested, corresponding microoperation is very difficult, and therefore efficiency is very low, repeatability is very poor.Adopt that automatic technology realizes high efficiency between fiber waveguide and optical fiber, low-loss Butt-coupling is a kind of effective means, conventional automatic core-adjusting method follows manually operated thinking, obtains coupling efficiency large as far as possible by the relative position of scanning fine setting waveguide optical axis and optical axis.

Conventional automatic core-adjusting method depends on the accurate test to micro various signal, not only requires that the guide rail of adjusting bracket has very high traveling precision, also requires that detector has very high sensitivity.Generally, due to precision reason and various random error, usually cause the repeatability of aligning bad.In recent years report a kind of new barycenter core regulating method for addressing this problem, the method is avoided directly finding peak, adopts the barycenter of the distribution of the coupling efficiency near survey calculation peak value to determine peak.Overcome conventional aligning process from principle and depend on the difficulty that micro various signal is accurately tested.Depend on and accurately test difficulty for solving conventional aligning process to micro various signal, the operation principle of existing automatic core-adjusting device is based upon on the feedback servo basis of light signal, shows very strong logicality.Must make to realize logical light between fiber-to-waveguide-optical fiber by initial adjustment, the feedback path between computer and power meter enter can running status time, automatic core-adjusting program can come into force.Light work is led in initial adjustment still needs manual operation, and this work is very high to the technical requirement of operating personnel, and rely on the qualification of operator, the time cannot lump together.In view of this, also been proposed a kind of automatic core-adjusting method and system adopting genetic algorithm afterwards, important improvement is that initial adjustment is led to light and achieved automation, significantly reduces the dependence of people.And this core regulating method and system do not limit by the strong logicality of feedback processing, allow the parallel work-flow adopting space respectively to tie up, but often must be individual containing tens of ~ hundreds of for population, mean that every hereditary generation at least must carry out tens of groups of parallel work-flows, the optimizing of fiber array and fiber waveguide array aligning restrains the heredity need going through nearly 100 generations, total parallel work-flow reaches thousands of groups, time-consumingly wants several tens minutes.In addition, what array docking automatic core-adjusting related to is a biobjective scheduling problem, has both required that each passage had high coupling efficiency, and has wished again have good uniformity between each passage coupling efficiency.Because discrete array device inevitably exists channel spacing error, during actual aligning, the implementation procedure of two targets is usually afoul mutually, can not be met simultaneously.In two indices, how light by about weighting what is heavy, but the reasonable set of weight proportion is still a difficult thing so far, and examination repeatedly generally can only be adopted to adjust the method revised gradually.Therefore, it is excessive that prior art can produce automatic core-adjusting operational ton, the problem of weight proportion setting difficulty.

Therefore, how the parallel index optimization method of a kind of fiber waveguide array-fiber array automatic butt coupling is provided, excessive to solve core regulating method automatic operation amount of the prior art, the many disadvantages such as weight proportion setting difficulty, have become practitioner in the art's technical problem urgently to be resolved hurrily in fact.

Summary of the invention

The shortcoming of prior art in view of the above, the object of the present invention is to provide the parallel index optimization method of a kind of fiber waveguide array-fiber array automatic butt coupling, excessive for solving core regulating method automatic operation amount in prior art, the problem of weight proportion setting difficulty.

For achieving the above object and other relevant objects, the invention provides the parallel index optimization method of a kind of fiber waveguide array-fiber array automatic butt coupling, comprise: set up the mapping between fiber waveguide array-fiber array automatic core-adjusting physical parameter and parallel index optimization model, individuality wherein during aligning in waveguide optical axis parallel index optimization model corresponding to certain relative space position between optical axis and direction, population in the parallel index optimization model of set correspondence in described relative space position and direction, chromosome in the parallel index optimization model of solid axes parameter correspondence, first object function in the parallel index optimization model of luminous power sum correspondence of two or more waveguide channels of aligning Procedure Acquisition, the second target function that the maximum of the luminous power of two or more waveguide channels of aligning Procedure Acquisition and the difference correspondence of minimum value walk abreast in index optimization model, adopt computer control, driving stepper motor mode adjusts input monomode fiber array and output monomode fiber array, make it respectively with the deviation of the alignment of fiber waveguide input port and output port in predetermined alignment deviation range, set maximum evolutionary generation, random generation initial population, input monomode fiber array and output monomode fiber array enter the locus of specifying by the regulation of each individuality by the direction attitude of specifying, and read, record, store the optical power value of two or more waveguide channels, this reading, record, storing process carries out successively by individuality order, the optical power value obtained is utilized to calculate the Pareto sum of ranks crowding distance of each individuality, the principle that individuality that in two identical individualities of the large individuality of Pareto ordinal number, Pareto ordinal number, crowding distance is large is better than the little individuality of crowding distance is better than according to the individuality that Pareto ordinal number is little, after completing the evaluation to individualities all in generation population, employing intersects, make a variation and the step of elimination of the last one produces population of future generation, the step iteration of described intersection, variation and elimination of the last one is carried out, accumulative evolutionary generation, judge whether evolution number of times reaches maximum evolutionary generation, if, end process, completes a waveguide-optical fibre automatic core regulating process, if not, the step continuing to perform described intersection is returned.

Alternatively, described predetermined alignment deviation range is-50 μm to+50 μm scopes; In described initial population, individual amount is more than 30 or 30; Described maximum evolutionary generation is more than or equal to 35.

Alternatively, described solid axes parameter is the reference axis parameter that space 11 is tieed up, and comprise the 5 dimension coordinate axle parameters of input side except rotating around optical axis and 6 dimension coordinate axle parameters of output side, this 11 dimension coordinate axle variable real number array is encoded.

Alternatively, i-th in the described parallel index optimization model 11 dimension coordinate axle variablees individual by any i-th described real number array coding are formed; Wherein, i is more than or equal to 1.

