CN116435861A - Device and method for automatically identifying and switching state of middle-infrared ultrashort pulse laser - Google Patents
Device and method for automatically identifying and switching state of middle-infrared ultrashort pulse laser Download PDFInfo
- Publication number
- CN116435861A CN116435861A CN202310396047.3A CN202310396047A CN116435861A CN 116435861 A CN116435861 A CN 116435861A CN 202310396047 A CN202310396047 A CN 202310396047A CN 116435861 A CN116435861 A CN 116435861A
- Authority
- CN
- China
- Prior art keywords
- pulse
- output
- state
- mode
- laser
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 230000010287 polarization Effects 0.000 claims abstract description 106
- 239000013307 optical fiber Substances 0.000 claims abstract description 55
- 239000000835 fiber Substances 0.000 claims description 57
- 238000001228 spectrum Methods 0.000 claims description 49
- 238000005086 pumping Methods 0.000 claims description 20
- 238000003062 neural network model Methods 0.000 claims description 16
- 238000004891 communication Methods 0.000 claims description 15
- 238000005457 optimization Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 14
- 230000002068 genetic effect Effects 0.000 claims description 13
- 238000005259 measurement Methods 0.000 claims description 9
- 230000009191 jumping Effects 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 8
- 230000003595 spectral effect Effects 0.000 claims description 7
- 238000004088 simulation Methods 0.000 claims description 6
- 229910052775 Thulium Inorganic materials 0.000 claims description 5
- FRNOGLGSGLTDKL-UHFFFAOYSA-N thulium atom Chemical compound [Tm] FRNOGLGSGLTDKL-UHFFFAOYSA-N 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 4
- 238000004804 winding Methods 0.000 claims description 4
- 230000003287 optical effect Effects 0.000 description 28
- 230000000694 effects Effects 0.000 description 19
- 238000005516 engineering process Methods 0.000 description 14
- 239000006185 dispersion Substances 0.000 description 12
- 238000012545 processing Methods 0.000 description 11
- 238000013527 convolutional neural network Methods 0.000 description 8
- 239000006096 absorbing agent Substances 0.000 description 7
- 230000005540 biological transmission Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000011160 research Methods 0.000 description 6
- 238000010521 absorption reaction Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000009022 nonlinear effect Effects 0.000 description 5
- 230000003321 amplification Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 230000018109 developmental process Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000003199 nucleic acid amplification method Methods 0.000 description 4
- 230000000737 periodic effect Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 238000003754 machining Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 206010063385 Intellectualisation Diseases 0.000 description 2
- 238000001069 Raman spectroscopy Methods 0.000 description 2
- 230000002547 anomalous effect Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000005459 micromachining Methods 0.000 description 2
- 230000010355 oscillation Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 239000002002 slurry Substances 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000005684 electric field Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000008140 language development Effects 0.000 description 1
- 230000004001 molecular interaction Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000012634 optical imaging Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000010363 phase shift Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01S—DEVICES USING THE PROCESS OF LIGHT AMPLIFICATION BY STIMULATED EMISSION OF RADIATION [LASER] TO AMPLIFY OR GENERATE LIGHT; DEVICES USING STIMULATED EMISSION OF ELECTROMAGNETIC RADIATION IN WAVE RANGES OTHER THAN OPTICAL
- H01S3/00—Lasers, i.e. devices using stimulated emission of electromagnetic radiation in the infrared, visible or ultraviolet wave range
- H01S3/10—Controlling the intensity, frequency, phase, polarisation or direction of the emitted radiation, e.g. switching, gating, modulating or demodulating
- H01S3/11—Mode locking; Q-switching; Other giant-pulse techniques, e.g. cavity dumping
- H01S3/1106—Mode locking
- H01S3/1112—Passive mode locking
- H01S3/1115—Passive mode locking using intracavity saturable absorbers
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01S—DEVICES USING THE PROCESS OF LIGHT AMPLIFICATION BY STIMULATED EMISSION OF RADIATION [LASER] TO AMPLIFY OR GENERATE LIGHT; DEVICES USING STIMULATED EMISSION OF ELECTROMAGNETIC RADIATION IN WAVE RANGES OTHER THAN OPTICAL
- H01S3/00—Lasers, i.e. devices using stimulated emission of electromagnetic radiation in the infrared, visible or ultraviolet wave range
- H01S3/05—Construction or shape of optical resonators; Accommodation of active medium therein; Shape of active medium
- H01S3/08—Construction or shape of optical resonators or components thereof
- H01S3/08018—Mode suppression
- H01S3/08022—Longitudinal modes
- H01S3/08031—Single-mode emission
- H01S3/08036—Single-mode emission using intracavity dispersive, polarising or birefringent elements
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01S—DEVICES USING THE PROCESS OF LIGHT AMPLIFICATION BY STIMULATED EMISSION OF RADIATION [LASER] TO AMPLIFY OR GENERATE LIGHT; DEVICES USING STIMULATED EMISSION OF ELECTROMAGNETIC RADIATION IN WAVE RANGES OTHER THAN OPTICAL
- H01S3/00—Lasers, i.e. devices using stimulated emission of electromagnetic radiation in the infrared, visible or ultraviolet wave range
- H01S3/05—Construction or shape of optical resonators; Accommodation of active medium therein; Shape of active medium
- H01S3/08—Construction or shape of optical resonators or components thereof
- H01S3/08054—Passive cavity elements acting on the polarization, e.g. a polarizer for branching or walk-off compensation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Engineering & Computer Science (AREA)
- Plasma & Fusion (AREA)
- Optics & Photonics (AREA)
- Lasers (AREA)
Abstract
An automatic identification and switching device and method for the state of a middle infrared ultrashort pulse laser, the device: the pump input arm of the beam combiner is connected with the pump source, one end of the gain optical fiber is connected with the signal output arm of the beam combiner, one end of the single-mode optical fiber is connected with the other end of the gain optical fiber, one input arm of the output coupler is connected with the other end of the single-mode optical fiber, one output arm of the output coupler is connected with the spectrometer, the input end of the isolator is connected with the other output arm of the output coupler, the input end of the polarizer is connected with the output end of the isolator, the input end of the electric polarization controller is connected with the output end of the polarizer, and the output end of the electric polarization controller is connected with the signal input arm of the beam combiner. The method comprises the following steps: the spectrometer receives the output signal of the mode-locked laser and sends the output signal to the control terminal, and the control terminal optimizes the control parameters according to the matching degree of the output laser pulse state and the set target pulse state. The device and the method can realize the automatic adjustment and the switching of the noise-like mode locking pulse and the soliton mode locking pulse.
Description
Technical Field
The invention relates to a device for automatically identifying and switching a medium infrared ultrashort pulse laser state, and belongs to the technical field of mode-locked lasers and automatic control.
Background
The ultra-fast fiber laser has important application in the fields of communication, sensing, precision machining and the like. How to improve the power and energy of the laser pulse, grasp and utilize the complex nonlinearity of the ultra-short pulse fiber laser, develop the key technical research of the application of the high-energy and high-peak-power all-fiber ultra-short pulse laser, and become an important scientific problem of the fiber pulse laser.
Mode locking is one of the important means to achieve ultra-fast lasers. A nonlinear polarization rotation mode-locked laser is one type of passive mode-locked fiber laser, and has been attracting attention because of its simple structure. Compared with other mode-locked lasers, the mode-locked laser has an all-fiber structure, and does not need to introduce external electric signals for modulation, so that the all-fiber mode-locked laser can realize all-optical transmission, is not limited by modulation bandwidth, can realize the output of mode-locked pulses in a very wide wavelength range, has response speed determined by the electric polarization origin of the nonlinear effect of the optical fiber, and can output pulse width reaching the femtosecond magnitude. These superior properties exhibited by mode-locked lasers have led researchers to conduct extensive research on them.
