CN103513350B - A kind of array waveguide device based on particle cluster algorithm aims at coupling process and device - Google Patents

Abstract

The invention discloses a method and device for aligning and coupling an arrayed waveguide device based on a particle swarm algorithm. The method includes: performing initial light-seeking on the arrayed waveguide; finding a peak position based on a particle swarm algorithm-mathematical optimization, and adding adaptive change inertia weight and perform simulation analysis; use finite measured optical power values and perform optimization iterations; find the actual peak position through optimization iterations; the device includes: CCD system module, laser light source, first positioning module, second positioning module, waveguide chip, second An array of optical fibers, a second array of optical fibers, an optical power detection module, a control system, and an alignment coupling algorithm module; based on the particle swarm algorithm, the present invention can find the peak point only through a small number of iterations, which improves the search efficiency; The optimization process does not involve human authority setting, which solves the shortcomings of the current automatic alignment coupling, and improves the convergence speed, local search ability and search accuracy.

Description

A particle swarm algorithm-based alignment coupling method and device for arrayed waveguide devices

technical field

The invention belongs to the field of arrayed waveguide devices, and in particular relates to a particle swarm algorithm-based alignment coupling method and device for arrayed waveguide devices.

Background technique

Arrayed waveguide devices are key components in optical network communication systems, mainly used for conversion and transmission of optical signals. With the rapid development of optical communication systems, the demand for arrayed waveguide devices in this field is increasing, and the quality requirements for devices are also getting higher and higher. However, the high cost of arrayed waveguide devices has become one of the factors restricting the development of optical network communications. The manufacturing cost of arrayed waveguide devices is mainly concentrated in the packaging process, and the key technology of the packaging process is the coupling and alignment of photonic devices, so efficient The proposed coupling alignment algorithm is of great significance to the cost reduction and quality improvement of integrated photonic devices.

At present, in the arrayed waveguide device packaging manufacturing industry, the manual process is widely used in China to carry out the alignment coupling of the arrayed waveguide device. The traditional alignment and coupling methods of arrayed waveguide devices have the problems of low alignment accuracy, more coupling times, low search efficiency, and low automation of low-loss fast alignment coupling between waveguide chips and array fibers. The speed of automatic alignment is fast, the coupling result is good, and the labor cost is low, which is the development direction of the alignment coupling process of arrayed waveguide devices. However, the fully automatic packaging of arrayed waveguide devices is monopolized by a few foreign companies. Therefore, the research on automatic alignment coupling devices and algorithms for arrayed waveguide devices can effectively break the foreign monopoly and reduce the cost of arrayed waveguide devices.

Contents of the invention

The purpose of the present invention is to provide an arrayed waveguide device alignment coupling method and device based on the particle swarm algorithm, aiming to solve the problem of low alignment accuracy and the need for more coupling times in the traditional arrayed waveguide device alignment coupling method. Low efficiency, low-loss fast alignment coupling of waveguide chips and array fibers, and low automation.

The present invention is achieved in this way, a particle swarm algorithm-based arrayed waveguide device alignment coupling method, the particle swarm algorithm-based arrayed waveguide device alignment coupling method includes:

Step 1: Initially search for light on the arrayed waveguide device;

Step 2: Based on the particle swarm algorithm-mathematical optimization to find the peak position, add adaptive variable inertia weight and conduct simulation analysis;

Step 3: Use limited measured optical power values and perform optimization iterations;

Step 4: Find the actual peak position through optimization iterations.

Further, the method of finding the peak value based on the particle swarm optimization algorithm is to initialize the particle population. After the initialization is completed, the position and velocity of the particles are updated by calculating the fitness of each particle. When the search target meets the threshold requirements, the search is terminated.

Further, the improved particle swarm optimization algorithm can be used instead of the particle swarm optimization algorithm. The improved particle swarm optimization algorithm is a particle swarm optimization algorithm that combines the hybrid PSO model with the introduction of crossover mechanism and the adaptive variable inertia weight. The hybrid PSO model introduces the selection mechanism into the basic particle swarm optimization algorithm. In , the new particles generated in each iteration are selected according to the fitness, and some particles with high fitness are used to replace some particles with low fitness.