Alternatively, described first object function representation is f ₁(U _i), described second target function is expressed as f ₂(U _i); f ₁(U _i)=-[P ₁(U _i)+P ₂(U _i)]; f ₂(U _i)=| P ₁(U _i)-P ₂(U _i) |; Wherein, P ₁(U _i) and P ₂(U _i) be respectively aligning position and be in by i-th individual U _iduring the dimensional orientation of regulation, the Output optical power that recorded by two samplings.

Alternatively, the Pareto order of described individuality is calculated by the first pre-defined rule; Described first pre-defined rule is: for any i-th individual U _iif, with another jth individual U _{j ≠ i}there is a kind of target function value relation; Described target function value closes: f ₁(U _i) > f ₁(U _{j ≠ i}), and f ₂(U _i)>=f ₂(U _{j ≠ i}); Then i-th individual U _ipareto ordinal number increase by 1, otherwise i-th individual U _ipareto ordinal number increase by 0; By with i-th individual U _ioutside all individualities compare one by one, accumulative obtain i-th individual U _ithe value of Pareto ordinal number.

Alternatively, i-th individual U _icrowding distance calculated by the second pre-defined rule; For having identical Pareto ordinal number in population, k the individual U with identical Pareto ordinal number _iaccording to f ₁(U _m) < f ₁(U _m+1), wherein, m=1,2 ..., k-1; f ₂(U _n) < f ₂(U _n+1), wherein, n=1,2 ..., k-1; Order respectively with cambic m subscript and n subscript arrangement; Calculate and described first object function f ₁(U _m) relevant crowding distance d ₁(U _m), calculate and described second target function f ₂(U _n) crowding distance the arrangement of m subscript and n subscript are arranged correspondence respectively and be returned to original i subscript arrangement, each individual U _iobtain individual U _ithe first crowding distance d ₁(U _i) and the second crowding distance d ₂(U _i), i-th individual U _icrowding distance d (U _i)=d ₁(U _i)+d ₂(U _i).

Alternatively, the step of described elimination of the last one comprises the 3rd pre-defined rule, the individuality that described 3rd pre-defined rule refers in the groups of individuals that rejecting Pareto ordinal number is maximum, crowding distance is minimum.

Alternatively, the step of described intersection refers to and is better than according to the individuality that Pareto ordinal number is little the principle that individuality that in two identical individualities of the large individuality of Pareto ordinal number, Pareto ordinal number, crowding distance is large is better than the little individuality of crowding, from size be select t the row of evaluation the population of M before individuality, newly individual wherein, U _ibe i-th individuality, a _ifor weight coefficient, a _imeet new individual U _oform with former M individuality the transition population that a size is M+1, calculate according to described first pre-defined rule and the second pre-defined rule the individual U that makes new advances _opareto order and crowding distance, eliminate last position according to described 3rd pre-defined rule individual, obtain the population of future generation that body quantity is one by one M.

Alternatively, be random selecting body one by one in the population of future generation of M at described individual amount, made a variation by essence, increase a new individuality near parent individuality, again construct the transition population that body quantity is one by one M+1; Calculate Pareto order and the crowding distance of all individualities in described transition population according to the first pre-defined rule and the second pre-defined rule, continue to eliminate last position individuality according to described 3rd pre-defined rule, obtain the new population of future generation that body quantity is one by one M.

As mentioned above, the parallel index optimization method of fiber waveguide array of the present invention-fiber array automatic butt coupling, has following beneficial effect:

The parallel index optimization method of 1, fiber waveguide array of the present invention-fiber array automatic butt coupling, improves the deficiency of automatic core-adjusting method in prior art from mechanism.

The remarkable result of the parallel index optimization method of 2, fiber waveguide array of the present invention-fiber array automatic butt coupling to make automatic core-adjusting device number of operations significantly reduce when being applied in automatic core-adjusting device, and without the need to artificially participating in the setting of target weight, display improves operability, effectively improves the automaticity and operating efficiency that realize fiber array-waveguide device-Array System low-loss rapid abutting joint coupling.

Accompanying drawing explanation

Fig. 1 is shown as the analytic modell analytical model schematic diagram of the end coupling of monomode fiber and single mode Luciola substriata.

Fig. 2 is shown as 1 × 8 Waveguide branching coupler and fiber array Butt-coupling structural representation.

Fig. 3 is shown as the parallel index optimization method flow schematic diagram of fiber waveguide of the present invention array-fiber array automatic butt coupling.

Fig. 4 is shown as multi-objective evolutionary algorithm aligning numerical simulation progress curve schematic diagram.

Fig. 5 is shown as the theory structure schematic diagram of the index parallel optimization automatic core-adjusting device of the parallel index optimization method application of fiber waveguide of the present invention array-fiber array automatic butt coupling.

Element numbers explanation

1 fiber end face

2 Waveguide end faces

31 × 8 Waveguide branching couplers and fiber array Butt-coupling structure

31 monomode fibers

32 1 × 8 Waveguide branching couplers

33 monomode fiber arrays

4 index parallel optimization automatic core-adjusting devices

41 lasing light emitters

42 monitors

43 light power meters

44 sextuple adjusting brackets

45 fiber waveguide supports

46 antidetonation platforms

47 first drivers

48 computers

S1 ~ S3 step

Embodiment

Below by way of specific instantiation, embodiments of the present invention are described, those skilled in the art the content disclosed by this specification can understand other advantages of the present invention and effect easily.The present invention can also be implemented or be applied by embodiments different in addition, and the every details in this specification also can based on different viewpoints and application, carries out various modification or change not deviating under spirit of the present invention.

Refer to accompanying drawing.It should be noted that, the diagram provided in the present embodiment only illustrates basic conception of the present invention in a schematic way, then only the assembly relevant with the present invention is shown in graphic but not component count, shape and size when implementing according to reality is drawn, it is actual when implementing, and the kenel of each assembly, quantity and ratio can be a kind of change arbitrarily, and its assembly layout kenel also may be more complicated.