The noise-like mode locking is a special pulse generated by a mode locking laser under a certain condition, has the characteristics of high pulse envelope energy, high average power, low coherence, good environmental stability and the like, can be applied to the field of basic physical research, and has important application value in production practice due to a plurality of special properties. The low coherence is an inherent characteristic of noise-like pulses, and the characteristic of pulses can be applied to typical low coherence spectrum interference technologies such as fiber grating demodulation technology, fiber information storage and reproduction technology and the like. The noise-like pulse can also obtain ultra-wide flat spectrum by using dispersion management technology, raman effect, double refraction effect, soliton self-frequency shift effect and the like. In recent years, with the development of fiber amplification technology and the continuous increase of pumping pulse power, the noise-like pulse energy available in experiments is also continuously increased, and single pulse energy of hundreds of nanojoules can be realized in a fiber laser. The high energy noise-like pulses can be effectively used not only in machining (micromachining) operations, but also as a pumping source to produce supercontinuum.
Soliton mode locking is also a special pulse generated by a mode-locked laser under certain conditions, the formation of an optical soliton in an optical fiber is based on mutual balance of anomalous Group-velocity dispersion (GVD) of the optical fiber and Self-phase modulation (SPM) in nonlinear effects of the optical fiber, and the optical soliton pulse can be transmitted in the optical fiber for long distances without distortion. Due to the characteristic of the optical soliton, the optical soliton communication system has great advantages in ultra-long-distance and ultra-large-capacity optical communication. Both theoretical analysis and experimental investigation of optical solitons have been developed until now. In the fields of optical communication, signal processing, and the like, various types of optical solitons and applications thereof are being developed and studied.
In a soliton mode-locked laser, soliton pulses circulate in the laser cavity. If the effects of dispersion and nonlinearity in the cavity are weak, the pulse may experience some weak periodic disturbances due to the dispersion and nonlinearity dispersion. Some additional interference may also be due to periodic losses or amplification within the resonant cavity. Such regularly occurring disturbances may cause solitons to propagate together with the dispersive wave. This will not tend to have a significant effect because only solitons will experience non-linearities, so that the relative phases of the solitary and dispersive waves are constantly changing. However, resonant coupling (or quasi-phase matching) sometimes occurs in fiber lasers such that at certain frequencies the phase between the dispersive wave and the optical soliton will vary by an integer multiple of 2pi in each cycle, appearing as a series of paired narrow peaks on the soliton spectrum, such narrow peaks being known as kelvin sidebands. In the soliton state of the fourier transform limit, the distance between the keley sidebands reflects the magnitude of the chromatic dispersion within the cavity.
However, the polarization control of the mode-locked fiber laser based on the nonlinear polarization rotation effect always has problems, most of the current control is manual control, whether the mode-locked laser output depends on the working experience of a user or not can be realized, the mode-locked fiber laser based on the nonlinear polarization rotation effect is sensitive to temperature and noise, and the same output cannot be ensured by control parameters under different conditions. A stable mode-locked fiber laser is lost when the environment is obviously changed such as temperature rise, and once lost, a professional maintainer is required to recover the mode, so that the mode-locked fiber laser is greatly restricted from being used in the military and civil fields.
In order to solve the above problems, in recent years, few experiments for realizing automatic mode locking by using electric control polarization have been reported, wherein U.Andral et al at Bogong university and R.I. Woodward et al at Imperial university sequentially use a genetic algorithm and combine an electric control polarization technology to realize automatic mode locking, but the former has a complicated experimental structure and needs two electric control polarization controllers and 6-path voltage control; the latter mode locking recognition process is extremely complex, and the time domain, frequency domain and spectrum information are needed to be used simultaneously for comprehensive recognition. In addition, the modes are controlled and identified for single mode locking state noise-like mode locking. The fast automatic mode locking method for covering multi-state pulse identification is provided by Chinese patent publication No. CN108539571A, which uses a high-speed oscilloscope as detection equipment, adjusts the polarization state according to an optimization algorithm, solves the problem of polarization control in a passive mode locking laser based on nonlinear polarization evolution, and the signal detected by the high-speed oscilloscope is still a time domain signal, and the noise-like mode locking and the soliton mode locking are both mode locking states, so that the time domain characteristics are highly consistent. However, the noise-like pulse has the same pulse envelope as the soliton locking mold, and is limited by the low bandwidth of the photoelectric detector (particularly, in the mid-infrared band of 2-3 microns, the bandwidth of the time domain detection device is strictly limited, the commercial maximum bandwidth of 2 microns is 22GHz, and the maximum bandwidth of the 3 micron detector is 1GHz level), so that the two states cannot be directly distinguished by the existing photoelectric detection system. The publication date is 20170111, and the Chinese patent publication number is CN106329303A proposes an automatic mode locking control method of an automatic mode locking optical fiber laser, and the method can judge the current working state of the laser by counting and detecting the level of a signal of a light detector, and still can not realize effective identification and control of noise-like mode locking and soliton mode locking.
Therefore, development of an intelligent recognition method capable of realizing accurate polarization control and effectively recognizing noise-like pulses and soliton pulses is needed, so that automatic recognition and switching of noise-like mode locking states and soliton mode locking states are achieved, different ultrafast laser outputs are achieved, and application scenes of the mode locking laser are expanded.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides a device and a method for automatically identifying and switching the state of a middle infrared ultrashort pulse laser, which have compact structure, can realize the output of ultrashort pulse laser with high power, high beam quality, high efficiency and high stability, and can conveniently realize the automatic adjustment and switching of noise-like mode locking pulse and soliton mode locking pulse; the method can automatically realize accurate and efficient identification of the noise-like pulse and soliton pulse states, can solve the problem of state misjudgment in the application of the mid-infrared ultrafast fiber laser pulse, and can realize automatic adjustment and switching of the two pulse states.
In order to achieve the above purpose, the invention provides a device for automatically identifying and switching the state of a mid-infrared ultrashort pulse laser, which comprises a pumping source, a mode-locked laser, a spectrometer and a control terminal;
The mode-locked laser consists of a beam combiner, a gain optical fiber, a single mode fiber, an output coupler, an isolator, a polarizer and an electric polarization controller;
the pump input arm of the beam combiner is connected with the output end of the pump source through a passive optical fiber, one end of the gain optical fiber is connected with the signal output arm of the beam combiner, one end of the single-mode optical fiber is connected with the other end of the gain optical fiber, one input arm of the output coupler is connected with the other end of the single-mode optical fiber, one output arm of the output coupler is connected with the input end of the spectrometer, the input end of the isolator is connected with the other output arm of the output coupler, the input end of the polarizer is connected with the output end of the isolator, the input end of the electric polarization controller is connected with the output end of the polarizer, and the output end of the electric polarization controller is connected with the signal input arm of the beam combiner;
and the control terminal is respectively connected with the pumping source, the electric polarization controller and the spectrometer.
Preferably, the control terminal is a computer.
As a preferable mode, the measurable range of the spectrometer is 1200 nm-2400 nm, the optical power of-70 dBm can be measured, 55dB dynamic measurement under the resolution of 0.05nm can be realized, the maximum scanning speed under the span of 100nm is 0.5s, and a logarithmic coordinate mode is selected; the spectrometer is provided with GPIB, RS232 and an Ethernet interface, and supports an external PC to carry out remote control and data acquisition.
Preferably, the electric polarization controller is a polarization controller based on a slurry structure, the outer diameter of a winding disc is 18mm, each slurry can rotate 170 degrees, the minimum step size is 0.12 degrees, and any polarization state on the bungjia ball can be covered.