Further, the crossover mechanism is similar to the crossover inheritance in the genetic algorithm. The particles to be crossed are selected according to the crossover probability, and new particles are generated after the pairwise crossover operation. The position and velocity vector of the new particles are shown in the following formula:

${\mathrm{<span\; class="notranslate">\; child</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1.0</span>}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; child</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1.0</span>}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; child</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>$

${\mathrm{<span\; class="notranslate">\; child</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>$

in is the position vector of D dimension; and k=1,2, respectively used to indicate the positions of newborn particles and particles to be intersected; is a D-dimensional uniformly distributed random number vector, and the value range of the components is [0,1].

Further, the arrayed waveguide device alignment coupling method based on the particle swarm optimization algorithm adaptively changes the size of the inertia weight ω value to meet the particle’s ability to inherit the current speed, changing the inertia weight value can control the search ability of the algorithm, and adaptively changing the inertia weight ω value formula

$\mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}\mathrm{<span\; class="notranslate">\; t</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; r</span>}\mathrm{<span\; class="notranslate">\; t</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}\mathrm{<span\; class="notranslate">\; t</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; r</span>}\mathrm{<span\; class="notranslate">\; t</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; e</span>}\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; )</span>{<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; k</span>}}{{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

Among them, ω _start is the initial inertia weight; ω _end is the inertia weight when iterating to the maximum number; k is the current iteration number; T _max is the maximum iteration algebra.

Another object of the present invention is to provide an alignment and coupling device for arrayed waveguide devices based on particle swarm optimization algorithm. The alignment and coupling device for arrayed waveguide devices based on particle swarm optimization algorithm includes: CCD system module, laser light source, first positioning module, The second positioning module, waveguide chip, first array optical fiber, second array optical fiber, optical power detection module, control system and alignment coupling algorithm module;

Connected with the control system and the alignment coupling algorithm module, the CCD system module used to realize the initial alignment of the alignment coupling device of the arrayed waveguide device;

The laser light source is arranged on the left side of the first positioning module;

The first positioning module and the second positioning module for installing the input optical fiber and the output optical fiber are respectively installed on the left and right sides of the control system and the alignment coupling algorithm module;

The waveguide chip is set in the middle of the control system and the alignment coupling algorithm module;

The first array of optical fibers is connected to the laser light source, and is arranged above the first positioning module;

The optical power detection module for precise alignment of the arrayed waveguide device packaging device is arranged on the right side of the second array of optical fibers; the second array of optical fibers is connected to the optical power detection module and is arranged on the top of the second positioning module; A control system for controlling an alignment coupling device and an alignment coupling algorithm module.

Further, the arrayed waveguide device alignment coupling device based on the particle swarm optimization algorithm also includes:

Six-axis ultra-precise positioning module for installing input fiber, waveguide chip and output fiber;

Precision optical fixture module for six-axis high-precision adjustment module fixation.

Further, the alignment steps of the arrayed waveguide device alignment coupling device based on the particle swarm optimization algorithm are as follows:

Fix the waveguide chip with a precision fixture, and move the input array fiber under the monitoring of the CCD system module;

The laser light source is coupled to the first array of optical fibers at the input end, and the infrared camera detects the light output from the waveguide chip, and moves the first array of optical fibers at the input end in the order of coarse alignment algorithm plus precise alignment algorithm until the infrared camera detects N regular circles shaped spot;

Under the guidance of the CCD system module, move the second array fiber at the output end to the waveguide chip, move the second array fiber at the output end according to the coarse alignment algorithm until the optical power value is detected, and then move the second array fiber at the output end according to the fine alignment algorithm Array optical fibers until the optical power value of 1 or N channels reaches the maximum;

Rotate the second array of optical fibers at the output end according to the multi-channel alignment scheme until the optical power values of the 1 and N channels reach the best balance.

Further, the arrayed waveguide device alignment coupling device based on the particle swarm algorithm adopts an automatic alignment algorithm.