Below in conjunction with embodiment and accompanying drawing, the present invention is described in detail.

The parallel index optimization method of fiber waveguide array of the present invention-fiber array automatic butt coupling, the general principle of this optimization method is:

What usually run in optical communication system is docking between single mode waveguide array and monomode fiber array, at docking end face, realizes carrier transmission by the coupling power between eigen mode.Be filled with between waveguide and optical fiber and absorb the little and binding agent to be solidified of index matching to operation wavelength, end face reflection is less than-50dB, can ignore, and single core end coupling efficiency theory shows as the overlap integral between the guided mode field distribution known.Without loss of generality, refer to Fig. 1, be shown as the analytic modell analytical model schematic diagram of the end coupling of monomode fiber and single mode Luciola substriata, as shown in Figure 1, if the optical axis of Luciola substriata in the z-direction, have an angular deviation θ between optical axis and Luciola substriata optical axis, the two optical axis hearts are perpendicular to z-axis direction there being deviation delta x and Δ y, and fiber end face 1 and Waveguide end face 2 spacing are along the z-axis direction Δ z.Consider that in fiber waveguide manufacturing process, some unpredictable factors cause the asymmetric of field distribution of waveguide, use W _xoand W _yorepresent single mode waveguide guided mode girdling the waist in x and y orientation respectively, girdling the waist as W of optical fiber mode _fo.Single core end coupling efficiency eta can be expressed as:

η=η _xη _yformula (1)

Wherein,

${\mathrm{\&eta;}}_x=\frac{2\mathrm{ep}[-\frac{k^2(W_{\mathrm{fo}}^2+W_{\mathrm{xo}}^2){\mathrm{\&theta;}}^2}{2{\mathrm{\&eta;}}_{\mathrm{xo}}\left(\mathrm{\&Delta;z}\right)}-\frac{2}{{\mathrm{\&eta;}}_{\mathrm{xo}}\left(\mathrm{\&Delta;z}\right)}(\frac{{\mathrm{\&Delta;x}}^2}{W_{\mathrm{fo}}^2}+\frac{{(\mathrm{\&Delta;z\&theta;}-\mathrm{\&Delta;x})}^2}{W_{\mathrm{xo}}^2})]}{\sqrt{{\mathrm{\&eta;}}_{\mathrm{xo}}\left(\mathrm{\&Delta;z}\right)}}$ Formula (2)

${\mathrm{\&eta;}}_y=\frac{2\exp [-\frac{{2\mathrm{\&Delta;y}}^2}{{\mathrm{\&eta;}}_{\mathrm{yo}}\left(\mathrm{\&Delta;z}\right)}(\frac{1}{W_{\mathrm{fo}}^2}+\frac{1}{W_{\mathrm{yo}}^2})]}{\sqrt{{\mathrm{\&eta;}}_{\mathrm{yo}}\left(\mathrm{\&Delta;z}\right)}}$ Formula (3)

Formula (2) and the middle η of formula (3) _xo(Δ z) and η _yo(Δ z) represents respectively by formula (4) and formula (5).

${\mathrm{\&eta;}}_{\mathrm{xo}}\left(\mathrm{\&Delta;z}\right)={(\frac{W_{\mathrm{xo}}}{W_{\mathrm{fo}}}+\frac{W_{\mathrm{fo}}}{W_{\mathrm{xo}}})}^2+{\left(\frac{2\mathrm{\&Delta;z}}{k{W}_{\mathrm{fo}}{W}_{\mathrm{xo}}}\right)}^2$ Formula (4)

${\mathrm{\&eta;}}_{\mathrm{yo}}\left(\mathrm{\&Delta;z}\right)={(\frac{W_{\mathrm{yo}}}{W_{\mathrm{fo}}}+\frac{W_{\mathrm{fo}}}{W_{\mathrm{yo}}})}^2+{\left(\frac{2\mathrm{\&Delta;z}}{k{W}_{\mathrm{fo}}{W}_{\mathrm{yo}}}\right)}^2$ Formula (5)

Wherein, k is the wave number in the medium of light wave between waveguide and optical fiber, relevant with the refractive index of medium, reflects the dielectric impact of binding agent.Coupling efficiency η increases with the reduction of θ, Δ x, Δ y and Δ z, when θ, Δ x, Δ y and Δ z are zero, has the maximum coupling efficiency η of single core end face _max:

${\mathrm{\&eta;}}_{\max }=\frac{4}{(\frac{W_{\mathrm{fo}}}{W_{\mathrm{xo}}}+\frac{W_{\mathrm{xo}}}{W_{\mathrm{fo}}})(\frac{W_{\mathrm{fo}}}{W_{\mathrm{yo}}}+\frac{W_{\mathrm{yo}}}{W_{\mathrm{fo}}})}$ Formula (6)

If now waveguide is identical with the mode distributions of optical fiber, formula (3) provides the end coupling efficiency of 100% in theory.Although actual conditions are not so desirable, but theory analysis points out a principle, for given fiber waveguide and optical fiber, obtain high end coupling efficiency, both optical axis alignment must be made as far as possible and keeping parallelism, both end surfaces should be tried one's best little and fill suitable index-coupling medium.

It is the popularization that above-mentioned single core is coupled that multichannel array is coupling in principle, due to the Waveguide array of reality and fiber array not the same in channel pitch, can not accomplish that all passages obtain the highest coupling efficiency all simultaneously, practical compromise adopts two indices to carry out balance evaluation to aligning result, one is whether the insertion loss sum of all passages is enough little, another is whether the difference of insertion loss maxima and minima in all passages is enough little, and the latter is used to define uniformity.Because the loss of discrete device self is determined, when entering aligning docking flow process, affect the factor mainly end coupling loss of insertion loss, the end coupling loss decibel value of end coupling efficiency represents.