Preferably, the pump source supports an external level signal to control the output of pump power, and the adjustable power range is 0-30W.
Preferably, the gain fiber 3 is a thulium doped gain fiber.
In the invention, through the arrangement of the beam combiner, the pump light and the signal light can be coupled into the same optical fiber; through the arrangement of the gain fiber, the laser of the micrometer wave band can be excited after the light emitted by the pumping source is received; by arranging the output coupler, light in the optical fiber annular cavity can be divided into two beams, one beam is output to facilitate the spectrometer to collect spectrum data, and the other beam of light continues to perform continuous oscillation feedback in the cavity; through the arrangement of the isolator and the polarizer, the pump light and the signal light in the optical fiber annular cavity can be transmitted in one direction. By setting the electric polarization controller, the saturable absorption effect caused by nonlinear polarization rotation in the optical fiber annular cavity can be changed, and then a stable mode locking effect can be realized. The spectrometer is connected with one output arm of the output coupler, so that the output signal can be conveniently sampled to obtain spectrum data, and meanwhile, the spectrometer is connected with the control terminal, so that the spectrometer can conveniently send the obtained spectrum data to the control terminal for data processing. The control terminal is further connected with the pump source and the electric polarization controller respectively, so that the control terminal can conveniently issue control parameters to the pump source and the electric polarization controller to execute, and can conveniently realize the switching of noise-like mode locking pulse and soliton mode locking pulse by controlling the electric polarization controller. The device has compact structure, can realize the output of ultrashort pulse laser with high power, high beam quality, high efficiency and high stability, and simultaneously can be convenient for realizing the automatic adjustment and switching of noise-like mode locking pulse and soliton mode locking pulse.
The invention also provides a method for automatically identifying and switching the state of the mid-infrared ultrashort pulse laser, which comprises the following steps:
step one: selecting a target pulse through a control terminal, and controlling the mode-locked laser to start working;
step two: receiving an output signal of the mode-locked laser by utilizing a spectrometer, and sending the spectrum data obtained by sampling to a control terminal;
step three: the control terminal uses a trained neural network model to perform mode locking state identification on spectrum data, calculates the matching degree of a measured pulse spectrum and a target spectrum, considers that the matching degree of an output laser pulse state and a set target pulse state is more than 80 percent and is consistent, and directly executes the fifth step; if the matching degree of the output laser pulse state and the set target pulse state is less than or equal to 80%, the output laser pulse state is not consistent with the set target pulse state, and the fourth step is directly executed;
step four: the control terminal generates corresponding control parameters according to a genetic optimization algorithm, and sends the control parameters to the pumping source and the electric polarization controller through a serial port communication protocol, wherein the control parameters comprise the pumping power of the pumping source and the blade angle of the electric polarization controller;
the specific method of the genetic optimization algorithm is as follows:
S1: randomly generating 20 groups of initial control parameters through a control terminal, and respectively sending the initial control parameters to a pump source and an electric polarization controller to enable the pump source and the electric polarization controller to execute the received control parameters;
s2: calculating, namely calculating the matching degree of the spectrum data corresponding to the 20 groups of initial control parameters in the S1 and the target pulse by using a control terminal;
s3: sequencing, namely sequencing 20 groups of initial parameters according to the matching degree by using a control terminal;
s4: selecting, namely selecting the top 6 groups of control parameters with highest matching degree ranking by using a control terminal;
s5: inheritance, 6 groups of parameters are randomly and alternately combined by utilizing a control terminal, and a new 10 groups of control parameters are generated;
s6: a step of mutating, in the newly generated control parameters in the step S5, randomly selecting 4 groups of control parameters by using a control terminal, and randomly modifying the control parameter values in the control parameters to generate new 20 groups of control parameters;
s7: the newly generated 20 groups of control parameters are sent to a pumping source and an electric polarization controller through a control terminal;
step five: if the output laser pulse state accords with the target pulse, pulse state monitoring is carried out, wherein the pulse state monitoring means that the current pulse state is compared with the set target pulse state at regular intervals, and whether the current output accords with the set or not is judged; if the output laser state is suddenly changed and is not consistent with the target pulse, jumping to the third step, and searching the target state again;
Step six: if the control terminal switches the target pulse, the step II is executed in a jumping mode;
in step three, the neural network model has a total of 1000 sample training and is spectral data, wherein 600 are simulation data, 400 are experimental data, wherein the number of noise-like mode locking pulse data is 300, the number of soliton mode locking pulse data is 300, and the rest is non-mode locking state data.
Preferably, in step one, the target pulse is a selected noise-like mode-locked pulse or soliton mode-locked pulse.
The invention uses the electric polarization controller as a polarization control device in the mode-locked laser, uses the spectrometer as a measuring device and receives the output signal of the mode-locked laser, simultaneously, carries out mode-locked state identification on the sampled spectrum data through a neural network model on a control terminal, then determines whether to carry out optimization processing according to the matching degree of the output laser pulse state and the set target pulse state, adopts a genetic algorithm in the optimization processing process, can effectively ensure the stability of the output signal of the laser in the target state through a closed loop feedback structure and effective state identification, and can realize the switching of noise-like mode-locked pulse and soliton mode-locked pulse through controlling the electric polarization controller. Because the noise-like mode locking pulse and the soliton mode locking pulse are highly consistent in time domain characteristics, the noise-like mode locking pulse and the soliton mode locking pulse are difficult to effectively distinguish through an oscilloscope or a photoelectric detector, the invention can automatically realize accurate and efficient identification of the noise-like pulse and the soliton pulse state through the reinforced training of the spectrum state of the ultrafast laser pulse by the convolutional neural network, solves the problem of misjudgment of the states in the application of the mid-infrared ultrafast fiber laser pulse, and realizes automatic adjustment and switching of the two pulse states. The neural network training algorithm adopted by the invention has the advantages of high identification speed, high identification precision, intellectualization and the like, and the reliability of the mid-infrared ultrafast fiber laser in industrial application is remarkably improved. The genetic algorithm adopted intelligently controls the laser pulse state, realizes the rapid and accurate switching of multiple laser states of complex noise-like pulse states and soliton states, and compared with the existing manual control method, the obtained laser state is more efficient and the cost of manual control is greatly reduced. Compared with the existing mid-infrared ultrafast laser system, the invention only needs to adopt a common commercial single-mode fiber instead of the polarization maintaining fiber with high cost adopted by the main stream method, thereby greatly reducing the cost. The intelligent identification and control method adopted by the invention also omits a complicated temperature control system, a mechanical vibration control system and an environment disturbance prevention control system, can be quickly adjusted according to the environment change, and greatly enhances the environment adaptability. By adopting a full-automatic intelligent control system, various pulse-shaped outputs such as various noise types, solitons, harmonic mode locking solitons and the like can be realized, and various customized laser pulse outputs can be generated according to the requirements of customers, so that the application range of the laser system is greatly increased, and the high-precision laser processing problems of materials with complex structures can be solved by rapidly switching pulses in different states.
Drawings
FIG. 1 is a schematic diagram of the structure of the present invention;
FIG. 2 is a schematic diagram of setting noise-like mode locking as a target pulse at the LabVIEW platform end;
FIG. 3 is a noise-like mode-locked pulse spectral measurement signal in accordance with the present invention;
FIG. 4 is a noise-like mode-locked pulse time domain measurement signal in accordance with the present invention;
FIG. 5 is a schematic diagram of setting soliton mode locking as a target pulse at the LabVIEW platform end;
FIG. 6 is a soliton mode-locked pulse spectrometry signal in the present invention;
fig. 7 is a soliton mode-locked pulse time domain measurement signal in the present invention.