The arrayed waveguide device alignment and coupling method and device based on the particle swarm algorithm provided by the present invention solve the problem of low alignment accuracy, more coupling times, and low search efficiency in the traditional arrayed waveguide device alignment and coupling method. The low-loss fast alignment coupling of the array fiber has a low degree of automation. Based on the particle swarm optimization algorithm, only a small number of iterations can be used to find the peak point, which can greatly improve the search efficiency. The present invention does not involve artificial authority setting in the optimization process, solves the shortcomings of the existing automatic alignment coupling, has faster convergence speed, stronger local search ability and higher search accuracy, and effectively improves the realization of The degree of automation and work efficiency of low-loss fast alignment coupling of waveguide chip and array fiber.

Description of drawings

FIG. 1 is a schematic structural diagram of an arrayed waveguide device alignment device based on a particle swarm algorithm provided by an embodiment of the present invention;

In the figure: 1. CCD system module; 2. Laser light source; 3. First positioning module; 4. Second positioning module; 5. Waveguide chip; 6. First array optical fiber; 7. Second array optical fiber; 8. Light Power detection module; control system and alignment coupling algorithm module;

Fig. 2 is a flow chart of the basic particle swarm optimization algorithm search principle provided by the embodiment of the present invention;

Fig. 3 is a flow chart of the method for aligning and coupling arrayed waveguide devices based on the particle swarm algorithm provided by the embodiment of the present invention;

Fig. 4 is an application effect diagram of the arrayed waveguide device alignment device based on the particle swarm algorithm provided by the embodiment of the present invention.

detailed description

In order to make the object, technical solution and advantages of the present invention more clear, the present invention will be further described in detail below in conjunction with the examples. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

The application principle of the present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

Fig. 1 and Fig. 4 show the structure of the arrayed waveguide device alignment coupling device based on the particle swarm algorithm provided by the present invention. For ease of illustration, only the parts relevant to the present invention are shown.

The arrayed waveguide device alignment coupling device based on the particle swarm algorithm in the embodiment of the present invention, the structure of the device is mainly composed of: CCD system module 1, laser light source 2, first positioning module 3, second positioning module 4, waveguide chip 5, The first array of optical fibers 6, the second array of optical fibers 7, an optical power detection module 8, a control system and an alignment coupling algorithm module 9;

Connected with the control system and the alignment coupling algorithm module 9, the CCD system module 1 used to implement the initial alignment of the arrayed waveguide device alignment coupling device;

The laser light source 2 is arranged on the left side of the first positioning module 3;

The first positioning module 3 and the second positioning module 4 for installing the input optical fiber and the output optical fiber are respectively installed on the left and right sides of the control system and the alignment coupling algorithm module 9;

The waveguide chip 5 is arranged in the middle of the control system and the alignment coupling algorithm module 9;

The first array of optical fibers 6 is connected to the laser light source 2 and arranged on the first positioning module 3;

The optical power detection module 8 for precise alignment of the arrayed waveguide device packaging device is arranged on the right side of the second array optical fiber 7; the second array optical fiber 7 is connected to the optical power detection module 8 and is arranged on the second positioning module 4 above;

A control system and an alignment coupling algorithm module 9 for controlling the alignment coupling device;

The arrayed waveguide device alignment coupling device based on the particle swarm algorithm also includes:

Six-axis ultra-precise positioning module for installing input fiber, waveguide chip and output fiber;

Precision optical fixture module for six-axis high-precision adjustment module fixation.

The alignment and coupling device for arrayed waveguide devices based on the particle swarm algorithm in the embodiment of the present invention is as follows:

1) Use a precision fixture to fix the waveguide chip 5, and move the input array fiber to a suitable position under the monitoring of the CCD system module 1;

2), the laser light source 2 is coupled to the first array fiber 6 at the input end, the infrared camera detects the light output by the waveguide chip 5, and moves the first array fiber 6 at the input end in the order of coarse alignment algorithm plus precise alignment algorithm until the infrared camera Detect N regular circular light spots;

3) Under the guidance of the CCD system module 1, move the second array optical fiber 7 at the output end to the vicinity of the waveguide chip 5, and move the second array optical fiber 7 at the output end according to the coarse alignment algorithm until a certain optical power value is detected, Then move the second array optical fiber 7 at the output end according to the precise alignment algorithm until the optical power value of the 1 or N channel reaches the maximum;

4) According to the multi-channel alignment scheme, rotate the second array optical fiber 7 at the output end until the optical power values of the 1 and N channels reach the best balance state.