During array aligning can the output of all passages of Real-Time Monitoring certainly best, but because the development of waveguide channels number with device technology is on the increase and is subject to the restriction of package system cost, aligning system is not generally equipped with a large amount of light power meters, real current situation adopts as shown in Figure 2,1 × 8 Waveguide branching coupler and fiber array Butt-coupling structure 3, wherein this structure comprises monomode fiber 31,1 × 8 Waveguide branching coupler 32, monomode fiber array 33.Get the output of the 1st and the 8th paths (can also be other two paths separated by a distance) as sampled signal, the Butt-coupling of other passage leans on the precision of each passage geometric position of discrete device to ensure.For fiber waveguide array and the fiber array being less than 0.5 μm of positional precision of 0.1 μm of positional precision, twin-core core regulating method is adopted not have obstacle in practical.

End face due to multicore array has larger length-width ratio, and the fine setting of end surfaces must be associated with faces parallel, requires between end face almost parallel, otherwise can not carry out because end collision in limit causes interval to be finely tuned.Therefore the optical axis of commercial waveguide chip or fiber array and the perpendicularity deviation of its end face are defined within 0.2 °, to ensure that the depth of parallelism between waveguide optical axis and optical axis can be controlled by the depth of parallelism adjusted between two array end faces.Between the end face of exit end, the adjustment of the depth of parallelism and end distance relates to θ _x, θ _ythree-dimensional with z, the docking of array twin-core relates to x, y and θ _zthree-dimensional.In incidence end, due to the circular symmetry of single fiber, θ _zadjustment lose meaning.Therefore the space dimensionality that twin-core automatic core-adjusting relates to is 11 dimensions, space five dimension being in waveguide chip input side, adjustment input optical fibre core position respectively, and be in waveguide chip output side, the six-dimensional space of positioning optical waveguides core position of adjustment fiber array.

To be solved by this invention what be that array docking automatic core-adjusting relates to is the problem of a Two indices parallel optimization, has both required that each passage had high coupling efficiency, and has wished again have good uniformity between each passage coupling efficiency.The index parallel optimization technique that the present invention relates to is a kind of Noninferior Solution Set algorithm based on stable replacement policy, multi-target evolution, adopt and distribute Pareto order without dominated Sorting rule to each individuality in population, after individuality presses the classification of Pareto ordinal number, its crowding distance is calculated to all individualities with identical ordinal number, the individuality of low Pareto ordinal number means the composite request of the more multiple target of convergence, and the individuality of large crowding distance means richer diversity.Compared with genetic algorithm, the present invention does not adopt the single fitness function set up and comprise multiple weighted target to evaluate the individuality in population, but determine to eliminate order by the size of individual Pareto ordinal number and crowding distance thereof, little and crowding distance the is large individuality of Pareto order is endowed more multiplies chance.

In order to improve convergence rate, searching process combine many parents intersect and Ants climbing hill two heuristic.First a new offspring individuals is produced with many parents cross method to M parental generation is individual, form the transitional population of one M+1 individuality, after eliminating a last position individuality by the Pareto ordinal number of individuality and the size of crowding distance thereof, a new offspring individuals is produced again with real-valued variation method, eliminate a last position individuality by identical criterion, obtain thus by M the individual population of future generation formed.Obviously, when method of the present invention is put to practicality, evolve to every generation and only need do two individual machine operations, operational ton comparatively genetic algorithm significantly reduces, and can significantly improve the efficiency of automatic core-adjusting.

Import index parallel optimization technique to need to solve the genetic model of index parallel optimization technique and the mapping of actual aligning process, certain space relative bearing between waveguide optical axis and optical axis when describing actual aligning with the individual concept in index parallel optimization model; Population concept reflects the set of considered relative bearing of having living space.The benefit of method of the present invention allows to adopt real coding scheme, contradiction between the precision existed when not only avoid binary processing continuous function and search volume, and computational speed improves greatly, therefore represent chromosome by the real number array containing multiple decision variable, be equivalent to the coordinate of physical space.Aligning process expects that the two indices reached is respectively as two target function values.Through Conceptual Projection like this, the searching process of array docking automatic core-adjusting to coupling efficiency and uniformity is converted into the process that index parallel optimization technique is optimized two target function values.

For the Butt-coupling structure 1 of 1 × 8 Waveguide branching coupler shown in Fig. 2 and fiber array, the space dimensionality that twin-core automatic core-adjusting relates to is 11 dimensions.Usually, under micro-supervisory work, initial deviation between waveguide optical axis and optical axis easily controls within 50 μm, the spacing of Waveguide end face and fiber end face easily controls within 4 μm, determine that the excursion of x-axis and y-axis parameter is [-50 μm thus, 50 μm], the excursion of z-axis parameter is [0 μm, 4 μm].Get θ _zthe hunting zone at angle is ± 5 °.To occur up and down in order to avoid waveguide chip and fiber array or the limit end of left and right collides, θ _xangle and θ _ythe hunting zone at angle is determined by formula (7):

$-\frac{z_{\max }}{h}\&le;\sin {\mathrm{\&theta;}}_x\&le;\frac{z_{\max }}{h},-\frac{z_{\max }}{w}\&le;\sin {\mathrm{\&theta;}}_y\&le;\frac{z_{\max }}{w}$ Formula (7)

H in formula (7), w, and z _maxrepresent the thickness of fiber array, width respectively, based on the largest interval between fiber array both ends of the surface.

Embodiment

This enforcement provides the parallel index optimization method of a kind of fiber waveguide array-fiber array automatic butt coupling, refers to Fig. 3, is shown as the parallel index optimization method flow schematic diagram of fiber waveguide array-fiber array automatic butt coupling.As shown in Figure 3, the parallel index optimization method of described fiber waveguide array-fiber array automatic butt coupling comprises the following steps:

S1, set up the mapping between fiber waveguide array-fiber array automatic core-adjusting physical parameter and parallel index optimization model, individuality wherein during aligning in waveguide optical axis parallel index optimization model corresponding to certain relative space position between optical axis and direction, population in the parallel index optimization model of set correspondence in described relative space position and direction, chromosome in the parallel index optimization model of solid axes parameter correspondence, first object function in the parallel index optimization model of luminous power sum correspondence of two or more waveguide channels of aligning Procedure Acquisition, the second target function that the maximum of the luminous power of two or more waveguide channels of aligning Procedure Acquisition and the difference correspondence of minimum value walk abreast in index optimization model.Described solid axes parameter is the reference axis parameter that space 11 is tieed up, and comprise the 5 dimension coordinate axle parameters of input side except rotating around optical axis and 6 dimension coordinate axle parameters of output side, this 11 dimension coordinate axle variable real number array is encoded.Wherein, i-th in the described parallel index optimization model 11 dimension coordinate axle variablees individual by any i-th described real number array coding are formed; Wherein, i is more than or equal to 1.

Above-mentioned space 11 dimension coordinate axle variable corresponds to 11 decision variables of evolution algorithmic, and the decision variable of one group of 11 dimension forms body U one by one _i, form population by setting up M group of individuals.Each individual U _iin aligning process, the corresponding coupling efficiency η of corresponding generation, is converted into above-mentioned two target function values and evaluates by coupling efficiency η.Be the luminous power of sampled signal due to what obtain during practical operation, it and insertion loss are inversely proportional to, then uniformity directly represents by the difference of two sampled signal power.Two target functions are respectively first object function f ₁(U _i) and the second target function f ₂(U _i).

F ₁(U _i)=-[P ₁(U _i)+P ₂(U _i)] formula (8)

F ₂(U _i)=| P ₁(U _i)-P ₂(U _i) | formula (9)

Wherein, P ₁(U _i) and P ₂(U _i) be respectively aligning position and be in by i-th individual U _iduring the dimensional orientation of regulation, the Output optical power that recorded by two samplings.The ratio of they and input optical power is insertion loss when adopting decibel to represent.Obvious P ₁and P ₂larger, first object function f ₁less; P ₁with P ₂difference less, the second target function f ₂less.

S2, adopt computer control, driving stepper motor mode adjusts input monomode fiber array and output monomode fiber array, make it respectively with the deviation of the alignment of fiber waveguide input port and output port in predetermined alignment deviation range, set maximum evolutionary generation, random generation initial population, input monomode fiber array and output monomode fiber array enter the locus of specifying by the regulation of each individuality by the direction attitude of specifying, and read, record, store the optical power value of two or more waveguide channels, this reading, record, storing process carries out successively by individuality order, the optical power value obtained is utilized to calculate the Pareto sum of ranks crowding distance of each individuality.The Pareto order of described individuality is calculated by the first pre-defined rule; Described first pre-defined rule is: for any i-th individual U _iif, with another jth individual U _{j ≠ i}there is a kind of target function value relation; Described target function value closes:

F ₁(U _i) > f ₁(U _{j ≠ i}), and f ₂(U _i)>=f ₂(U _{j ≠ i}) formula (10)

Then i-th individual U _ipareto ordinal number increase by 1, otherwise i-th individual U _ipareto ordinal number increase by 0; By with i-th individual U _ioutside all individualities compare one by one, accumulative obtain i-th individual U _ipareto ordinal number.Investigation process like this, throughout all individualities, completes the classification of individuality by Pareto ordinal number of population.Obviously, the criterion of formula (10) and ordinal number allocation rule, namely the first pre-defined rule ensure that the individuality of low Pareto ordinal number has more the tendency simultaneously optimized for Bi-objective.

To the individuality in population with identical ordinal number, calculate crowding distance according to the second pre-defined rule: the k under identical ordinal number individual U _iobtain by described second pre-defined rule.Described second pre-defined rule refers to:

For having identical Pareto ordinal number in population, k the individual U with identical Pareto ordinal number _iaccording to

F ₁(U _m) < f ₁(U _m+1), wherein, m=1,2 ..., k-1;

F ₂(U _n) < f ₂(U _n+1), wherein, n=1,2 ..., k-1; The order of formula (11) is respectively with cambic m subscript and the arrangement of n subscript.

Calculate and described first object function f ₁(U _m) relevant crowding distance d ₁(U _m),

formula (12)

Calculate and described second target function f ₂(U _n) crowding distance d ₂(U _n),

formula (13)

The arrangement of m subscript and n subscript are arranged correspondence respectively and be returned to original i subscript arrangement, each individual U _iobtain individual U _ithe first crowding distance d ₁(U _i) and the second crowding distance d ₂(U _i), i-th individual U _icrowding distance d (U _i)=d ₁(U _i)+d ₂(U _i).

Process for transition subscript m: be provided with three individual U simply _i: U _i=1, U _i=2and U _i=3, calculate f respectively ₁(U _i=1), f ₁(U _i=2) and f ₁(U _i=3), found that such relation: f ₁(U _i=1) < f ₁(U _i=3) < f ₁(U _i=2).So we change subscript transiently, according to the size order that result of calculation provides, arrangement is set as: U _m=1=U _i=1, U _m=2=U _i=3and U _m=3=U _i=2, and program can remember the one-to-one relationship of former subscript i and transition subscript m.Then d is calculated by above-mentioned rule ₁(U _m=1), d ₁(U _m=2) and d ₁(U _m=3).Then according to the corresponding relation of former subscript i and transition subscript m, obtain at once: d ₁(U _i=1)=d ₁(U _m=1), d ₁(U _i=2)=d ₁(U _m=3) and d ₁(U _i=3)=d ₁(U _m=2).Process about transition subscript n is identical therewith, and the second target function only related to is f ₂.

All individualities in above-mentioned procedure ergodic population.Obviously, crowding distance here refer in the two-dimensional space determined by two target dimensions, in population certain investigated individuality and its estimation with the space length between ordinal number adjacent body.The individuality that crowding distance is large, its two target function values are large with the difference with two target function values of ordinal number adjacent body, therefore large to the diversity contribution with ordinal number individuality in population.Physically, diversity contributes to the local convergence that aligning process avoids being absorbed in suboptimum.In a word, for any two individualities in population, the individuality that ordinal number is little is better than the large individuality of ordinal number; For two individualities that ordinal number is identical, the individuality that crowding distance is large is better than the little individuality of crowding distance.So far the evaluation to individualities all in generation population is completed by Pareto sum of ranks crowding distance two parameters.

In the present embodiment, described predetermined alignment deviation range is-50 μm to+50 μm scopes; In described initial population, individual amount is more than 30 or 30.

S3, the principle that individuality that in two identical individualities of the large individuality of Pareto ordinal number, Pareto ordinal number, crowding distance is large is better than the little individuality of crowding distance is better than according to the individuality that Pareto ordinal number is little, after completing the evaluation to individualities all in generation population, employing intersects, make a variation and the step of elimination of the last one produces population of future generation, the step iteration of described intersection, variation and elimination of the last one is carried out, complete and once evolve, accumulative evolutionary generation; Judge whether evolution number of times reaches maximum evolutionary generation, if so, terminates process, complete a waveguide-optical fibre automatic core regulating process; If not, the step continuing to perform described intersection is returned.Wherein, described maximum evolutionary generation is more than or equal to 35.

In this step, the step of described elimination of the last one comprises the 3rd pre-defined rule, the individuality that described 3rd pre-defined rule refers in the groups of individuals that rejecting Pareto ordinal number is maximum, crowding distance is minimum.

The step of described intersection refers to and is better than according to the individuality that Pareto ordinal number is little the principle that individuality that in two identical individualities of the large individuality of Pareto ordinal number, Pareto ordinal number, crowding distance is large is better than the little individuality of crowding, from size be select t the row of evaluation the population of M before individuality, newly individual

$U_0=\underset{i=1}{\overset{t}{\mathrm{\&Sigma;}}}{a}_i{U}_i$ Formula (14)

Wherein, U _ibe i-th individuality, a _ifor weight coefficient, a _imeet:

$\underset{i=1}{\overset{t}{\mathrm{\&Sigma;}}}{a}_i=1,-0.5\&le;a_i\&le;1.5$ Formula (15)

New individual U _oform with former M individuality the transition population that a size is M+1, calculate according to described first pre-defined rule and the second pre-defined rule the individual U that makes new advances _opareto order and crowding distance, eliminate last position according to described 3rd pre-defined rule individual, obtain the population of future generation that body quantity is one by one M.

It is random selecting body one by one in the population of future generation of M at described individual amount, made a variation by essence, namely adopt real number value to change real number array coding individual in step S1, increase a new individuality near parent individuality, again construct the transition population that body quantity is one by one M+1; Calculate Pareto order and the crowding distance of all individualities in described transition population according to the first pre-defined rule and the second pre-defined rule, continue to eliminate last position individuality according to described 3rd pre-defined rule, obtain the new population of future generation that body quantity is one by one M.Adopt Ants climbing hill so, by generations of intersection, variation, stablize genetic manipulations such as eliminating, constantly evolve, until converge on Best Coupling position.

In this enforcement, 1 × 8 Waveguide branching coupler provided using Fig. 2 and the Butt-coupling structure 1 of fiber array are as simulation model, sampling optical fiber is the 1st and the 8th passage, in order to outstanding end face coupling efficiency, without loss of generality, ignore the loss of 1 × 8 Waveguide branching coupler, bending loss and Y branching loss, this model reaches 100% ideally in end coupling efficiency, and the insertion loss of each passage of Waveguide branching coupler equals 1 point of 8 principle loss of 9.031dB.Get the W with a tight waist of Standard single-mode fiber under 1550nm wavelength _fo=5.63 μm, mould field asymmetric factor α=1-(W of quartz single mode Luciola substriata _xo/ W _yo) general in 0.1% magnitude, the W with a tight waist of the quartz single mode Luciola substriata of refractive index contrast Δ=0.3% is got here by 0.4% _yo=5.63 μm, W _xo=5.61 μm, input optical power is 0dBm.The method of initial population completely random produces, and Population Size is 30.

In order to hold the convergence property of index parallel optimization aligning process, first consider that the core interval of fiber array and Waveguide array is the ideal situation of 250 μm, the spacing between the 1st and the 8th sampling optical fiber is 1250 μm.The medium of filling in the gap between waveguide both ends of the surface and I/O fiber array end face is with reference to the parameter of BP300 ultraviolet binding agent, and the refractive index on 1550nm wavelength is 1.444, is less than 0.01 with the refringence of quartz, and transmitance is 98%/mm.Done and repeated emulation for 10 times, hereditary 100 generations at every turn, initial population adopts random device to generate.Fiber array and the Waveguide array twin-core Butt-coupling simulation result of desired pitch refer to table 1.

Table 1: the fiber array of desired pitch and Waveguide array twin-core Butt-coupling simulation result

Two core insertion loss mean values are 9.042dB, and the loss of deduction principle, end coupling loss mean value is about 0.011dB.Because the channel pitch setting optical fiber and waveguide is equal, homogeneity error is 0dB.Refer to Fig. 4, be shown as multi-objective evolutionary algorithm aligning numerical simulation progress curve schematic diagram, in simulation process, premature convergence is very fast as can be seen from the figure, and approximately pass through the evolution of 35 generations and can enter the region being less than 9.1dB, search efficiency and convergence are all fine.