In the figure: 1. the device comprises a pumping source, a beam combiner, a gain optical fiber, a single-mode optical fiber, an output coupler, an isolator, a polarizer, an electric polarization controller, a spectrometer, a control terminal and a control terminal, wherein the pumping source, the beam combiner, the gain optical fiber, the single-mode optical fiber, the output coupler, the isolator, the polarizer and the electric polarization controller are sequentially arranged in sequence, and the electric polarization controller, the spectrometer, the gain optical fiber, the single-mode optical fiber, the output coupler, the isolator, the polarizer and the polarizer are sequentially arranged in sequence.
Detailed Description
The invention will be further described with reference to the accompanying drawings and examples.
The ultra-fast fiber laser can generate ultra-short pulse with high peak power and wide spectrum, and plays an indispensable role in basic scientific research and industrial application, such as serving as a light source of a nonlinear optical imaging system in the biomedical field; as another example, high energy short pulse fiber lasers may be used for laser fine microstructure processing.
Mode locking (Mode locking) is one of methods for generating ultrashort pulses by a laser, and similar to Q-Switching, mode locking also modulates the laser cavity to split the original Continuous Wave (CW) to generate pulses. However, mode locking and Q-switching are similar in physical phenomenon only, and the physical principles are quite different. Mode locking techniques are generally classified into active mode locking and passive mode locking. The active mode locking adopts a method for periodically modulating the parameters of the resonant cavity. The basis is that the loss or optical path (amplitude modulation and phase modulation) of the resonant cavity is periodically changed with a certain modulation frequency by using a modulator controlled by an external signal in the resonant cavity. When the modulation frequency is chosen to be equal to the spacing of the longitudinal modes, the modulation of each mode produces sidebands whose frequencies coincide with the frequencies of two adjacent longitudinal modes. Due to the interaction between the longitudinal modes, all the modes are synchronized under the strong enough modulation, and the modes are coherently overlapped to form a mode locking sequence pulse. Another effective method of generating ultrashort pulses is passive mode locking. The method is implemented by placing a saturable absorber in a laser resonator. The saturable absorber is a nonlinear absorption medium which has absorption transition characteristics and a large absorption cross section for laser frequencies. Once the radiation pulse of the laser is incident on such an absorber, the absorber molecules absorb the laser radiation and as the laser intensity increases, its upper level particle count increases, and when the laser intensity is greater than the saturation intensity of the absorber, the absorber becomes saturated, causing the laser pulse of the greatest intensity to undergo minimal loss and pass freely through it, thus obtaining a very strong mode-locked pulse. It is similar to a passive Q-switch, but differs. Passive mode locking requires that the upper energy level lifetime of the saturable absorber be extremely short, while placing in the cavity must be in close proximity to the total reflection mirror.
Nonlinear polarization rotation mode locking belongs to one type of passive mode locking, nonlinear polarization rotation (Nonlinear Polarization Rotation, NPR), nonlinear polarization rotation, also called NPR, is a typical unidirectional ring cavity. The mode locking element is formed by the nonlinear polarization rotation effect of light in the weak birefringent fiber and two polarization state mutually perpendicular polarization controllers or wave plate groups (polarizer and analyzer). The polarization state of the light can be changed by adjusting the wave plate or the polarization controller, and the light is changed from elliptical polarization to linear polarization and then to elliptical polarization. During pulse transmission, the phase shifts induced from the phase modulation effect and the cross-phase modulation effect act on the orthogonal polarization components to continuously rotate their polarization vectors along the fiber. Because of the intensity dependence of the rotation angle of the nonlinear polarization, when the lower pulse intensity part passes through the optical fiber, the polarization vector hardly rotates due to small nonlinear effect, and the transmission loss is large because the polarization states of the two polarization controllers are perpendicular. While when the central portion of the light pulse with higher intensity passes through the optical fiber, the nonlinear effect is strong and the polarization vector rotates so that it can pass through the polarizer less. As a result, the loss of the two wings of the pulse with low intensity is large, the central loss of the pulse with high intensity is small, and the pulse is narrowed periodically. Although all-fiber mode-locked laser pulses on the order of femtoseconds are realized by means of nonlinear polarization rotation technology, the nonlinear polarization rotation environment instability is caused by the extremely sensitive polarization state to the environment due to the fact that pulse formation and output are polarization-dependent, and commercial production of the all-fiber mode-locked laser pulses is greatly limited. It is highly determined that the optimal polarization setting will vary with temperature and therefore need to be readjusted, and it is difficult to experimentally quantify and achieve a particular modulation depth and saturation power.
Noise-like pulse (NLP) is a special pulse generated by a mode-locked laser under a certain condition, and has the characteristics of high energy, wide pulse width, low coherence and the like. The noise-like pulse can be applied to the field of basic physical research, and has important application value in production practice due to a plurality of special properties of the noise-like pulse. The low coherence is an inherent characteristic of noise-like pulses, and the characteristic of pulses can be applied to typical low coherence spectrum interference technologies such as fiber grating demodulation technology, fiber information storage and reproduction technology and the like. The noise-like pulse can also obtain ultra-wide flat spectrum by using dispersion management technology, raman effect, double refraction effect, soliton self-frequency shift effect and the like. In recent years, with the development of fiber amplification technology and the continuous increase of pumping pulse power, the noise-like pulse energy available in experiments is also continuously increased, and single pulse energy of hundreds of nanojoules can be realized in a fiber laser. The high energy noise-like pulses can be used not only effectively in machining (micromachining) but also as a pumping source to produce supercontinuum.
In most cases, the noise-like pulse cannot be used as an ideal high-energy pulse due to a wider substrate and a lower signal-to-noise ratio, but is subjected to mode locking in an anomalous dispersion region by using a nonlinear fiber loop mirror technology, so that the noise-like pulse with a high signal-to-noise ratio can be obtained, and has great application potential in the femtosecond pulse with high energy, and the application range can relate to the fields of laser radar, electric field and molecular interaction and the like. The dynamic evolution process of the nonlinear optical system can be stabilized and controlled theoretically by using the dispersion management technology.
Soliton (Soliton), also known as Soliton, is a special form of ultrashort pulse, or a pulsed traveling wave that maintains its shape, amplitude, and velocity during propagation. An optical soliton (soliton) is a pulse of light that can propagate in an optical fiber for a long period of time while maintaining its form, amplitude, and velocity. The optical soliton characteristic can be utilized to realize ultra-long distance and ultra-large capacity optical communication. In soliton mode-locked lasers, soliton pulses circulate in the laser cavity. If the effects of dispersion and nonlinearity in the cavity are weak, the pulse may experience some weak periodic disturbances due to the dispersion and nonlinearity dispersion. Some additional interference may also be due to periodic losses or amplification within the resonant cavity. Resonant coupling (or quasi-phase matching) sometimes occurs in fiber lasers such that at certain frequencies the phase between the dispersive wave and the optical soliton will vary by an integer multiple of 2pi in each cycle.
The characteristics of the optical soliton determine the application prospect of the optical soliton in the field of communication. The fundamental soliton is typically used for communication because it does not change anything throughout the propagation. The optical soliton communication has the following characteristics:
(1) The capacity is large: the transmission code rate can reach 20Gb/s generally and can reach more than 100Gb/s at most;
(2) Low error rate and strong anti-interference capability: the optical solitons of the base order are kept unchanged in the transmission process, and the adiabatic characteristic of the optical solitons determines that the error rate of the optical solitons transmission is greatly lower than that of conventional optical fiber communication, and even the error-free optical fiber communication with the error rate lower than 10 < -12 > can be realized;
(3) Relay stations may not be used: the optical signal can be transmitted for a very long distance without distortion by only performing gain compensation on the optical fiber loss, so that the complex processes of photoelectric conversion, reshaping and amplifying, error code checking, photoelectric conversion, retransmission and the like are avoided.