The optical power detection module in the device is used to achieve precise alignment, using an automatic alignment algorithm. Automatic alignment algorithm is the key technology for automatic alignment of arrayed waveguide devices. A fast and reliable alignment algorithm can effectively reduce alignment time and ensure alignment accuracy;

The invention adopts the basic particle swarm algorithm, and proposes an improved particle swarm algorithm with reference to the characteristics of the genetic algorithm.

The present invention is described in conjunction with accompanying drawings 2 and 3, the present invention is realized in this way, a kind of arrayed waveguide device alignment coupling method based on particle swarm algorithm, this method comprises:

S301: Initially search for light on the arrayed waveguide device;

S302: Based on the particle swarm algorithm-mathematical optimization to find the peak position, add adaptive change inertia weight and conduct simulation analysis;

S303: Using a limited measured optical power value and performing optimization iterations;

S304: Find the actual peak position through optimization iterations.

The method of finding the peak value based on the particle swarm optimization algorithm is to initialize the particle swarm. After the initialization is completed, the position and speed of the particles are updated by calculating the fitness of each particle. When the search target meets the threshold requirements, the search is terminated;

The improved particle swarm optimization algorithm can be used instead of the particle swarm optimization algorithm. The improved particle swarm optimization algorithm is a particle swarm optimization algorithm that combines the hybrid PSO model (HybridPSO, HPSO) that introduces a crossover mechanism and the adaptive variable inertia weight. The hybrid PSO model introduces the selection mechanism into the basic particle swarm optimization algorithm, and selects the new particles generated in each iteration according to the fitness, and replaces some particles with lower fitness with some particles with higher fitness. The crossover mechanism is similar to the crossover inheritance in the genetic algorithm. The particles to be crossed are selected with a certain crossover probability, and new particles are generated after pairwise crossover operations. The position and velocity vectors of the new particles are given by:

${\mathrm{<span\; class="notranslate">\; child</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1.0</span>}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; child</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1.0</span>}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; child</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>$

${\mathrm{<span\; class="notranslate">\; child</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; parents</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>$

in is the position vector of D dimension; and k=1,2, respectively used to indicate the positions of newborn particles and particles to be intersected; is a D-dimensional uniformly distributed random number vector, and its component value range is [0,1]. The introduction of the crossover mechanism can theoretically improve the convergence speed of the basic particle swarm optimization algorithm and enhance the local search ability of the algorithm.

Adaptively change the size of the inertia weight ω value to meet the standard particle’s ability to inherit the current speed. Changing the inertia weight value can control the search ability of the algorithm. The formula for adaptively changing the inertia weight ω value

$\mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}\mathrm{<span\; class="notranslate">\; t</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; r</span>}\mathrm{<span\; class="notranslate">\; t</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}\mathrm{<span\; class="notranslate">\; t</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; r</span>}\mathrm{<span\; class="notranslate">\; t</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; e</span>}\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; d</span>}}<span\; class="notranslate">\; )</span>{<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; k</span>}}{{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

Among them, ω _start is the initial inertia weight; ω _end is the inertia weight when iterating to the maximum number; k is the current iteration number; T _max is the maximum iteration algebra.

Using the above formula, the inertia weight ω can be adaptively reduced nonlinearly. The value of ω keeps a larger value in the early iteration process to ensure that the particles have a larger search range. The value of ω changes less in the later iterations, further reducing the The search range balances the global and local search capabilities of the algorithm.

Through simulation analysis and experiments, it can be seen that the crossover PSO algorithm (HPSO) introduces the concept of crossover mechanism, which is equivalent to the genetic crossover in the genetic algorithm, which can speed up the particle optimization process, adaptively change the inertia weight and the crossover PSO algorithm. Combining the advantages of particle swarm optimization and genetic algorithm at the same time, the simulation results also prove that this improved particle swarm algorithm can converge to the optimal solution faster.