In the present embodiment, select the automatic core-adjusting of single-core fiber-Luciola substriata-single-core fiber system, investigate the test of end coupling loss, the problem that waveguide own loss brings can be got rid of, be convenient to reflect the performance of aligning system in hardware and software two more objectively.Owing to being the coupling of single core, uniformity index is desensitized.Fiber waveguide selects silica based to bury type single mode Luciola substriata, and waveguide core Cross section Design is of a size of 8 × 8 μm ², refractive index contrast is 0.3%, both ends of the surface plane lapping.Optical fiber is plane lapping monomode fiber.Be equipped with the index-matching fluid of trace between clean dry fiber end face and Waveguide end face, effectively suppress end face reflection.Initial adjustment is carried out under the micro-supervision of x, y bidimensional of orthogonal, and adopt mouse action, 0.1 μm of stepper drive, easily control within the scope of ± 50 μm by the core deviation of the alignment of optical fiber and fiber waveguide, the spacing of Waveguide end face and fiber end face controls within 4 μm.Now start automatic core-adjusting, two ends aligning is parallel carries out, and terminates rear record power output, completes that aligning is time-consuming is not more than 2min.After aligning, output monomode fiber changes multimode fiber into, repeats above-mentioned aligning process.Core diameter due to multimode fiber reaches 50 μm, can be similar to and think that multimode fiber all have received the power output of single mode Luciola substriata.The decibel difference of the measured value of multimode fiber and monomode fiber is the end coupling loss of single mode Luciola substriata and monomode fiber.Automatic core-adjusting experiment number is 7 times, the results are shown in table 4.Coupling loss mean value on 1542.4nm wavelength is 0.1136dB, and the maximum of single coupling loss is less than 0.13dB, and standard deviation is less than 0.02dB.Show that this system is highly effective realizing in high efficiency, low-loss fiber-to-waveguide automatic butt.

And the parallel index optimization method of the fiber waveguide array described in the present embodiment-fiber array automatic butt coupling is realized by index parallel optimization automatic core-adjusting device 2.Refer to Fig. 5, be shown as the theory structure schematic diagram of the index parallel optimization automatic core-adjusting device of the parallel index optimization method application of fiber waveguide array-fiber array automatic butt coupling.Described index parallel optimization automatic core-adjusting device 4 comprises lasing light emitter 41, monitor 42, light power meter 43, sextuple adjusting bracket 44, fiber waveguide support 45, antidetonation platform 46, driver 47 and computer 48.Chip of light waveguide is fixed on middle fiber waveguide support 45, and this fiber waveguide support 45 manually can adjust in x, y, z three dimensional realization.Each side be equipped with a sextuple adjusting bracket 44 of precision at fiber waveguide support 45, be respectively used to fixing and adjustment input optical fibre array and output optical fibre array dimensional orientation.The three-dimensional shifting axle of x, y, z of accurate sextuple adjusting bracket and the θ around shifting axle rotation _x, θ _y, θ _zthree dimensional rotation axle is by driving stepper motor.The minimum amount of movement of the three-dimensional shifting axle of x, y, z is 0.05 μm, θ _xthe minimum rotation amount of rotation axis is 0.0030 degree, θ _ythe minimum rotation amount of rotation axis is 0.0032 degree, θ _zthe minimum rotation amount of rotation axis is 0.0025 degree.Two drivers 47 are for driving the stepping motor of two sextuple adjusting brackets 44 of precision, and driver is connected with computer by gpib interface, and available software programming controls.Monitor 42 by two cover along x, y orthogonal arrange CCD microscope and micropositioning stage, display and illumination cold light source form, microexamination during for just adjusting.Above-mentioned fiber waveguide support 45, two sextuple adjusting brackets 44 of precision and monitor 42 are placed on antidetonation platform 47.Lasing light emitter 42 in order to provide stable work light wave, the output interface of lasing light emitter 42 can with fiber alignment.Light power meter 43, for detecting the Output optical power of output optical fibre, is furnished with two input interfaces, can test the power output coming from two optical fiber simultaneously.Light power meter is equipped with gpib interface, can be connected realize communicating with computer.Command system is taken on by computer, controls aligning operation, accept light power meter in real time and detect the optical power signals obtained according to software programming, completes differentiation, memory and feedback operation in real time.Systems soft ware VB language is worked out.

The parallel index optimization method of fiber waveguide array of the present invention-fiber array automatic butt coupling, improves the deficiency of automatic core-adjusting method in prior art from mechanism.The remarkable result of the parallel index optimization method of fiber waveguide array of the present invention-fiber array automatic butt coupling to make automatic core-adjusting device number of operations significantly reduce when being applied in automatic core-adjusting device, and without the need to artificially participating in the setting of target weight, display improves operability, effectively improves the automaticity and operating efficiency that realize fiber array-waveguide device-Array System low-loss rapid abutting joint coupling.

In sum, the present invention effectively overcomes various shortcoming of the prior art and tool high industrial utilization.

Above-described embodiment is illustrative principle of the present invention and effect thereof only, but not for limiting the present invention.Any person skilled in the art scholar all without prejudice under spirit of the present invention and category, can modify above-described embodiment or changes.Therefore, such as have in art usually know the knowledgeable do not depart from complete under disclosed spirit and technological thought all equivalence modify or change, must be contained by claim of the present invention.

Claims (10)

1. a parallel index optimization method for fiber waveguide array-fiber array automatic butt coupling, is characterized in that, comprising:

Set up the mapping between fiber waveguide array-fiber array automatic core-adjusting physical parameter and parallel index optimization model, individuality wherein during aligning in waveguide optical axis parallel index optimization model corresponding to certain relative space position between optical axis and direction, population in the parallel index optimization model of set correspondence in described relative space position and direction, chromosome in the parallel index optimization model of solid axes parameter correspondence, first object function in the parallel index optimization model of luminous power sum correspondence of two or more waveguide channels of aligning Procedure Acquisition, the second target function that the maximum of the luminous power of two or more waveguide channels of aligning Procedure Acquisition and the difference correspondence of minimum value walk abreast in index optimization model,