The optical soliton maintains its shape during propagation and is stable even when subjected to a small amount of disturbance, and has been considered as an ideal optical information carrier. The self-stabilizing property of the soliton is derived from the balance between the dispersive effect of light and the nonlinear effect, and the concept of an "attractor" in nonlinear science is realized in an optical system.
The convolutional neural network (Convolutional Neural Network, CNN) is a feed-forward neural network whose artificial neurons can respond to surrounding cells in a part of the coverage area with excellent performance for large image processing. Convolutional neural networks consist of one or more convolutional layers and a top fully connected layer (corresponding to classical neural networks) and also include associated weights and pooling layers (pooling layers).
LabVIEW (Laboratory Virtual instrument Engineering Workbench) is a graphical programming language development environment that is widely accepted by industry, academia, and research laboratories as a standard data acquisition and instrumentation control software. LabVIEW integrates all the functions of communicating with hardware and data acquisition cards that meet the GPIB, VXI, RS-232 and RS-485 protocols. It also has built in library functions that facilitate application of software standards such as TCP/IP, activeX, etc. This is a powerful and flexible software. By using the method, the virtual instrument can be conveniently built, and the graphical interface of the virtual instrument enables the programming and the using process to be vivid and interesting.
The virtual instrument (virtual instrument) is a computer-based instrument. The close combination of computers and instruments is an important direction of instrument development at present. Roughly speaking, there are two ways to combine this, one is to load a computer into an instrument, a typical example of which is a so-called intelligent instrument. With the increasing functions and the shrinking of the volumes of computers, such instruments are becoming more and more powerful, and instruments with embedded systems are now being presented. Another way is to load the instrument into a computer. Based on general computer hardware and operation system, various instrument functions are realized.
As shown in fig. 1 to 7, the invention provides an automatic identification and switching device for a mid-infrared ultrashort pulse laser state, which comprises a pumping source 1, a mode-locked laser, a spectrometer 9 and a control terminal 10;
the mode-locked laser consists of a beam combiner 2, a gain optical fiber 3, a single-mode optical fiber 4, an output coupler 5, an isolator 6, a polarizer 7 and an electric polarization controller 8;
the pump input arm of the beam combiner 2 is connected with the output end of the pump source 1 through a passive optical fiber, one end of the gain optical fiber 3 is connected with the signal output arm of the beam combiner 2, one end of the single-mode optical fiber 4 is connected with the other end of the gain optical fiber 3, one input arm of the output coupler 5 is connected with the other end of the single-mode optical fiber 4, one output arm of the output coupler is connected with the input end of the spectrometer 9, the input end of the isolator 6 is connected with the other output arm of the output coupler 5, the input end of the polarizer 7 is connected with the output end of the isolator 6, the input end of the electric polarization controller 8 is connected with the output end of the polarizer 7, and the output end of the electric polarization controller is connected with the signal input arm of the beam combiner 2;
the control terminal 10 is respectively connected with the pump source 1, the electric polarization controller 8 and the spectrometer 9.
Preferably, the beam combiner 2 is a (2+1) ×1 high-power multimode pump+signal beam combiner, and the working wavelength of the beam combiner is 1960-2020nm;
as a preferred mode, the gain fiber 3 adopts 2m thulium doped gain fiber, and the type of the thulium doped gain fiber is 4% doped concentration thulium fiber;
preferably, the single-mode optical fiber 4 is an SMF28e optical fiber;
as a preferred mode, the output coupler 5 is a 2×2 broadband fiber coupler, and the coupling ratio is 2000±200nm,90:10, and the SM2000 fiber;
preferably, the polarizer 7 is a coaxial optical fiber polarizer, 2000+/-50 nm, and an SM/SM tail fiber;
preferably, the model of the isolator 6 is IO-F-2000, 2000nm, single mode, 10W;
preferably, the control terminal 10 is a computer, the CPU of the computer is AMD Ryzen 7 4800h, the computer has a 16G running memory, and a LabVIEW 2018 platform is configured; the LabVIEW platform provides communication interfaces with the electric polarization controller 8, the pump source 1 in the mode-locked laser and the spectrometer 9, and an optimization algorithm framework and a program framework are built in the LabVIEW platform, wherein the method comprises the following steps: a target pulse setting window, a control parameter display window and an output laser measurement window. The LabVIEW platform is connected with the electric polarization controller 8 by a USB A type to Micro USB B type cable, the LabVIEW platform is connected with the spectrometer 9 by a USB A type to GP-IB type cable, and the LabVIEW platform is connected with the pump source 1 in the mode-locked laser by a USB A type to RS232 type cable.
The measurable range of the spectrometer 9 is 1200 nm-2400 nm, the optical power of-70 dBm can be measured, 55dB dynamic measurement under the resolution of 0.05nm can be realized, the maximum scanning speed under the span of 100nm is 0.5s, and a logarithmic coordinate mode is selected; the spectrometer 9 is provided with GPIB, RS232 and Ethernet interfaces, and supports an external PC to carry out remote control and data acquisition. As a preferable mode, the spectrometer 9 is AQ6375, 1200 nm-2400 nm, +20dBm-70 dBm, high wavelength resolution (0.05 nm) and large dynamic range (55 dB), and is provided with GP-IB, RS-232 and Ethernet (10/100 Base-T) interfaces for connecting an external PC for remote control and constructing an automatic test system.
Preferably, the electric polarization controller is a pulp structure-based polarization controller, model MPC320, forOptical fiber of sheath, its winding disk external diameter +.>Three paddles, each of which can be rotated 170 deg., with a minimum step size of 0.12 deg., can cover any polarization state on the bungjia sphere. The electro-polarization controller is based on the principle of stress induced birefringence, changing the polarization state of light passing through a single mode fiber. By arranging the optical fibres in two or three separate waysOn the standing winding disc, two or three independent wave plates (optical fiber retarders) are formed, which effectively operate at wavelengths of 300nm to 2100nm.
Preferably, the pump source 1 is a 793nm semiconductor laser, the length of an output tail fiber connected with the pump source is 1.5m, the pump power output is controlled by supporting an external level signal, the adjustable power range is 0-30W, and the maximum output power is 30W.
In the invention, through the arrangement of the beam combiner, the pump light and the signal light can be coupled into the same optical fiber; through the arrangement of the gain fiber, the laser of the micrometer wave band can be excited after the light emitted by the pumping source is received; by arranging the output coupler, light in the optical fiber annular cavity can be divided into two beams, one beam is output to facilitate the spectrometer to collect spectrum data, and the other beam of light continues to perform continuous oscillation feedback in the cavity; through the arrangement of the isolator and the polarizer, the pump light and the signal light in the optical fiber annular cavity can be transmitted in one direction. By setting the electric polarization controller, the saturable absorption effect caused by nonlinear polarization rotation in the optical fiber annular cavity can be changed, and then a stable mode locking effect can be realized. The spectrometer is connected with one output arm of the output coupler, so that the output signal can be conveniently sampled to obtain spectrum data, and meanwhile, the spectrometer is connected with the control terminal, so that the spectrometer can conveniently send the obtained spectrum data to the control terminal for data processing. The control terminal is further connected with the pump source and the electric polarization controller respectively, so that the control terminal can conveniently issue control parameters to the pump source and the electric polarization controller to execute, and can conveniently realize the switching of noise-like mode locking pulse and soliton mode locking pulse by controlling the electric polarization controller. The device has compact structure, can realize the output of ultrashort pulse laser with high power, high beam quality, high efficiency and high stability, and simultaneously can be convenient for realizing the automatic adjustment and switching of noise-like mode locking pulse and soliton mode locking pulse.
