Optical fiber welding and cutting high-efficiency processing system and method based on intelligent algorithm

Abstract

The invention relates to the technical field of optical fiber communication, discloses an optical fiber welding and cutting high-efficiency processing system and method based on an intelligent algorithm, a computer readable storage medium and terminal equipment, and aims at solving the problems that a traditional optical fiber processing procedure is discrete, parameters depend on manpower, and environmental adaptability is weak. The system comprises a motion control module, a welding parameter optimization module, a parameter compensation module and an optional data interaction unit. The method executes the steps corresponding to the modules. The method can improve operation fluency and welding quality stability, enhance severe environment adaptability and reduce manual dependence, and is suitable for optical fiber construction and maintenance scenes.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of optical fiber communication technology, in particular to an optical fiber fusion cutting high-efficiency processing system based on intelligent algorithm, a corresponding processing method, a computer readable storage medium and a computer terminal device, which is suitable for high-precision and automatic processing of the whole process of optical fiber stripping, cutting and fusion in optical fiber line construction and maintenance scenes. BACKGROUND

[0002] In the construction and operation process of optical fiber communication network, optical fiber fusion and cutting are core processes, and their operation efficiency and quality directly determine the transmission performance of optical fiber link. The current optical fiber processing technology in the industry has the following key problems:

[0003] Traditional optical fiber processing needs to be completed in steps through independent stripping machines, cutting machines and fusion machines, and the optical fiber needs to be manually transferred between processes, which not only leads to long intervals between operations (the single-process connection time is usually more than 2s), but also easily causes optical fiber positioning errors (usually more than 0.05mm) due to manual operation deviation, ultimately affecting the fusion quality.

[0004] The parameters (such as discharge current and fusion time) of the existing fusion machine need to be manually set according to the optical fiber type (single-mode / multi-mode / special optical fiber) and experience, and lack dynamic optimization capability. When the optical fiber material or operating environment changes, the parameter adaptation lags behind, which easily leads to an increase in fusion loss (more than 0.05dB in some scenarios), or even fusion failure.

[0005] Optical fiber operation often faces harsh environments such as high temperature (≥40℃), high humidity (≥85% RH) and high dust, and the existing equipment lacks real-time environmental monitoring and parameter compensation mechanism, which easily leads to a decrease in cutting surface flatness (roughness Ra>0.3μm) or fusion parameter drift due to environmental interference, and the operation stability is less than 80%.

[0006] There is a lack of unified data interaction mechanism between stripping, cutting and fusion modules, and environmental data and position data cannot be shared in real time, which leads to disconnection between parameter adjustment and process control, further reducing the efficiency of the whole process.

[0007] In view of the above problems, there is an urgent need for an optical fiber fusion cutting system that can realize process integration, intelligent optimization of parameters and environmental self-adaptive compensation, to break through the technical bottlenecks of the existing technology. SUMMARY

[0008] The present application aims to overcome the defects of the existing optical fiber fusion cutting technology, such as "process dispersion, parameter manualization and weak environmental adaptability", and provides a high-efficiency processing system and method based on intelligent algorithm, which realizes automatic and high-precision control of the whole process of optical fiber stripping, cutting and fusion, and improves the real-time performance and environmental adaptability of parameter optimization.

[0009] To solve the above technical problems, the technical scheme adopted by the present application is:

[0010] An optical fiber fusion cutting high-efficiency processing system based on an intelligent algorithm, comprising:

[0011] A motion control module is used to integrally and coherently control the motion process of the optical fiber stripping, cutting and fusion processes, dynamically corrects errors based on preset error correction rules by collecting motion position data of each process in real time;

[0012] A fusion parameter optimization module is used to pre-construct a sample training library, train the data in the sample training library, and generate a fusion parameter optimization model; the fusion parameter optimization model obtains type data of the current optical fiber to be processed and current job environment data in real time, and automatically outputs the optimal fusion parameters based on the data;

[0013] A parameter compensation module is used to obtain environmental state parameters, synchronously interact the environmental state parameters to the fusion parameter optimization module, and call preset environmental parameter compensation rules to real-time correct the motion control parameters of the motion control module and the optimal fusion parameters output by the fusion parameter optimization module.