Adopt computer control, driving stepper motor mode adjusts input monomode fiber array and output monomode fiber array, make it respectively with the deviation of the alignment of fiber waveguide input port and output port in predetermined alignment deviation range, set maximum evolutionary generation, random generation initial population, input monomode fiber array and output monomode fiber array enter the locus of specifying by the regulation of each individuality by the direction attitude of specifying, and read, record, store the optical power value of two or more waveguide channels, this reading, record, storing process carries out successively by individuality order, the optical power value obtained is utilized to calculate the Pareto sum of ranks crowding distance of each individuality,

The principle that individuality that in two identical individualities of the large individuality of Pareto ordinal number, Pareto ordinal number, crowding distance is large is better than the little individuality of crowding distance is better than according to the individuality that Pareto ordinal number is little, after completing the evaluation to individualities all in generation population, employing intersects, make a variation and the step of elimination of the last one produces population of future generation, the step iteration of described intersection, variation and elimination of the last one is carried out, accumulative evolutionary generation, judge whether evolution number of times reaches maximum evolutionary generation, if, end process, completes a waveguide-optical fibre automatic core regulating process; If not, the step continuing to perform described intersection is returned.

2. the parallel index optimization method of fiber waveguide array according to claim 1-fiber array automatic butt coupling, is characterized in that: described predetermined alignment deviation range is-50 μm to+50 μm scopes; In described initial population, individual amount is more than 30 or 30; Described maximum evolutionary generation is more than or equal to 35.

3. the parallel index optimization method of fiber waveguide array according to claim 1-fiber array automatic butt coupling, it is characterized in that: described solid axes parameter is the reference axis parameter that space 11 is tieed up, comprise the 5 dimension coordinate axle parameters of input side except rotating around optical axis and 6 dimension coordinate axle parameters of output side, this 11 dimension coordinate axle variable real number array is encoded.

4. the parallel index optimization method of fiber waveguide array according to claim 3-fiber array automatic butt coupling, is characterized in that: i-th in the described parallel index optimization model 11 dimension coordinate axle variablees individual by any i-th described real number array coding are formed; Wherein, i is more than or equal to 1.

5. the parallel index optimization method of fiber waveguide array according to claim 4-fiber array automatic butt coupling, is characterized in that: described first object function representation is f ₁(U _i), described second target function is expressed as f ₂(U _i);

f ₁(U _i)＝-[P ₁(U _i)+P ₂(U _i)]；

f ₂(U _i)＝|P ₁(U _i)-P ₂(U _i)|；

Wherein, P ₁(U _i) and P ₂(U _i) be respectively aligning position and be in by i-th individual U _iduring the dimensional orientation of regulation, the Output optical power that recorded by two samplings.

6. the parallel index optimization method of fiber waveguide array according to claim 5-fiber array automatic butt coupling, is characterized in that: the Pareto order of described individuality is calculated by the first pre-defined rule; Described first pre-defined rule is:

For any i-th individual U _iif, with another jth individual U _{j ≠ i}there is a kind of target function value relation; Described target function value closes:

F ₁(U _i) > f ₁(U _{j ≠ i}), and f ₂(U _i)>=f ₂(U _{j ≠ i});

Then i-th individual U _ipareto ordinal number increase by 1, otherwise i-th individual U _ipareto ordinal number increase by 0; By with i-th individual U _ioutside all individualities compare one by one, accumulative obtain i-th individual U _ithe value of Pareto ordinal number.

7. the parallel index optimization method of fiber waveguide array according to claim 6-fiber array automatic butt coupling, is characterized in that: i-th individual U _icrowding distance calculated by the second pre-defined rule;

For having identical Pareto ordinal number in population, k the individual U with identical Pareto ordinal number _iaccording to

F ₁(U _m) < f ₁(U _m+1), wherein, m=1,2 ..., k-1;

F ₂(U _n) < f ₂(U _n+1), wherein, n=1,2 ..., k-1;

Order respectively with cambic m subscript and n subscript arrangement;

Calculate and described first object function f ₁(U _m) relevant crowding distance d ₁(U _m),

Calculate and described second target function f ₂(U _n) crowding distance d ₂(U _n),

The arrangement of m subscript and n subscript are arranged correspondence respectively and be returned to original i subscript arrangement, each individual U _iobtain individual U _ithe first crowding distance d ₁(U _i) and the second crowding distance d ₂(U _i), i-th individual U _icrowding distance d (U _i)=d ₁(U _i)+d ₂(U _i).

8. the parallel index optimization method of fiber waveguide array according to claim 7-fiber array automatic butt coupling, it is characterized in that: the step of described elimination of the last one comprises the 3rd pre-defined rule, described 3rd pre-defined rule refer to reject in the maximum groups of individuals of Pareto ordinal number, individuality that crowding distance is minimum.

9. the parallel index optimization method of fiber waveguide array according to claim 8-fiber array automatic butt coupling, it is characterized in that: the step of described intersection refers to and is better than according to the individuality that Pareto ordinal number is little the principle that individuality that in two identical individualities of the large individuality of Pareto ordinal number, Pareto ordinal number, crowding distance is large is better than the little individuality of crowding, from size be select t the row of evaluation the population of M before individuality, newly individual wherein, U _ibe i-th individuality, a _ifor weight coefficient, a _imeet -0.5≤a _i≤ 1.5; New individual U _oform with former M individuality the transition population that a size is M+1, calculate according to described first pre-defined rule and the second pre-defined rule the individual U that makes new advances _opareto order and crowding distance, eliminate last position according to described 3rd pre-defined rule individual, obtain the population of future generation that body quantity is one by one M.

10. the parallel index optimization method of fiber waveguide array according to claim 9-fiber array automatic butt coupling, it is characterized in that: be random selecting body one by one in the population of future generation of M at described individual amount, made a variation by essence, increase a new individuality near parent individuality, again construct the transition population that body quantity is one by one M+1; Calculate Pareto order and the crowding distance of all individualities in described transition population according to the first pre-defined rule and the second pre-defined rule, continue to eliminate last position individuality according to described 3rd pre-defined rule, obtain the new population of future generation that body quantity is one by one M.