The invention also provides a method for automatically identifying and switching the state of the mid-infrared ultrashort pulse laser, which comprises the following specific embodiments:
example 1:
the method utilizes a device for automatically identifying and switching the state of the middle infrared ultrashort pulse laser to realize the identification and control of noise-like pulses in a 2-micrometer-band fiber laser, wherein a thulium-doped fiber laser based on nonlinear polarization rotation is used as a control object, and specifically comprises the following steps:
step one: selecting a target pulse on a LabVIEW platform in the control terminal 10, and controlling a mode-locked laser to start working;
the LabVIEW platform block diagram is shown in figure 2, optional noise-like mode locking pulse or soliton mode locking pulse is shown, specifically, a target pulse type, a target pulse center wavelength range and a target pulse spectrum half-width range are set, if the noise-like mode locking pulse is selected, a 3dB bandwidth optional range under a logarithmic coordinate of a target pulse spectrum is 10-30nm, if the soliton mode locking pulse is selected, a 3dB bandwidth optional range under the logarithmic coordinate of the target pulse spectrum is 4-10nm, a reference spectrum signal is generated according to a set parameter, as shown in figure 2, in the embodiment, the target pulse type is set as noise-like mode locking, the expected center wavelength is 1972nm, the 3dB bandwidth under the logarithmic coordinate is 20nm, the spectral resolution is set to be 0.05nm, and the single scanning range is 100nm;
Step two: receiving an output signal of the mode-locked laser by utilizing a spectrometer 9, and sending the spectrum data obtained by sampling to a LabVIEW platform in a control terminal 10;
the output signal of the mode-locked laser has a wavelength range of 1800-2940nm, the repetition frequency of pulse is 10kHz-1GHz, and the pulse width range is 100fs-50 ps.
Step three: the control terminal 10 uses a trained neural network model on the LabVIEW platform to perform mode locking state identification on the spectrum data, calculates the matching degree of the measured pulse spectrum and the target spectrum, and if the matching degree of the output laser pulse state and the set target pulse state is more than 80%, the output laser pulse state and the set target pulse state are considered to be consistent, and the fifth step is directly executed; if the matching degree of the output laser pulse state and the set target pulse state is less than or equal to 80%, the output laser pulse state is not consistent with the set target pulse state, and the fourth step is directly executed;
the neural network model has total 1000 sample training, which are all spectrum data, 600 simulation data and 400 experimental data, wherein the noise-like mode locking pulse data quantity is 300, the soliton mode locking pulse data quantity is 300, and the rest is non-mode locking state data. The neural network model adopts convolutional neural network training, 900 data is selected as a training set, 100 data is selected as a testing set, and the accuracy is 99.8% and 97% respectively.
Step four: the control terminal 10 generates corresponding control parameters according to a genetic optimization algorithm, and sends the control parameters to the pump source 1 and the electric polarization controller 8 through a serial port communication protocol, wherein the control parameters comprise the pump power of the pump source 1 and the blade angle of the electric polarization controller 8;
the specific method of the genetic optimization algorithm is as follows:
s1: the control terminal 10 randomly generates 20 groups of initial control parameters and sends the initial control parameters to the pump source 1 and the electric polarization controller 8 respectively, so that the pump source 1 and the electric polarization controller 8 execute the received control parameters;
s2: calculating, namely calculating the matching degree of the spectrum data corresponding to the 20 groups of initial control parameters in the S1 and the target pulse by using the control terminal 10;
s3: sorting, namely sorting 20 groups of initial parameters according to the matching degree by using the control terminal 10;
s4: selecting, by using the control terminal 10, the top 6 groups of control parameters with the highest matching degree ranking;
s5: inheritance, 6 groups of parameters are randomly crossed and combined by the control terminal 10 to generate new 10 groups of control parameters;
s6: a step of randomly selecting 4 groups of control parameters by using the control terminal 10 from the newly generated control parameters in the step S5, and randomly modifying the control parameter values in the control parameters to generate new 20 groups of control parameters;
S7: the newly generated 20 groups of control parameters are sent to the pump source 1 and the electric polarization controller 8 through the control terminal 10;
step five: if the output laser pulse state accords with the target pulse, pulse state monitoring is carried out, wherein the pulse state monitoring means that the current pulse state is compared with the set target pulse state at regular intervals, and whether the current output accords with the set or not is judged; if the output laser state is suddenly changed and is not consistent with the target pulse, jumping to the third step, and searching the target state again;
step six: if the control terminal 10 switches the target pulse, the step two is executed in a jumping manner;
in the third step, the neural network model has a total of 1000 sample training, and is spectral data, wherein 600 are simulation data, 400 are experimental data, wherein the number of noise-like mode locking pulse data is 300, the number of soliton mode locking pulse data is 300, and the rest is non-mode locking state data.
The final output pulse spectrum measurement signal is shown in figure 4, the corresponding control parameter is pumping power 9.5W, the three blade angles of the electric polarization controller are 7.9 degrees, 16.0 degrees and 6.0 degrees respectively, and the matching degree calculated by the neural network model is 95.6 percent, as shown in figure 2;
The oscilloscope is used for verifying the time domain state of the output pulse, the output laser is input into a high-speed photoelectric sensor, and the high-speed photoelectric sensor transmits signals to the oscilloscope end, as shown in figure 5.
Example 2:
the identification and control of soliton-like pulses in a 2-micrometer-band fiber laser are realized by using a device for automatically identifying and switching the state of a middle infrared ultrashort pulse laser, wherein a thulium-doped fiber laser based on nonlinear polarization rotation is used as a control object, and the method specifically comprises the following steps:
step one: selecting a target pulse on a LabVIEW platform in the control terminal 10, and controlling a mode-locked laser to start working;
the LabVIEW platform block diagram is shown in figure 2, optional noise-like mode locking pulse or soliton mode locking pulse is shown, specifically, a target pulse type, a target pulse center wavelength range and a target pulse spectrum half-width range are set, if the noise-like mode locking pulse is selected, a 3dB bandwidth optional range under a logarithmic coordinate of a target pulse spectrum is 10-30nm, if the soliton mode locking pulse is selected, a 3dB bandwidth optional range under the logarithmic coordinate of the target pulse spectrum is 4-10nm, a reference spectrum signal is generated according to a set parameter, as shown in figure 5, in the embodiment, the target pulse type is soliton mode locking, the expected center wavelength is 1960nm, the 3dB bandwidth under the logarithmic coordinate is 6nm, the spectral resolution is set to be 0.05nm, and the single scanning range is 100nm;
Step two: receiving an output signal of the mode-locked laser by utilizing a spectrometer 9, and sending the spectrum data obtained by sampling to a LabVIEW platform in a control terminal 10;
the output signal of the mode-locked laser has a wavelength range of 1800-2940nm, the repetition frequency of pulse is 10kHz-1GHz, and the pulse width range is 100fs-50 ps.