The alignment and coupling method of arrayed waveguide devices based on particle swarm algorithm provided by the present invention, the method for automatic alignment and coupling of arrayed waveguide devices is based on particle swarm algorithm-mathematical optimization to find the peak position, using limited measurement of optical power values and performing optimization iterations , and then find the actual peak position. The present invention is based on the particle swarm algorithm to find the peak point only through a small number of iterations, which can greatly improve the search efficiency. The present invention does not involve artificial authority setting in the optimization process, solves the shortcomings of the existing automatic alignment coupling, has faster convergence speed, stronger local search ability and higher search accuracy, and effectively improves the realization of The degree of automation and work efficiency of low-loss fast alignment coupling of waveguide chip and array fiber.

Of course, the present invention also has other embodiments, without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and deformations according to the present invention, but these corresponding changes and All deformations should belong to the protection scope of the claims of the present invention.

Claims (9)

1. the array waveguide device based on particle cluster algorithm aims at a coupling process, it is characterized in that, should aim at coupling process comprise based on the array waveguide device of particle cluster algorithm:

Step one: pair array waveguide device initially seeks light;

Step 2: find peak based on particle cluster algorithm-mathematical optimization, add adaptive change inertia weight and carry out simulation analysis;

Step 3: adopt definite measured optical power value and carry out Optimized Iterative;

Step 4: find actual peak location by Optimized Iterative.

2. aim at coupling process based on the array waveguide device of particle cluster algorithm as claimed in claim 1, it is characterized in that, should be initialization particle colony based on the method for particle cluster algorithm searching peak value, fitness by calculating each particle after initialization completes comes position and the speed of more new particle, after search target meets threshold requirements, search stops.

3. aim at coupling process based on the array waveguide device of particle cluster algorithm as claimed in claim 1, it is characterized in that, particle cluster algorithm can be replaced with improve PSO algorithm, improve PSO algorithm is the particle cluster algorithm adopting the hybridization PSO model introducing crossover mechanism to combine with adaptive change inertia weight, hybridization PSO model is introduced select mechanism in basic particle group algorithm, the new particle that iteration produces each time is selected according to fitness, replaces the low some particles of fitness with the some particles that fitness is high.

4. aim at coupling process based on the array waveguide device of particle cluster algorithm as claimed in claim 3, it is characterized in that, crossover mechanism is similar to the crisscross inheritance in genetic algorithm, select to wait to intersect particle with crossover probability, produce new particle after combination of two interlace operation, position and the velocity of new particle are shown below:

${\mathrm{child}}_1\left(\stackrel{\&RightArrow;}{x}\right)=p*{\mathrm{parent}}_1\left(\stackrel{\&RightArrow;}{x}\right)+(1.0-\stackrel{\&RightArrow;}{p})*{\mathrm{parent}}_2\left(\stackrel{\&RightArrow;}{x}\right)$

${\mathrm{child}}_2\left(\stackrel{\&RightArrow;}{x}\right)=p*{\mathrm{parent}}_2\left(\stackrel{\&RightArrow;}{x}\right)+(1.0-\stackrel{\&RightArrow;}{p})*{\mathrm{parent}}_1\left(\stackrel{\&RightArrow;}{x}\right)$

${\mathrm{child}}_1\left(\stackrel{\&RightArrow;}{v}\right)=\frac{{\mathrm{parent}}_1\left(\stackrel{\&RightArrow;}{v}\right)+{\mathrm{parent}}_2\left(\stackrel{\&RightArrow;}{v}\right)}{|{\mathrm{parent}}_1\left(\stackrel{\&RightArrow;}{v}\right)+{\mathrm{parent}}_2\left(\stackrel{\&RightArrow;}{v}\right)|}\left|{\mathrm{parent}}_1\left(\stackrel{\&RightArrow;}{v}\right)\right|$

${\mathrm{child}}_2\left(\stackrel{\&RightArrow;}{v}\right)=\frac{{\mathrm{parent}}_1\left(\stackrel{\&RightArrow;}{v}\right)+{\mathrm{parent}}_2\left(\stackrel{\&RightArrow;}{v}\right)}{|{\mathrm{parent}}_1\left(\stackrel{\&RightArrow;}{v}\right)+{\mathrm{parent}}_2\left(\stackrel{\&RightArrow;}{v}\right)|}\left|{\mathrm{parent}}_1\left(\stackrel{\&RightArrow;}{v}\right)\right|$

Wherein the position vector of D dimension; with k=1,2, be used for respectively showing newborn particle and the position for the treatment of intersection particle; be that D ties up equally distributed random number vector, component span is [0,1].

5. aim at coupling process based on the array waveguide device of particle cluster algorithm as claimed in claim 1, it is characterized in that, the size particle up to standard of coupling process adaptive change inertia weight ω value should be aimed to the inheritance capability of present speed based on the array waveguide device of particle cluster algorithm, changing inertia weight value can the search capability of control algolithm, the formula of adaptive change inertia weight ω value

$\mathrm{\&omega;}\left(k\right)={\mathrm{\&omega;}}_{start}-({\mathrm{\&omega;}}_{start}-{\mathrm{\&omega;}}_{end}){\left(\frac{k}{T_{max}}\right)}^2$

Wherein, ω _startit is initial inertia weight; ω _endthat iteration is to inertia weight during maximum times; K is current iteration number of times; T _maxgreatest iteration algebraically.

6. the array waveguide device based on particle cluster algorithm aims at coupling device, it is characterized in that, coupling device should be aimed at based on the array waveguide device of particle cluster algorithm and comprise: CCD system module, LASER Light Source, the first locating module, the second locating module, waveguide chip, the first array fibre, the second array fibre, luminous power detection module, control system and aligning coupling algorithm module;

With control system and aim at coupling algorithm model calling, aim at for pair array waveguide device the CCD system module that coupling device realizes initial alignment;

LASER Light Source is arranged on the left side of the first locating module;

For installing input optical fibre, the first locating module of output optical fibre and the second locating module, being arranged on control system respectively and aiming at the left and right sides of coupling algorithm module;

Waveguide chip is arranged on control system and aims at the centre position of coupling algorithm module;

First array fibre connects LASER Light Source, is arranged on above the first locating module;

Realize the luminous power detection module of fine registration for pair array waveguide device packaging system, be arranged on the right side of the second array fibre; Second array fibre connects luminous power detection module, is arranged on above the second locating module; For to aiming at the control system that controls of coupling device and aiming at coupling algorithm module.

7. aim at coupling device based on the array waveguide device of particle cluster algorithm as claimed in claim 6, it is characterized in that, coupling device should be aimed at based on the array waveguide device of particle cluster algorithm and also comprise:

For installing six axle superfinishing locating modules of input optical fibre, waveguide chip and output optical fibre;

For the precision optics fixture module that six axle high precision adjusting modules are fixing.

8. aim at coupling device based on the array waveguide device of particle cluster algorithm as claimed in claim 7, it is characterized in that, should be as follows based on the array waveguide device aligning coupling device alignment procedures of particle cluster algorithm:

Precise clamp is utilized to be fixed by waveguide chip, mobile input array optical fiber under the monitoring of CCD system module;

LASER Light Source is coupled with input end first array fibre, and infrared camera detects the light that waveguide chip exports, and adds order mobile input end first array fibre of accurate alignment algorithm, until infrared camera detects the circular light spot of N number of rule according to coarse alignment algorithm;

Under the guiding of CCD system module, mobile output terminal second array fibre is to waveguide chip, output terminal second array fibre is moved according to coarse alignment algorithm, until optical power value detected, then move output terminal second array fibre according to accurate alignment algorithm, until 1 or N channel optical power value reach maximum;

According to hyperchannel alignment scheme rotary output second array fibre, until 1 and N channel optical power value reach optimal equalization state.

9. as claimed in claim 6 aim at coupling device based on the array waveguide device of particle cluster algorithm, it is characterized in that, what should aim at that coupling device adopts based on the array waveguide device of particle cluster algorithm is robotization alignment algorithm.