[0014] Further, as an improvement of the technical scheme of the present application, the motion control module comprises:

[0015] A position collection unit is used to collect motion position data of each process of optical fiber stripping, cutting and fusion in real time;

[0016] An error correction unit is built-in with preset error correction rules, calculates real-time deviation values according to the motion position data, and completes dynamic error correction when the deviation values exceed the preset threshold;

[0017] A process coordination unit is used to realize integrally and coherently control the optical fiber stripping, cutting and fusion processes.

[0018] Further, as an improvement of the technical scheme of the present application, the position collection unit comprises:

[0019] A multi-axis sensor assembly is configured with a plurality of displacement detection modules corresponding to the optical fiber stripping station, the cutting station and the fusion station respectively, and is used to synchronously collect motion coordinate data of each station;

[0020] A data processing subunit is built-in with a noise filtering algorithm and a dynamic calibration model, and can perform real-time noise reduction processing and zero-point drift compensation on the collected motion coordinate data, and output standardized position data;

[0021] A synchronous transmission subunit transmits the standardized position data to the error correction unit of the motion control module.

[0022] Further, as an improvement of the technical scheme of the present application, the error correction unit comprises:

[0023] a deviation calculation subunit, configured to receive real-time motion position data of each process, calculate a deviation value between an actual position and a theoretical reference position through a preset algorithm, and generate a deviation signal;

[0024] a correction rule storage subunit, internally provided with a multi-scene error correction model, wherein the model comprises differentiated correction coefficients based on the differences in fiber material and process type, and can call a matching correction strategy according to the deviation signal;

[0025] a dynamic execution subunit, configured to receive the deviation signal and the matching correction strategy, and drive an execution mechanism to complete motion parameter adjustment.

[0026] Further, as an improvement of the technical scheme of the present application, the fusion parameter optimization module comprises:

[0027] a sample training subunit, configured to construct a sample library comprising fiber type characteristics, environmental influence factors and corresponding fusion quality data, and perform model training by using a gradient boosting tree algorithm to generate a fusion parameter optimization model adapted to multiple scenes;

[0028] a model inference subunit, configured to receive type data of a current fiber to be processed and real-time monitoring data of a working environment in real time, call the fusion parameter optimization model to perform inference calculation, and generate a preliminary fusion parameter combination;

[0029] a parameter output subunit, configured to perform compliance verification on the preliminary fusion parameter combination, and finally output optimal fusion parameters comprising a discharge current, a fusion time and a preheating temperature.

[0030] Further, as an improvement of the technical scheme of the present application, the model inference subunit comprises:

[0031] a data receiving module, configured to synchronously acquire type characteristic data of a current fiber to be processed and real-time monitoring data of a working environment;

[0032] a scene matching module, internally provided with a scene classification model, capable of automatically identifying a current working scene type according to the received fiber type characteristic data and environmental monitoring data, and calling model parameter weights of a corresponding scene;

[0033] a real-time inference module, configured to load a gradient boosting tree optimization model pre-trained by the sample training subunit, combine the weight parameters output by the scene matching module, perform inference calculation on the received fiber type characteristic data and environmental monitoring data, and generate a preliminary fusion parameter combination comprising an initial value of a discharge current, a fusion time interval and a reference value of a preheating temperature;

[0034] An intermediate result output module is configured to combine the preliminary fusion parameters and transmit them to the parameter output subunit in a structured data format.

[0035] Further as an improvement of the technical scheme of the present application, the parameter compensation module comprises:

[0036] An environment sensing subunit is configured with a temperature sensor, a humidity sensor and a dust concentration sensor, and is configured to collect the environmental state parameters of the working environment in real time.

[0037] A rule storage subunit is internally provided with a multi-dimensional environmental parameter compensation model, which comprises a discharge current correction coefficient based on the environmental temperature, a fusion time compensation curve based on the environmental humidity and a motion precision correction weight based on the dust concentration, and the model parameters can be dynamically updated according to the historical operation data.

[0038] A parameter correction subunit is configured to receive the environmental state parameters and call the matched compensation model, and to collaboratively correct the stripping intensity, cutting speed motion parameters output by the motion control module, and the discharge current and preheating temperature fusion parameters output by the fusion parameter optimization module.

[0039] A data synchronization subunit is configured to feed back the corrected motion parameters and fusion parameters to the motion control module and the fusion parameter optimization module in real time, respectively.

[0040] An efficient processing method for optical fiber fusion and cutting based on intelligent algorithm, comprising the following steps:

[0041] The motion process of the optical fiber stripping, cutting and fusion processes is integrally and coherently controlled, the motion position data of each process is collected in real time, and dynamic error correction is performed based on a preset error correction rule.

[0042] A sample training library is constructed in advance, the data in the sample training library is subjected to model training, and a fusion parameter optimization model is generated; the fusion parameter optimization model acquires the type data of the current optical fiber to be processed and the current working environment data in real time, and automatically outputs the optimal fusion parameters based on the data.

[0043] The environmental state parameters are acquired, the environmental state parameters are synchronously and interactively fed to the fusion parameter optimization module, and a preset environmental parameter compensation rule is called to correct the motion control parameters of the motion control module and the optimal fusion parameters output by the fusion parameter optimization module in real time.

[0044] A computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the steps of the efficient processing method for optical fiber fusion and cutting based on intelligent algorithm.

[0045] A computer terminal device, characterized by comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the intelligent algorithm-based high-efficiency processing method for fiber fusion cutting when executing the computer program.

[0046] Compared with the prior art, the present application has the following beneficial effects:

[0047] The present application realizes integrated and coherent control of the whole process of fiber processing, breaks through the process barriers of traditional step-by-step operation, eliminates the need for manual transfer of optical fibers, reduces the operation interval and human intervention between processes, effectively reduces the positioning deviation risk caused by manual transfer, and significantly improves the smoothness and operation consistency of the operation process.

[0048] The present application relies on a machine learning-driven fusion parameter optimization mechanism, which can automatically adapt to the material, structural characteristics and real-time operating environment conditions of different types of optical fibers, without the need for manual experience-based setting or adjustment of fusion parameters, thereby avoiding fusion quality fluctuations caused by parameter adaptation lag or human setting deviation, and effectively ensuring the stability and reliability of fusion quality.

[0049] Through real-time environmental monitoring and dynamic parameter compensation design, the present application can actively perceive changes in temperature, humidity, dust and other changes in the operating environment, and make targeted corrections to motion control parameters and fusion parameters, thereby avoiding the adverse effects of harsh environments on cutting flatness and fusion stability, significantly improving the operation adaptability and work reliability of the system in complex outdoor or industrial environments, and reducing operation interruptions or quality problems caused by environmental factors.

[0050] The full-process automatic operation mode of the present application greatly reduces the dependence on the experience of operating personnel, simplifies the operation process, and reduces the number and intensity of human intervention, thereby not only reducing the probability of human operation errors, but also improving the operation amount per unit time, reducing the demand for professional operating personnel, and indirectly reducing operation cost and training cost.

[0051] The present application realizes real-time sharing of information and synchronous updating of parameters between modules through a cooperative data interaction mechanism, ensures that the motion control, parameter optimization and environmental compensation are highly matched in terms of action and parameter adjustment, avoids the problem of uncoordinated parameters caused by data discontinuity between modules, further ensures the stability and efficiency of the overall operation of the system, and improves the cooperativity of the whole process. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0053] Fig. 1 is a schematic diagram of the composition of a high-efficiency processing system for optical fiber fusion cutting based on an intelligent algorithm according to an embodiment of the present application.

[0054] Fig. 2 is a schematic diagram of the framework flow of a high-efficiency processing method for optical fiber fusion cutting based on an intelligent algorithm according to an embodiment of the present application.

[0055] Fig. 3 Fig. 3 is a schematic diagram of the composition of a computer terminal device according to an embodiment of the present application. DETAILED DESCRIPTION

[0056] The technical solutions in the embodiments of the present application will be described clearly and completely below. Obviously, the described embodiments only constitute some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0057] As shown in Fig. 1, the high-efficiency processing system for optical fiber fusion cutting based on an intelligent algorithm according to an embodiment of the present application comprises: Figs. 1-3

[0058] The system realizes high-efficiency processing of the whole process of optical fiber through the cooperative work of three core modules, specifically including:

[0059] The motion control module, as the "process execution core" of the system, is used for integrated and coherent control of the motion process of the optical fiber stripping, cutting and fusion processes, and solves the positioning deviation and efficiency problems of traditional step-by-step operation. It contains:

[0060] The position acquisition unit: 3 groups of multi-axis displacement detection modules corresponding to the stripping, cutting and fusion stations are configured, and a grating displacement sensor is used to realize synchronous acquisition of the motion coordinate data of each station at a frequency of not less than 1 kHz; a built-in noise filtering algorithm and a dynamic calibration model, specifically a Kalman filtering algorithm, are used to perform noise reduction and zero drift compensation on the original data, and the output standardized position data has an error of not more than 0.005 microns; the data is transmitted to the error correction unit through a high-speed serial bus, and the data transmission delay is not more than 10 milliseconds.

[0061] ​Error correction unit: after receiving the standardized data from the position acquisition unit, the deviation value between the actual position and the theoretical reference is calculated through a preset algorithm, which is a PID combined with a fuzzy control algorithm; a built-in multi-scene correction model is provided, which includes correction coefficients for differences in two materials (quartz optical fiber and plastic optical fiber) and three processes (stripping, cutting and fusion), and can call the matching correction strategy according to the deviation value; the driving executive mechanism completes parameter adjustment within 50 milliseconds, controls the position deviation within 0.01 millimeters, and the correction process does not interrupt the process.

[0062] Process coordination unit: through time sequence control logic, the connection of stripping, cutting and fusion processes is realized, the time interval is not more than 0.5 seconds, the integrated process of “starting the next process as soon as the last process is completed” is achieved, and the efficiency loss caused by manual intervention is avoided.

[0063] Fusion parameter optimization module, as the “intelligent decision core” of the system, realizes automatic adaptation of fusion parameters based on machine learning, and solves the stability problem of manual parameter setting. Its internal includes:

[0064] Sample training subunit: a sample library covering “optical fiber type characteristics - environmental influence factors - fusion quality data” is constructed, the sample library contains 5000 groups of data, including 2000 groups of single-mode optical fiber, 1500 groups of multi-mode optical fiber and 1500 groups of special optical fiber, and the data content includes the material and structure parameters of optical fiber, temperature and humidity environment parameters, and corresponding fusion loss and strength data; gradient boosting tree algorithm is used for model training, and the iteration number is more than 1000 times to generate a fusion parameter optimization model adapted to multiple scenes.

[0065] Model inference subunit: real-time acquisition of current optical fiber type data (such as single-mode quartz optical fiber with a diameter of 0.25 millimeters) and environmental data (such as temperature 25 degrees Celsius and relative humidity 60%) is realized through the data interaction unit; the scene classification model is called to identify the working scene, and after loading the corresponding weight parameters, the preliminary fusion parameter combination (such as discharge current 18 milliamperes, fusion time 1.2 seconds and preheating temperature 80 degrees Celsius) is generated through the pre-training model inference, and the inference time is not more than 50 milliseconds.

[0066] Parameter output subunit: the preliminary parameters are checked for compliance, the checking standard is to match the equipment hardware parameter range (such as the upper limit of discharge current 25 milliamperes), and the optimal fusion parameters are finally output to ensure that the fusion loss is not more than 0.02 decibels, and the total response time from data input to parameter output is not more than 100 milliseconds.

[0067] Parameter compensation module, as the "environmental adaptation core" of the system, improves the operation reliability in harsh environments through real-time environmental monitoring and parameter correction. It contains:

[0068] Environmental sensing subunit: temperature sensor, humidity sensor, dust concentration sensor, temperature sensor model DS18B20, temperature measurement accuracy ±0.5 Celsius; humidity sensor model SHT30, relative humidity measurement accuracy ±2%; dust concentration sensor model PMS5003, measurement accuracy ±10%; collect environmental state parameters at a frequency of not less than 10 hertz, the overall measurement error is not more than ±2%.

[0069] Rule storage subunit: built-in multi-dimensional environmental compensation model, including temperature-based discharge current correction coefficient (current down 0.5 mA per 10 degrees Celsius), humidity-based welding time compensation curve (time extension 0.1 second per 10% relative humidity), dust concentration-based motion accuracy correction weight; model parameters are updated every 30 days based on historical operation data, 1000 sets of historical operation data in harsh environments, to improve correction accuracy.

[0070] Parameter correction subunit: after receiving the environmental parameters, the matched compensation model is called to correct the motion control module's stripping force (force increased by 10% when dust concentration is greater than 5 mg per cubic meter), cutting speed (speed reduced by 20% at -20 degrees Celsius), and discharge current, preheating temperature of the welding parameter optimization module.

[0071] Data synchronization subunit: real-time feedback of corrected parameters to motion control module and welding parameter optimization module through CAN bus, data transmission delay not more than 30 milliseconds, ensuring synchronous updating of parameters in each module.

[0072] Data interaction unit: data interaction unit is an optional configuration, mainly to strengthen the collaborative work between modules, realize real-time data transmission between motion control module, welding parameter optimization module and parameter compensation module, support bidirectional interaction of position data, fiber type data, environmental data and corrected parameters, provide data support for collaborative work of each module.

[0073] The application also discloses a high-efficiency processing method for optical fiber fusion cutting based on intelligent algorithm, which is realized based on the above-mentioned system and specifically includes the following steps:

[0074] Motion control step

[0075] The motion control module is started, the motion coordinate data of stripping, cutting and welding stations is synchronously collected by the position collection unit, and after noise reduction and calibration, the motion coordinate data is transmitted to the error correction unit; the error correction unit calculates the position deviation, calls the corresponding correction strategy to drive the actuator to adjust, and simultaneously controls the connection of each process through the process cooperation unit to realize integrated and coherent operation.

[0076] Parameter optimization step

[0077] The welding parameter optimization module calls a pre-trained welding parameter optimization model, receives current optical fiber type data (such as optical fiber material, diameter) and initial operation environment data in real time, generates and outputs optimal welding parameters through model inference.

[0078] Parameter compensation step

[0079] The parameter compensation module collects real-time environmental state parameters (temperature, humidity, dust concentration) through the environment sensing subunit, synchronizes to the welding parameter optimization module; calls the environmental parameter compensation rule, and real-time corrects the motion control parameters (stripping force, cutting speed) and the optimal welding parameters (discharge current, welding time), and feeds back the corrected parameters to the corresponding module to ensure operation stability.

[0080] The application also discloses a computer readable storage medium and a computer terminal device.

[0081] The computer readable storage medium: a computer program is stored thereon, and the computer program is executed by a processor to realize the steps of the optical fiber fusion and cutting high-efficiency processing method based on the intelligent algorithm. The computer readable storage medium includes but is not limited to ROM, RAM, magnetic disk, and floppy disk.

[0082] The computer terminal device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to realize the steps of the optical fiber fusion and cutting high-efficiency processing method based on the intelligent algorithm. The memory of the device can be realized by any type of volatile or non-volatile storage device or their combination, including but not limited to magnetic disk, optical disk, EEPROM, EPROM, SRAM, ROM, magnetic memory, flash memory, and PROM. The memory of the device provides an environment for the operation system and the computer program stored therein. The communication interface of the device is a network interface, which is used for communication with external terminals through network connection. The computer program is executed by the processor to realize the steps of the optical fiber fusion and cutting high-efficiency processing method based on the intelligent algorithm.

[0083] Computer program product: including computer program / instruction, which is executed by processor to realize the intelligent algorithm-based optical fiber fusion cutting high-efficiency processing method steps.

[0084] Embodiment 1: system hardware configuration and parameters

[0085] Motion control module

[0086] Controller: STM32H743 microcontroller is selected, and the main frequency is 480 megahertz;

[0087] Position acquisition unit: 3 groups of grating displacement sensors, model RGH24Z, resolution 0.1 microns, acquisition frequency 1 kilohertz, data processing adopts Kalman filtering algorithm;

[0088] Error correction unit: PID fuzzy control algorithm is adopted, and the actuator is 42HS40 stepping motor, the step angle of the stepping motor is 1.8 degrees, the subdivision level is 16, the response time is 45 milliseconds, and the deviation is controlled to be 0.008 microns;

[0089] Process coordination unit: timing control logic is realized through microcontroller timer, and the process connection interval is 0.4 seconds.

[0090] Fusion parameter optimization module

[0091] Hardware carrier: NVIDIA Jetson Nano development board is adopted, and the GPU computing power is 472 GFLOPS;

[0092] Sample library: 5000 groups of data are included, covering 2000 groups of single-mode optical fibers, 1500 groups of multi-mode optical fibers and 1500 groups of special optical fibers, and the data content includes optical fiber material (quartz, plastic), diameter (0.125 microns to 0.9 microns), temperature (-20 degrees Celsius to 60 degrees Celsius), humidity (30% to 90%), and fusion loss (0.005 decibels to 0.1 decibels);

[0093] Model training: XGBoost algorithm is adopted, iteration number is 1200, learning rate is 0.05, and the inference time of the trained model is 40 milliseconds;

[0094] Parameter output: the optimal parameter response time is 80 milliseconds, and the fusion loss is stable at 0.012 decibels to 0.018 decibels.

[0095] Parameter compensation module

[0096] Sensors: Temperature sensor is DS18B20, measurement range -55 to 125 degrees Celsius, accuracy ±0.5 degrees Celsius; humidity sensor is SHT30, measurement range 0% to 100%, accuracy ±2%; dust sensor is PMS5003, measurement range 0 to 500 micrograms per cubic meter, accuracy ±10%;

[0097] Compensation model: adopt multiple linear regression model, temperature changes 10 degrees Celsius, discharge current correction ±0.5 mA; relative humidity changes 10%, fusion time correction ±0.1 seconds; when dust concentration is greater than 5 mg / m3, stripping intensity increases by 10%;

[0098] Data synchronization: transmit data through CAN bus, delay 25 ms, after correction, the system operating stability is 97% under the environment of -20 degrees Celsius, relative humidity 90%, dust concentration 8 mg / m3.

[0099] Data interaction unit: adopt EtherCAT bus, data transmission rate 100 Mbps, delay no more than 5 ms.

[0100] Example 2: system operation process

[0101] Initialization

[0102] After the system is powered on, the motion control module, the fusion parameter optimization module, the parameter compensation module self-check, and the data interaction unit establishes the inter-module communication connection; the parameter compensation module collects the initial environment data (e.g. temperature 25 degrees Celsius, relative humidity 60%, dust concentration 2 mg / m3) and transmits it to the fusion parameter optimization module.

[0103] Optical fiber clamping and identification

[0104] Clamp the optical fiber to be processed (e.g. single-mode quartz optical fiber with a diameter of 0.25 mm) to the system work station, the motion control module confirms the optical fiber positioning through the position acquisition unit, and the fusion parameter optimization module obtains the optical fiber type data.

[0105] Full-process processing

[0106] Stripping process: the motion control module drives the fiber stripping mechanism to act, the position acquisition unit monitors the stripping position in real time, and the error correction unit controls the deviation to be 0.008 mm, and the stripping time is 0.8 seconds;

[0107] Cutting process: start cutting 0.4 seconds after stripping is completed, cutting speed 1 mm / s (based on current environment data without correction), cutting surface roughness Ra is 0.08 microns;

[0108] The fusion procedure: the optimal parameters (discharge current 18.2 mA, fusion time 1.2 seconds, preheating temperature 82 degrees Celsius) are output by the fusion parameter optimization module, and the parameter compensation module does not need to be corrected based on the current environment. The detection loss is 0.015 decibels after fusion.

[0109] Job completion

[0110] The system outputs the job results (loss value, processing time), automatically resets to the initial state, and waits for the next job.

[0111] The present application realizes the integrated and coherent control of the whole process of optical fiber processing, breaks through the process barriers of traditional step-by-step operation, eliminates the need for manual transfer of optical fibers, reduces the operation interval and human intervention between processes, effectively reduces the positioning deviation risk caused by manual transfer, and significantly improves the smoothness and operation consistency of the operation process.

[0112] The present application relies on a machine learning driven fusion parameter optimization mechanism, which can automatically adapt to the material, structural characteristics and real-time operating environment conditions of different types of optical fibers, without the need for manual experience-based setting or adjustment of fusion parameters, avoiding fluctuations in fusion quality caused by parameter adaptation lag or human setting deviation, and effectively ensuring the stability and reliability of fusion quality.

[0113] The present application can actively perceive changes in temperature, humidity, dust and other changes in the operating environment through real-time environmental monitoring and dynamic parameter compensation design, and make targeted corrections to motion control parameters and fusion parameters, avoiding the adverse effects of harsh environments on cutting flatness and fusion stability, significantly improving the operation adaptability and work reliability of the system in complex outdoor or industrial environments, and reducing job interruptions or quality problems caused by environmental factors.

[0114] The full-process automatic operation mode of the present application greatly reduces the dependence on the experience of operating personnel, simplifies the operation process, and reduces the number and intensity of human involvement, not only reducing the probability of human operation errors, but also improving the amount of work per unit time, while reducing the demand for professional operators, indirectly reducing the cost of operation and training.

[0115] The present application realizes real-time information sharing and parameter synchronous updating among modules through a collaborative data interaction mechanism, ensures that the motion control, parameter optimization and environmental compensation are highly matched in terms of action and parameter adjustment, avoids the problem of parameter incoordination caused by data discontinuity between modules, further ensures the stability and efficiency of the overall operation of the system, and improves the collaboration of the whole process.

Claims

1. A smart algorithm based high performance processing system for fiber fusion splicing and cleaving, characterized by, Comprise: A motion control module for integrated and coherent control of the motion process of the optical fiber stripping, cutting and fusion process, real-time acquisition of motion position data of each process and dynamic error correction based on pre-set error correction rules; A fusion parameter optimization module for pre-constructing a sample training library, model training of data in the sample training library, and generation of a fusion parameter optimization model; the fusion parameter optimization model real-time acquires type data and current working environment data of the current optical fiber to be processed, and automatically outputs the optimal fusion parameters based on the data; A parameter compensation module for acquiring environmental state parameters, synchronously interacting the environmental state parameters to the fusion parameter optimization module, and calling a pre-set environmental parameter compensation rule to real-time correct the motion control parameters of the motion control module and the optimal fusion parameters output by the fusion parameter optimization module.

2. The system of claim 1, wherein, The motion control module comprises: A position acquisition unit for real-time acquisition of motion position data of each process of optical fiber stripping, cutting and fusion; An error correction unit with a pre-set error correction rule, which calculates real-time deviation value according to the motion position data, and completes dynamic error correction when the deviation value exceeds the pre-set threshold; A process coordination unit for integrated and coherent control of the optical fiber stripping, cutting and fusion process.

3. The system of claim 2, wherein, The position acquisition unit comprises: A multi-axis sensor assembly configured with a plurality of displacement detection modules corresponding to the optical fiber stripping station, the cutting station and the fusion station for synchronously acquiring motion coordinate data of each station; A data processing subunit with a noise filtering algorithm and a dynamic calibration model, which can real-time denoise and compensate for zero drift of the acquired motion coordinate data, and output standardized position data; A synchronous transmission subunit for transmitting the standardized position data to the error correction unit of the motion control module.

4. The system of claim 2, wherein, The error correction unit comprises: A deviation calculation subunit for receiving real-time motion position data of each process, calculating the deviation value of the actual position and the theoretical reference position by a pre-set algorithm, and generating a deviation signal; A correction rule storage subunit with a multi-scene error correction model, which contains differential correction coefficients based on the differences of optical fiber material and process type, and can call the matching correction strategy according to the deviation signal; A dynamic execution subunit for receiving the deviation signal and the matching correction strategy to drive the execution mechanism to complete the motion parameter adjustment.

5. The system of claim 1, wherein, The fusion parameter optimization module comprises: A sample training subunit for constructing a sample library containing optical fiber type characteristics, environmental influence factors and corresponding fusion quality data, and generating a fusion parameter optimization model adapted to multiple scenes by gradient boosting tree algorithm; A model inference subunit for real-time receiving type data of the current optical fiber to be processed and real-time monitoring data of the working environment, calling the fusion parameter optimization model for inference calculation, and generating a preliminary fusion parameter combination; A parameter output subunit for verifying the compliance of the preliminary fusion parameter combination and finally outputting the optimal fusion parameters including discharge current, fusion time and pre-heating temperature.

6. The system of claim 5, wherein, The model inference subunit comprises: The data receiving module is configured to synchronously acquire type characteristic data of the current optical fiber to be processed and real-time monitoring data of the working environment. The scene matching module is internally provided with a scene classification model, which can automatically identify the type of the current working scene according to the received type characteristic data of the optical fiber and the environmental monitoring data, and call the model parameter weight of the corresponding scene. The real-time inference module loads the gradient boosting tree optimization model pre-trained by the sample training subunit, combines the weight parameter output by the scene matching module, and performs inference calculation on the received type characteristic data of the optical fiber and the environmental monitoring data to generate a preliminary fusion parameter combination including the initial value of the discharge current, the fusion time interval and the preheating temperature reference value. The intermediate result output module is configured to transmit the preliminary fusion parameter combination to the parameter output subunit in a structured data format.

7. The system of claim 1, wherein, The parameter compensation module comprises: The environmental sensing subunit is configured with a temperature sensor, a humidity sensor and a dust concentration sensor, and is configured to collect environmental state parameters of the working environment in real time. The rule storage subunit is internally provided with a multi-dimensional environmental parameter compensation model, which comprises a discharge current correction coefficient based on environmental temperature, a fusion time compensation curve based on environmental humidity and a motion precision correction weight based on dust concentration, and the model parameters can be dynamically updated according to historical working data. The parameter correction subunit is configured to receive the environmental state parameters and call the matched compensation model to cooperatively correct the stripping force, the cutting speed motion parameter output by the motion control module and the discharge current, the preheating temperature fusion parameter output by the fusion parameter optimization module. The data synchronization subunit is configured to real-time feedback the corrected motion parameters and fusion parameters to the motion control module and the fusion parameter optimization module, respectively.

8. A high-efficiency processing method for fiber fusion cutting based on an intelligent algorithm, characterized in that, The method comprises the following steps: The motion process of the optical fiber stripping, cutting and fusion processes is integrally and coherently controlled, the motion position data of each process is collected in real time, and dynamic error correction is performed based on a preset error correction rule; A sample training library is pre-constructed, the data in the sample training library is subjected to model training, and a fusion parameter optimization model is generated; the fusion parameter optimization model acquires type data of the current optical fiber to be processed and current working environment data in real time, and automatically outputs adaptive optimal fusion parameters based on the data; The environmental state parameters are acquired, the environmental state parameters are synchronously interacted to the fusion parameter optimization module, and a preset environmental parameter compensation rule is called to real-time correct the motion control parameters of the motion control module and the optimal fusion parameters output by the fusion parameter optimization module.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the intelligent algorithm-based optical fiber fusion and cutting high-efficiency processing method in claim 8.

10. A computer terminal device, characterized by The computer program is executed by the processor to realize the steps of the intelligent algorithm-based optical fiber fusion and cutting high-efficiency processing method in claim 8. The computer program is executed by the processor to realize the steps of the intelligent algorithm-based optical fiber fusion and cutting high-efficiency processing method in claim 8.

Images