Step three: the control terminal 10 uses a trained neural network model on the LabVIEW platform to perform mode locking state identification on the spectrum data, calculates the matching degree of the measured pulse spectrum and the target spectrum, and if the matching degree of the output laser pulse state and the set target pulse state is more than 80%, the output laser pulse state and the set target pulse state are considered to be consistent, and the fifth step is directly executed; if the matching degree of the output laser pulse state and the set target pulse state is less than or equal to 80%, the output laser pulse state is not consistent with the set target pulse state, and the fourth step is directly executed;
the neural network model has total 1000 sample training, which are all spectrum data, 600 simulation data and 400 experimental data, wherein the noise-like mode locking pulse data quantity is 300, the soliton mode locking pulse data quantity is 300, and the rest is non-mode locking state data. The neural network model adopts convolutional neural network training, 900 data is selected as a training set, 100 data is selected as a testing set, and the accuracy is 99.8% and 97% respectively.
Step four: the control terminal 10 generates corresponding control parameters according to a genetic optimization algorithm, and sends the control parameters to the pump source 1 and the electric polarization controller 8 through a serial port communication protocol, wherein the control parameters comprise the pump power of the pump source 1 and the blade angle of the electric polarization controller 8;
the specific method of the genetic optimization algorithm is as follows:
s1: the control terminal 10 randomly generates 20 groups of initial control parameters and sends the initial control parameters to the pump source 1 and the electric polarization controller 8 respectively, so that the pump source 1 and the electric polarization controller 8 execute the received control parameters;
s2: calculating, namely calculating the matching degree of the spectrum data corresponding to the 20 groups of initial control parameters in the S1 and the target pulse by using the control terminal 10;
s3: sorting, namely sorting 20 groups of initial parameters according to the matching degree by using the control terminal 10;
s4: selecting, by using the control terminal 10, the top 6 groups of control parameters with the highest matching degree ranking;
s5: inheritance, 6 groups of parameters are randomly crossed and combined by the control terminal 10 to generate new 10 groups of control parameters;
s6: a step of randomly selecting 4 groups of control parameters by using the control terminal 10 from the newly generated control parameters in the step S5, and randomly modifying the control parameter values in the control parameters to generate new 20 groups of control parameters;
S7: the newly generated 20 groups of control parameters are sent to the pump source 1 and the electric polarization controller 8 through the control terminal 10;
step five: if the output laser pulse state accords with the target pulse, pulse state monitoring is carried out, wherein the pulse state monitoring means that the current pulse state is compared with the set target pulse state at regular intervals, and whether the current output accords with the set or not is judged; if the output laser state is suddenly changed and is not consistent with the target pulse, jumping to the third step, and searching the target state again;
step six: if the control terminal 10 switches the target pulse, the step two is executed in a jumping manner;
in the third step, the neural network model has a total of 1000 sample training, and is spectral data, wherein 600 are simulation data, 400 are experimental data, wherein the number of noise-like mode locking pulse data is 300, the number of soliton mode locking pulse data is 300, and the rest is non-mode locking state data.
The final output pulse spectrum measurement signal is shown in figure 6, the corresponding control parameter is pumping power 5W, the three blade angles of the electric polarization controller are 86.0 degrees, 65.8 degrees and 71.5 degrees respectively, and the matching degree calculated by the neural network model is 86.5%, as shown in figure 5;
The oscilloscope is used for verifying the time domain state of the output pulse, the output laser is input into a high-speed photoelectric sensor, and the high-speed photoelectric sensor transmits signals to the oscilloscope end, as shown in fig. 7.
The invention uses the electric polarization controller as a polarization control device in the mode-locked laser, uses the spectrometer as a measuring device and receives the output signal of the mode-locked laser, simultaneously, carries out mode-locked state identification on the sampled spectrum data through a neural network model on a control terminal, then determines whether to carry out optimization processing according to the matching degree of the output laser pulse state and the set target pulse state, adopts a genetic algorithm in the optimization processing process, can effectively ensure the stability of the output signal of the laser in the target state through a closed loop feedback structure and effective state identification, and can realize the switching of noise-like mode-locked pulse and soliton mode-locked pulse through controlling the electric polarization controller. Because the noise-like mode locking pulse and the soliton mode locking pulse are highly consistent in time domain characteristics, the noise-like mode locking pulse and the soliton mode locking pulse are difficult to effectively distinguish through an oscilloscope or a photoelectric detector, the invention can automatically realize accurate and efficient identification of the noise-like pulse and the soliton pulse state through the reinforced training of the spectrum state of the ultrafast laser pulse by the convolutional neural network, solves the problem of misjudgment of the states in the application of the mid-infrared ultrafast fiber laser pulse, and realizes automatic adjustment and switching of the two pulse states. The neural network training algorithm adopted by the invention has the advantages of high identification speed, high identification precision, intellectualization and the like, and the reliability of the mid-infrared ultrafast fiber laser in industrial application is remarkably improved. The genetic algorithm adopted intelligently controls the laser pulse state, realizes the rapid and accurate switching of multiple laser states of complex noise-like pulse states and soliton states, and compared with the existing manual control method, the obtained laser state is more efficient and the cost of manual control is greatly reduced. Compared with the existing mid-infrared ultrafast laser system, the invention only needs to adopt a common commercial single-mode fiber instead of the polarization maintaining fiber with high cost adopted by the main stream method, thereby greatly reducing the cost. The intelligent identification and control method adopted by the invention also omits a complicated temperature control system, a mechanical vibration control system and an environment disturbance prevention control system, can be quickly adjusted according to the environment change, and greatly enhances the environment adaptability. By adopting a full-automatic intelligent control system, various pulse-shaped outputs such as various noise types, solitons, harmonic mode locking solitons and the like can be realized, and various customized laser pulse outputs can be generated according to the requirements of customers, so that the application range of the laser system is greatly increased, and the high-precision laser processing problems of materials with complex structures can be solved by rapidly switching pulses in different states.
Claims (9)
1. The automatic identification and switching device for the state of the mid-infrared ultrashort pulse laser comprises a pumping source (1) and is characterized by further comprising a mode-locked laser, a spectrometer (9) and a control terminal (10);
the mode-locked laser consists of a beam combiner (2), a gain optical fiber (3), a single-mode optical fiber (4), an output coupler (5), an isolator (6), a polarizer (7) and an electric polarization controller (8);
the pump input arm of the beam combiner (2) is connected with the output end of the pump source (1) through a passive optical fiber, one end of the gain optical fiber (3) is connected with the signal output arm of the beam combiner (2), one end of the single-mode optical fiber (4) is connected with the other end of the gain optical fiber (3), one input arm of the output coupler (5) is connected with the other end of the single-mode optical fiber (4), one output arm of the output coupler is connected with the input end of the spectrometer (9), the input end of the isolator (6) is connected with the other output arm of the output coupler (5), the input end of the polarizer (7) is connected with the output end of the isolator (6), and the input end of the electric polarization controller (8) is connected with the output end of the polarizer (7) and the signal input arm of the beam combiner (2);
the control terminal (10) is respectively connected with the pump source (1), the electric polarization controller (8) and the spectrometer (9).
2. The automatic status recognition and switching device for the mid-infrared ultrashort pulse laser according to claim 1, wherein the control terminal (10) is a computer.
3. The automatic identification and switching device for the state of the mid-infrared ultrashort pulse laser according to claim 1, wherein the measurable range of the spectrometer (9) is 1200 nm-2400 nm, the light power of-70 dBm can be measured, 55dB dynamic measurement under the resolution of 0.05nm can be realized, the maximum scanning speed under the span of 100nm is 0.5s, and a logarithmic coordinate mode is selected; the spectrometer (9) is provided with GPIB, RS232 and an Ethernet interface, and supports an external PC to carry out remote control and data acquisition.
4. The automatic identification and switching device for the state of the mid-infrared ultrashort pulse laser according to claim 1, wherein the electric polarization controller is a polarization controller based on a pulp structure, the outer diameter of a winding disc is 18mm, each pulp can rotate 170 degrees, the minimum stepping size is 0.12 degrees, and any polarization state on a Pongary ball can be covered.
5. The automatic identification and switching device for the state of the mid-infrared ultrashort pulse laser according to claim 1, wherein the pumping source (1) supports an external level signal to control the output of pumping power, and the adjustable power range is 0-30W.
6. The automatic identification and switching device for the state of the mid-infrared ultrashort pulse laser according to claim 1, wherein the gain fiber (3) is a thulium doped gain fiber.
7. The method for automatically identifying and switching the state of the mid-infrared ultrashort pulse laser is characterized by comprising the following steps of:
step one: selecting a target pulse through a control terminal (10) and controlling the mode-locked laser to start working;
step two: receiving an output signal of the mode-locked laser by using a spectrometer (9), and transmitting the spectrum data obtained by sampling to a control terminal (10);
step three: the control terminal (10) uses a trained neural network model to carry out mode locking state identification on spectrum data, calculates the matching degree of a measured pulse spectrum and a target spectrum, considers the matching degree of an output laser pulse state and a set target pulse state to be consistent if the matching degree of the output laser pulse state and the set target pulse state is more than 80%, and directly executes the step five; if the matching degree of the output laser pulse state and the set target pulse state is less than or equal to 80%, the output laser pulse state is not consistent with the set target pulse state, and the fourth step is directly executed;
step four: the control terminal (10) generates corresponding control parameters according to a genetic optimization algorithm, and sends the control parameters to the pump source (1) and the electric polarization controller (8) through a serial port communication protocol, wherein the control parameters comprise the pump power of the pump source (1) and the blade angle of the electric polarization controller (8);
The specific method of the genetic optimization algorithm is as follows:
s1: the control terminal (10) randomly generates 20 groups of initial control parameters and sends the initial control parameters to the pump source (1) and the electric polarization controller (8) respectively, so that the pump source (1) and the electric polarization controller (8) execute the received control parameters;
s2: calculating, namely calculating the matching degree of the spectrum data corresponding to 20 groups of initial control parameters in the S1 and the target pulse by using the control terminal (10);
s3: sorting, namely sorting 20 groups of initial parameters according to the matching degree by using a control terminal (10);
s4: selecting, by using the control terminal (10), the top 6 groups of control parameters with highest matching degree ranking;
s5: inheritance, 6 groups of parameters are randomly and alternately combined by using a control terminal (10), and a new 10 groups of control parameters are generated;
s6: a step of mutating, in the newly generated control parameters in the step S5, randomly selecting 4 groups of control parameters by using the control terminal (10), and randomly modifying the control parameter values in the 4 groups of control parameters to generate new 20 groups of control parameters;
s7: the newly generated 20 groups of control parameters are sent to the pump source (1) and the electric polarization controller (8) through the control terminal (10);
step five: if the output laser pulse state accords with the target pulse, pulse state monitoring is carried out, wherein the pulse state monitoring means that the current pulse state is compared with the set target pulse state at regular intervals, and whether the current output accords with the set or not is judged; if the output laser state is suddenly changed and is not consistent with the target pulse, jumping to the third step, and searching the target state again;
Step six: if the control terminal (10) switches the target pulse, the step II is executed in a jumping mode.
8. The method for automatically identifying and switching states of a mid-infrared ultrashort pulse laser according to claim 7, wherein in the third step, the neural network model has a total of 1000 samples for training and is spectral data, 600 are simulation data, 400 are experimental data, the number of noise-like mode locking pulse data is 300, the number of soliton mode locking pulse data is 300, and the rest is non-mode locking state data.
9. The method of claim 8, wherein in step one, the target pulse is a selected noise-like mode-locked pulse or soliton mode-locked pulse.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310396047.3A CN116435861A (en) | 2023-04-14 | 2023-04-14 | Device and method for automatically identifying and switching state of middle-infrared ultrashort pulse laser |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310396047.3A CN116435861A (en) | 2023-04-14 | 2023-04-14 | Device and method for automatically identifying and switching state of middle-infrared ultrashort pulse laser |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116435861A true CN116435861A (en) | 2023-07-14 |
Family
ID=87088629
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310396047.3A Pending CN116435861A (en) | 2023-04-14 | 2023-04-14 | Device and method for automatically identifying and switching state of middle-infrared ultrashort pulse laser |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116435861A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117095806B (en) * | 2023-10-19 | 2023-12-15 | 邦士医疗科技股份有限公司 | Quality management system of medical equipment in operating room |
CN118137277A (en) * | 2024-05-06 | 2024-06-04 | 南京信息工程大学 | Rapid automatic mode locking method, system and equipment based on deep learning |
-
2023
- 2023-04-14 CN CN202310396047.3A patent/CN116435861A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117095806B (en) * | 2023-10-19 | 2023-12-15 | 邦士医疗科技股份有限公司 | Quality management system of medical equipment in operating room |
CN118137277A (en) * | 2024-05-06 | 2024-06-04 | 南京信息工程大学 | Rapid automatic mode locking method, system and equipment based on deep learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116435861A (en) | Device and method for automatically identifying and switching state of middle-infrared ultrashort pulse laser | |
Krzempek | Dissipative soliton resonances in all-fiber Er-Yb double clad figure-8 laser | |
Pu et al. | Automatic mode-locking fiber lasers: progress and perspectives | |
CN110829164B (en) | All-fiber ultrashort pulse light source capable of simultaneously generating soliton and noise-like pulses | |
CN106093599B (en) | Optical probe and electromagnetic field measuring equipment and measuring method thereof | |
CN108598856B (en) | Femtosecond laser | |
Li et al. | Self-starting mode-locked Tm-doped fiber laser using a hybrid structure of no core-graded index multimode fiber as the saturable absorber | |
CN113451869B (en) | Method for generating double-optical comb and multi-optical comb by single cavity | |
Luo et al. | Modulation instability induced by cross-phase modulation in a dual-wavelength dispersion-managed soliton fiber ring laser | |
Shao et al. | Cavity-assisted modulation instability lasing of a fiber ring laser | |
CN117374707A (en) | All-fiber device and method for generating three switchable near infrared broadband supercontinuum | |
CN112398545A (en) | Quantum signal modulation variance control and calibration method and system for CVQKD system | |
Huang et al. | Highly Birefringence‐Guided Microfiber Resonator for Ultra‐High Repetition Rate Ultrashort Pulse | |
CN208385830U (en) | Tunable single longitudinal mode optical-fiber laser, gas concentration detection apparatus | |
CN216624862U (en) | 1.5 micron wave band full polarization-maintaining bidirectional mode-locked fiber laser | |
CN106410586A (en) | High-power high-repetition-frequency mode-locking pulse optical fiber laser | |
US20220341760A1 (en) | System and method for arbitrary optical waveform generation | |
Spiess et al. | Chirped temporal solitons in driven optical resonators | |
Sulaiman et al. | Investigation of continuously adjustable extinction ratio in a multiwavelength SOA fiber laser based on intensity dependent transmission effect | |
CN114122877A (en) | 1.5 micron wave band full polarization-maintaining bidirectional mode-locked fiber laser | |
Peng et al. | Switchable multi-wavelength actively Q-switched erbium-doped fiber laser based on nonlinear polarization rotation and Sagnac filter | |
Mao et al. | Wavelength-tunable 10 GHz actively harmonic mode-locked fiber laser based on semiconductor optical amplifier | |
CN104733989A (en) | Dark pulse fiber laser | |
CN116885557B (en) | Vacuum compression photon generation system based on waveguide structure | |
CN220692518U (en) | Semiconductor saturable absorption mirror mode-locking fiber laser |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |