Fiber optic light source control method and system based on parameter optimization

Abstract

A fiber optic light source control method and system based on parameter optimization, belonging to the field of fiber optic light source control technology, which finds out the basic change points in the parameters by obtaining peak and valley points, and calculates the interference value amplitude and critical amplitude of the peak and valley points, and finally identifies the peak and valley points with not low interference value amplitude and not high critical amplitude as interference value points that are often to be deleted, so the accuracy of deleting interference value points is greatly improved by combining the interference value amplitude and the critical amplitude, and finally improves the accuracy of the collected light intensity parameters of the fiber optic light source or the ambient temperature value of the fiber optic light source.

Description

Technical Field

[0001] The present invention belongs to the technical field of optical fiber light source control, and in particular relates to an optical fiber light source control method and system based on parameter optimization. Background Art

[0002] Fiber optic light source devices are often core devices in the fields of medical treatment, industrial detection, sensing, optical lighting, etc. Their function is to convert electrical signals into optical signals.

[0003] For fiber optic light sources, the control structure of the fiber optic light source of the prior art solution with patent publication number "CN101377420A" is often used. It includes an optical signal acquisition circuit and a temperature sensor connected to a microprocessor. The temperature sensor is used to collect the ambient temperature value of the fiber optic light source and transmit it to the microprocessor. The microprocessor is used to calculate the voltage value parameter used to drive the laser driving circuit based on the collected ambient temperature value of the fiber optic light source. The optical signal acquisition circuit is also connected to the C-end of the monitoring detector. The B-end of the monitoring detector is the output end of the light source. The optical signal acquisition circuit is used to collect the photovoltage of the C-end of the monitoring detector and transmit it to the microprocessor as a light intensity parameter of the fiber optic light source. The ambient temperature value or photovoltage of the fiber optic light source is the fiber optic light source parameter.

[0004] However, during the acquisition of fiber optic light source parameters, the nonlinear error of the temperature sensor or the optical signal acquisition circuit is a common factor that is detrimental to the fiber optic light source parameters, which often causes the optical intensity parameters of the fiber optic light source or the ambient temperature value of the fiber optic light source to be inaccurate, thereby making the control performance of the fiber optic light source poor. Therefore, when the microprocessor collects the fiber optic light source parameters, the nonlinear error must be removed first. The fiber optic light source parameters generally contain various interference values ​​caused by external electric or magnetic fields, which is not conducive to the removal of nonlinear errors. Although the current filtering interference values, removing outliers and standardization methods can partially optimize the performance of fiber optic light source parameters, they cannot ensure that all interference values ​​are completely removed, especially some suspected outliers often reflect the normal method or behavior of the fiber optic light source parameters. Direct deletion is often not conducive to the final level of the fiber optic light source parameters and the effective operation of the fiber optic light source control system. Summary of the invention

[0005] In order to solve the defects in the prior art, the present invention proposes a fiber optic light source control method and system based on parameter optimization, which finds out the basic change points in the parameters by obtaining the peak and valley points, and calculates the interference value amplitude and the critical amplitude of the peak and valley points, and finally identifies the peak and valley points with not low interference value amplitude and not high critical amplitude as interference value points that are often to be deleted. Therefore, the accuracy of deleting interference value points is greatly improved by combining the interference value amplitude and the critical amplitude, and finally improves the accuracy of the collected light intensity parameters of the fiber optic light source or the ambient temperature value of the fiber optic light source.

[0006] The present invention utilizes the following technical solutions.

[0007] A method for controlling an optical fiber light source based on parameter optimization, comprising:

[0008] The temperature sensor collects the ambient temperature value of the optical fiber light source and transmits it to the microprocessor. The microprocessor processes the collected ambient temperature value of the optical fiber light source, and then calculates the voltage value used to drive the laser driving circuit based on the processed ambient temperature value of the optical fiber light source. The optical signal acquisition circuit collects the photovoltage at the C end of the monitoring detector and transmits it to the microprocessor for processing, so that the processed photovoltage is regarded as the light intensity of the optical fiber light source. The ambient temperature value or photovoltage of the optical fiber light source is the parameter of the optical fiber light source.

[0009] The microprocessor receives the fiber optic light source parameters to perform the processing method, including:

[0010] Step 1, the microprocessor obtains the parameter regression line of the collected and transmitted optical fiber light source parameters, and identifies the peak and valley points in the parameter regression line and the corresponding parameter values ​​of the peak and valley points;

[0011] Step 2, take the amount obtained by subtracting the parameter value of the peak and valley point from the mean of the peak and valley point as the deduction one, and determine the mean of the deviation amplitude of the peak and valley point;

[0012] Step 3, the amount obtained by subtracting the parameter value of the peak and valley point from the mean of the deviation amplitude is regarded as the reduction amount 2, and the interference value amplitude of each peak and valley point is determined;

[0013] Step 4, obtain the fluctuation degree of each peak and valley point, and determine the important amplitude of each peak and valley point;

[0014] Step 5, calculate the deletion probability of each peak and valley point;

[0015] Step 6, delete the peak and valley points whose deletion probability is higher than the pre-set critical amount to obtain the target parameter regression line, and execute the RLMD method (robust local mean decomposition method) on the target parameter regression line to remove the nonlinear error. Then, the result after removing the linear error is reconstructed into a new regression line using the Hilbert transform, and the corresponding regression line equation of the new regression line is used to calculate at each acquisition time to obtain the processed fiber light source parameters.

[0016] Furthermore, the temperature sensor or the optical signal acquisition circuit acquires the optical fiber light source parameters at a speed of five times per second.

[0017] Furthermore, in Step 1, after the microprocessor obtains the collected and transmitted fiber optic light source parameters, it can construct a parameter regression line of the fiber optic light source parameters on a rectangular coordinate system through a RANSAC algorithm (random sample consensus algorithm), and obtain peak and valley points in the parameter regression line.

[0018] Furthermore, in Step 1, when obtaining the parameter regression line of the fiber optic light source parameters and identifying the peak and valley points in the parameter regression line and the corresponding parameter values ​​of the peak and valley points, the slice span is set in advance. After obtaining the parameter regression line of the fiber optic light source parameters, the parameter regression line is sliced ​​according to the slice span to obtain several sub-regression lines, and the peak and valley points of each sub-regression line in the parameter regression line and the corresponding parameter values ​​of the peak and valley points are identified.

[0019] Furthermore, the calculation equation for determining the amplitude of the interference value at the peak and valley point is:

[0020] ;

[0021] In the equation, Representative The interference value amplitude of the peak and valley points, Representative The parameter value of the peak and valley point, Representative The parameter value of the peak and valley point, Represents the mean of peak and valley points, Represents the number of peak and valley points.

[0022] Furthermore, in Step 4, if the peak-to-valley point is a peak point, when obtaining the fluctuation degree of the peak point, the average of the difference between the peak point and the valley points adjacent to it on both sides is obtained, and its calculation equation is:

[0023] ;

[0024] In the equation, Representative The fluctuation of the peak point, Representative The parameter value of the peak point, Represents the parameter value of the valley point adjacent to the peak point. Represents the parameter value of the peak point next to the valley point;

[0025] If the peak-to-valley point is a valley point, when obtaining the fluctuation degree of the valley point, the average of the difference between the valley point and the peak points adjacent to it on both sides is obtained.

[0026] Furthermore, in Step 4, after determining the fluctuation degree of each peak and valley point, the important amplitude of each peak and valley point is determined according to the fluctuation degree:

[0027] When identifying the important amplitude of each peak and valley point, take any peak and valley point within the parameter regression line as the target peak and valley point, and use the interference value amplitude of the peak and valley point as the regression weight to perform regression on the fluctuation, and obtain the fluctuation regression value of each peak and valley point; under the condition that the fluctuation of the target peak and valley point is regressed, the fluctuation of all peak and valley points outside the target peak and valley point and the total amount of residual errors of the fluctuation regression value are obtained as regression attribute one; under the condition that the fluctuation of the target peak and valley point is not regressed, the fluctuation of all peak and valley points outside the target peak and valley point and the total amount of residual errors of the fluctuation regression value are obtained as regression attribute two; identify the important amplitude of each peak and valley point.

[0028] Furthermore, in Step 4, the calculation equation for determining the critical amplitude of the peak and valley points is:

[0029] ;

[0030] In the equation, Representative The critical amplitude of the peak and valley points, Representative The fluctuation of the peaks and valleys, Representative When the fluctuation degree of the peak and valley points is regressed The regression value of the fluctuation degree of the peak and valley points, Representative When the fluctuation degree of the peak and valley points is not regressed, The regression value of the fluctuation degree of the peak and valley points, Represents the number of peak and valley points, Representative The fluctuation degree of each peak-to-valley point is the same as that of the The quantity obtained by subtracting the fluctuations of the peaks and valleys.

[0031] Furthermore, in Step 5, the calculation equation for the probability of deleting the peak and valley points is:

[0032] ;

[0033] In the equation, Representative The probability of deleting peak and valley points, Representative The interference value amplitude of the peak and valley points, Representative The critical amplitude of the peak and valley points, Represents the use of Z-score method to Perform standardized operations.

[0034] Furthermore, in Step 6, the pre-set critical amount is defined as eighty-one percent.

[0035] A fiber optic light source control system based on parameter optimization, comprising:

[0036] An optical signal acquisition circuit and a temperature sensor connected to the microprocessor, the temperature sensor is used to collect the ambient temperature value of the optical fiber light source and transmit it to the microprocessor, the microprocessor is used to process the collected ambient temperature value of the optical fiber light source, and then calculate the voltage value used to drive the laser driving circuit according to the ambient temperature value of the optical fiber light source after the processing. The optical signal acquisition circuit is also connected to the C end of the monitoring detector, and the B end of the monitoring detector is the output end of the light source. The optical signal acquisition circuit is used to collect the photovoltage of the C end of the monitoring detector and transmit it to the microprocessor for processing, so that the processed photovoltage is regarded as the light intensity of the optical fiber light source, and the ambient temperature value or photovoltage of the optical fiber light source is the optical fiber light source parameter;

[0037] Modules running on a microprocessor include:

[0038] A recognition module, which is used by the microprocessor to obtain the parameter regression line of the collected and transmitted optical fiber light source parameters, and to recognize the peak and valley points in the parameter regression line and the corresponding parameter values ​​of the peak and valley points;

[0039] The decrement module is used to take the amount obtained by subtracting the parameter value of the peak and valley points from the mean of the peak and valley points as decrement one, and determine the mean of the deviation amplitude of the peak and valley points;

[0040] The interference module is used to take the amount obtained by subtracting the parameter value of the peak and valley points from the average of the deviation amplitude as the decrement two, and to determine the amplitude of the interference value of each peak and valley point;

[0041] The key module is used to obtain the fluctuation degree of each peak and valley point and determine the key amplitude of each peak and valley point;

[0042] A calculation module, which is used to calculate the deletion probability of each peak and valley point;

[0043] The processing module is used to delete the peak and valley points whose deletion probability is higher than the preset critical amount to obtain the target parameter regression line, and the RLMD method is used to remove the nonlinear error of the target parameter regression line. Then, the result after removing the linear error is reconstructed into a new regression line using the Hilbert transform, and the regression line equation corresponding to the new regression line is used to calculate at each collection time to obtain the processed fiber light source parameters.

[0044] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0045] When processing the collected fiber optic light source parameters, the interference value involved has a significant interference effect on the removal effect of nonlinear errors, which will eventually be detrimental to the accuracy of the fiber optic light source parameters. Based on this, the basic change points in the parameters are found through the obtained peak and valley points, and the interference value amplitude and the critical amplitude of the peak and valley points are calculated, and finally the peak and valley points with not low interference value amplitude and not high critical amplitude are identified as interference value points to be deleted. Therefore, the combination of the interference value amplitude and the critical amplitude can greatly improve the accuracy of deleting interference value points, and finally improve the accuracy of the collected optical intensity parameters of the fiber optic light source or the ambient temperature value of the fiber optic light source. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a flow chart of the optical fiber light source control method based on parameter optimization described in the present invention;

[0047] Figure 2 It is a partial structural diagram of the optical fiber light source control system based on parameter optimization described in the present invention. DETAILED DESCRIPTION

[0048] During the acquisition of fiber optic light source parameters, the nonlinear error of the temperature sensor or the optical signal acquisition circuit is a common factor that is detrimental to the fiber optic light source parameters, which often causes the optical intensity parameters of the fiber optic light source or the ambient temperature value of the fiber optic light source to be inaccurate, thereby making the control performance of the fiber optic light source poor. Therefore, the nonlinear error must be removed before the microprocessor collects the fiber optic light source parameters. The fiber optic light source parameters generally contain interference values ​​caused by various external electric or magnetic fields, which is not conducive to the removal of nonlinear errors. Although the current filtering interference values, removing outliers and standardization methods can partially optimize the performance of fiber optic light source parameters, they cannot ensure the complete removal of all interference values, especially some suspected outliers often reflect the normal state or behavior of the fiber optic light source parameters. Direct deletion is often not conducive to the final level of the fiber optic light source parameters and the effective operation of the fiber optic light source control system.

[0049] The microprocessor in the present application collects fiber optic light source parameters to perform processing methods, which obtains peak and valley points to find out the basic change points in the value, and calculates the interference value amplitude and the critical amplitude of the peak and valley points, and finally identifies the peak and valley points with large interference value amplitude and small critical amplitude as interference value points that are often to be deleted. Therefore, by combining the interference value amplitude and the critical amplitude, the accuracy of deleting the interference value points can be effectively improved, and finally the accuracy of the collected light intensity parameters of the fiber optic light source or the ambient temperature value of the fiber optic light source is improved, so that the control performance of the fiber optic light source is better, which is beneficial to the final level of the fiber optic light source parameters and the effective operation of the fiber optic light source control system.

[0050] In order to make the purpose, technical solution and advantages of the present invention clearer, the following will be combined with the drawings in the embodiments of the present invention to clearly and completely express the technical solution of the present invention. The embodiments expressed in this application are only some embodiments of the present invention, not all embodiments. According to the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the protection scope of the present invention.

[0051] like Figure 1 As shown, a method for controlling an optical fiber light source based on parameter optimization according to the present invention comprises:

[0052] The temperature sensor collects the ambient temperature value of the optical fiber light source and transmits it to the microprocessor. The microprocessor processes the collected ambient temperature value of the optical fiber light source, and then calculates the voltage value used to drive the laser driving circuit based on the processed ambient temperature value of the optical fiber light source. The optical signal acquisition circuit collects the photovoltage at the C end of the monitoring detector and transmits it to the microprocessor for processing, so that the processed photovoltage is regarded as the light intensity of the optical fiber light source. The ambient temperature value or photovoltage of the optical fiber light source is the parameter of the optical fiber light source.

[0053] The microprocessor receives the fiber optic light source parameters to perform the processing method, including:

[0054] Step 1, the microprocessor obtains the parameter regression line of the collected and transmitted optical fiber light source parameters, identifies the peak and valley points in the parameter regression line and the corresponding parameter values ​​of the peak and valley points; the microprocessor also obtains the collection time of the parameter transmitted along with the optical fiber light source parameters.

[0055] In a preferred but non-limiting embodiment of the present invention, the temperature sensor or the optical signal acquisition circuit acquires the optical fiber light source parameters at a speed of five times per second.

[0056] In a preferred but non-limiting embodiment of the present invention, in Step 1, after the microprocessor obtains the collected and transmitted fiber light source parameters, it can construct a parameter regression line of the fiber light source parameters on a rectangular coordinate system through a RANSAC algorithm, and obtain the peak and valley points in the parameter regression line. The horizontal coordinate of the rectangular coordinate system is the acquisition time of the parameter, and the vertical coordinate of the rectangular coordinate system is the fiber light source parameter. The parameter regression equation corresponds to the parameter regression line.

[0057] The peak and valley points in the parameter regression line are peak points or valley points. A peak point is when the parameter values ​​on both sides of the parameter regression line are lower than the parameter value of the peak point, and a valley point is when the parameter values ​​on both sides of the parameter regression line are higher than the parameter value of the peak point. The parameter points with zero derivative values ​​on the parameter regression line can be selected by the derivation method. If the selected parameter points have lower parameter values ​​on both sides of the parameter regression line than the parameter value of the peak point, it is a peak point. If the selected parameter points have higher parameter values ​​on both sides of the parameter regression line than the parameter value of the peak point, it is a valley point.

[0058] The parameter changes involved in the regression line of all parameters are often not small, so by processing the parameters into pieces, the changing properties of the parameters can be better analyzed.

[0059] In a preferred but non-limiting embodiment of the present invention, in Step 1, the parameter regression line of the optical fiber light source parameters is obtained, and when the peak and valley points in the parameter regression line and the corresponding parameter values ​​of the peak and valley points are identified, the slice span is set in advance. After the parameter regression line of the optical fiber light source parameters is obtained, the parameter regression line is sliced ​​according to the slice span to obtain several sub-regression lines, and the peak and valley points of each sub-regression line in the parameter regression line and the corresponding parameter values ​​of the peak and valley points are identified.

[0060] Here, the slicing span can be the size of the collection time, just like dividing the parameter regression line into several sub-regression lines according to the length of the collection time of its horizontal axis, such as five-minute intervals.

[0061] The peak and valley points often show significant interference value characteristics due to interference values. However, the peak and valley points often play an important role in parameter analysis. That is, if the peak and valley point interference value amplitude is not low, but its location in the parameter, fluctuation properties or the effect on the overall parameter structure are very important, then the peak and valley point should be retained; if the peak and valley point interference value amplitude is not low, but its importance to the waveform is not high, then the peak and valley point can be deleted. Based on this, the interference value amplitude and important amplitude of each peak and valley point can be obtained according to the following method to determine whether the peak and valley point should be deleted.

[0062] Step 2, take the amount obtained by subtracting the parameter value of the peak and valley point from the mean of the peak and valley point as the deduction one, and determine the mean of the deviation amplitude of the peak and valley point;

[0063] Here, the deviation amplitude is proportional to the modulus of the reduction 1. The peak-to-valley point mean is the mean of the parameter values ​​of the peak-to-valley points.

[0064] The deviation amplitude of the peak-valley point is related to the interference value amplitude of the peak-valley point. If the deviation amplitude is larger, the interference value amplitude of the peak-valley point is larger; if the deviation amplitude is smaller, the interference value amplitude of the peak-valley point is smaller.

[0065] Just as if the valley point is the peak point, we can first determine the mean of the peak points within the sub-regression line, and then obtain the value obtained by subtracting the parameter value of the peak point from the mean of the peak point to determine the mean of the deviation amplitude of the peak point.

[0066] Step 3, the amount obtained by subtracting the parameter value of the peak and valley point from the mean of the deviation amplitude is regarded as the reduction amount 2, and the interference value amplitude of each peak and valley point is determined;

[0067] Here, the amplitude of the interference value and the modulus of the decrement two are in direct proportion.

[0068] In a preferred but non-limiting embodiment of the present invention, the calculation equation for determining the interference value amplitude of the peak and valley points is:

[0069] ;

[0070] In the equation, Representative The interference value amplitude of the peak and valley points, Representative The parameter value of the peak and valley point, Representative The parameter value of the peak and valley point, Represents the mean of peak and valley points, Represents the number of peak and valley points.

[0071] In this equation, Representative The deviation amplitude of the peak-to-valley point is higher. The greater the deviation amplitude of each peak and valley point.

[0072] Represents the average of the deviation amplitudes of all peak and valley points. The higher the value, the greater the overall deviation amplitude of all peak and valley points.

[0073] After determining the interference value amplitude of each peak and valley point according to the above method, the important amplitude of each peak and valley point can be determined according to the following method.

[0074] Step 4, obtain the fluctuation degree of each peak and valley point, and determine the important amplitude of each peak and valley point;

[0075] Here, the modulus of the quantity obtained by subtracting the two adjacent values ​​of the important amplitude and the fluctuation degree of the peak and valley points is in a proportional relationship.

[0076] The higher the fluctuation of the peak-valley point, the higher the probability that the peak-valley point contains a critical value, and the larger its critical amplitude. Therefore, the critical amplitude of each peak-valley point can be obtained by obtaining the fluctuation of the peak-valley point and performing analysis.

[0077] In a preferred but non-limiting embodiment of the present invention, in Step 4, if the peak-valley point is a peak point, when obtaining the fluctuation degree of the peak point, the average of the difference between the peak point and the valley points adjacent to it on both sides is obtained, and the calculation equation is:

[0078] ;

[0079] In the equation, Representative The fluctuation of the peak point, Representative The parameter value of the peak point, Represents the parameter value of the valley point adjacent to the peak point. Represents the parameter value of the peak point next to the valley point;

[0080] If the peak-to-valley point is a valley point, when obtaining the fluctuation degree of the valley point, it is to obtain the average of the difference between the valley point and the peak points adjacent to it on both sides. The difference is the amount obtained by subtraction.

[0081] When determining the fluctuation degree of the first and last pair of peak-valley points, there is only one adjacent peak-valley point. Under this condition, the difference between the peak-valley point and the adjacent peak-valley point is regarded as the fluctuation degree of the peak-valley point. Just as, if the first peak-valley point is the peak point, when obtaining the fluctuation degree of the peak point, the difference between the valley point adjacent to the peak value is regarded as the fluctuation degree of the peak point.

[0082] In a preferred but non-limiting embodiment of the present invention, in Step 4, after the fluctuation degree of each peak-valley point is determined according to the above method, the critical amplitude of each peak-valley point is then determined according to the fluctuation degree:

[0083] In order to analyze the importance of the peak and valley points in the sub-regression line, the present application can determine the importance of the target peak and valley points in all the peak and valley points by obtaining the total residual error measurement when the regression is performed and not performed on the fluctuation of the target peak and valley points. The modulus of the quantity obtained by subtracting the total residual error measurement represents that the greater the importance of the target peak and valley point, the greater its effect on the fiber optic light source parameters.

[0084] When identifying the critical amplitude of each peak and valley point, take any peak and valley point within the parameter regression line as the target peak and valley point, take the interference value amplitude of the peak and valley point as the regression weight to perform regression on the fluctuation, and obtain the fluctuation regression value of each peak and valley point; under the condition that the fluctuation of the target peak and valley point is regressed, obtain the total residual error of the fluctuation and the fluctuation regression value of all peak and valley points outside the target peak and valley point, and treat it as regression attribute one; under the condition that the fluctuation of the target peak and valley point is not regressed, obtain the total residual error of the fluctuation and the fluctuation regression value of all peak and valley points outside the target peak and valley point, and treat it as regression attribute two; identify the critical amplitude of each peak and valley point, and the critical amplitude is proportional to the modulus of the amount obtained by subtracting regression attribute one from regression attribute two. The residual error of the fluctuation and the fluctuation regression value of the peak and valley point is the amount obtained by subtracting the fluctuation regression value from the fluctuation of the peak and valley point.

[0085] Here, performing regression on the fluctuation can be achieved via the RANSAC algorithm.

[0086] In a preferred but non-limiting embodiment of the present invention, in Step 4, the computational equation for determining the critical amplitude of the peak-valley point is:

[0087] ;

[0088] In the equation, Representative The critical amplitude of the peak and valley points, Representative The fluctuation of the peaks and valleys, Representative When the fluctuation degree of the peak and valley points is regressed The regression value of the fluctuation degree of the peak and valley points, Representative When the fluctuation degree of the peak and valley points is not regressed, The regression value of the fluctuation degree of the peak and valley points, Represents the number of peak and valley points, Representative The fluctuation degree of each peak-to-valley point is the same as that of the The quantity obtained by subtracting the fluctuations of the peaks and valleys.

[0089] In this equation, Representative The fluctuation degree of each peak-to-valley point is the same as that of the The modulus of the quantity obtained by subtracting the fluctuations of the peak and valley points. The higher the modulus, the higher the stability of the peak and valley point changes, and the stronger the confidence in the critical amplitude of the peak and valley points obtained; the lower the modulus, the lower the stability of the peak and valley point changes, and the weaker the confidence in the critical amplitude of the peak and valley points obtained.

[0090] represents the regression attribute one, represents the regression attribute 2, and The modulus of the quantity obtained by subtracting the two can represent the important amplitude of the peak and valley point. The higher the modulus, the larger the important amplitude of the peak and valley point.

[0091] After the interference value amplitude and the critical amplitude of each peak and valley point are determined according to the above Step 2 to Step 4, the peak and valley points can be selectively deleted according to the interference value amplitude and the critical amplitude of each peak and valley point.

[0092] Step 5, calculate the deletion probability of each peak and valley point;

[0093] If the interference value amplitude of the peak-valley point is not low and the critical amplitude is not high, it means that the probability that the peak-valley point is an interference value point is not low, and the probability of being deleted is also greater.

[0094] In a preferred but non-limiting embodiment of the present invention, in Step 5, the calculation equation for the probability of deleting the peak and valley points is:

[0095] ;

[0096] In the equation, Representative The probability of deleting peak and valley points, Representative The interference value amplitude of the peak and valley points, Representative The critical amplitude of the peak and valley points, Represents the use of Z-score method to Perform standardized operations.

[0097] Represents the outlier amplitude of the peak and valley point. The larger the outlier amplitude, the higher the probability of deletion; the smaller the outlier amplitude, the lower the probability of deletion.

[0098] According to the above method, the deletion probability of all peak points and valley points can be obtained respectively, and then the following process is executed.

[0099] Step 6, delete the peak and valley points whose deletion probability is higher than the pre-set critical amount to obtain the target parameter regression line, and use the RLMD method to remove the nonlinear error of the target parameter regression line. Then, the result after removing the linear error is reconstructed into a new regression line using the Hilbert transform, and the corresponding regression line equation of the new regression line is used to calculate at each collection time to obtain the processed fiber light source parameters.

[0100] In a preferred but non-limiting embodiment of the present invention, in Step 6, the pre-set critical amount is defined as eighty-one percent.

[0101] According to the above method, the peak and valley points whose deletion probability is higher than the pre-set critical value in all the peak and valley points of each sub-regression line can be deleted, and the accuracy of the obtained target parameter regression line can be greatly increased. Then, the target parameter regression line is processed by the RLMD method, which can greatly improve the accuracy of removing nonlinear errors.

[0102] Here, the RANSAC algorithm can be used to perform regression on the peak points and valley points in the peak-valley points respectively. The regression line obtained by the peak point is the upper regression line, and the regression line obtained by the valley point is the lower regression line. Then, the RLMD method is used to perform several cycles on the regression line of the target parameter to remove the nonlinear error.

[0103] In order to prevent the accumulation of interference values ​​in the future, the deletion operation is only performed during the first loop operation, and the change operation does not need to be performed in subsequent loop operations.

[0104] After executing the RLMD method on the target parameter regression line, in order to accurately analyze the target parameter regression line, the result obtained by the RLMD method after removing the nonlinear error can be reconstructed using the Hilbert transform, which is suitable for subsequent parameter detection and analysis.

[0105] Therefore, in the present application, the parameter regression line of the fiber optic light source parameters is obtained, and the peak and valley points in the parameter regression line and the corresponding parameter values ​​of the peak and valley points are identified; the amount obtained by subtracting the parameter value of the peak and valley point from the mean of the peak and valley point is taken as the decrement one, and the mean of the deviation amplitude of the peak and valley point is determined, and the deviation amplitude is proportional to the modulus of the decrement one; the two obtained by subtracting the parameter value of the peak and valley point from the mean of the deviation amplitude is taken as the decrement two, and the interference value amplitude of each peak and valley point is determined, and the interference value amplitude is proportional to the modulus of the decrement two; the fluctuation degree of each peak and valley point is obtained, the critical amplitude of each peak and valley point is determined, and the critical amplitude of each peak and valley point is determined. The modulus of the quantity obtained by subtracting the adjacent values ​​of the amplitude and the fluctuation degree of the peak and valley points is proportional to each other; the interference value amplitude and the important amplitude of the peak and valley points are combined to calculate the deletion probability of each peak and valley point; the peak and valley points with deletion probability higher than the preset critical value are deleted to obtain the target parameter regression line, and the RLMD method is used to remove the nonlinear error of the target parameter regression line, and then the result after removing the linear error is reconstructed into a new regression line using the Hilbert transform, and the regression line equation corresponding to the new regression line is used to calculate the processed fiber light source parameters at each acquisition time.

[0106] Therefore, the present invention finds out the basic change points within the parameters by obtaining the peak and valley points, and calculates the interference value amplitude and the critical amplitude of the peak and valley points, and finally identifies the peak and valley points whose interference value amplitude is not low and the critical amplitude is not high as interference value points that are often to be deleted. Therefore, the accuracy of deleting interference value points is greatly improved by combining the interference value amplitude and the critical amplitude, and finally improves the accuracy of the collected light intensity parameters of the optical fiber light source or the ambient temperature value of the optical fiber light source.

[0107] like Figure 2 As shown, the optical fiber light source control system based on parameter optimization described in the present invention includes:

[0108] An optical signal acquisition circuit and a temperature sensor connected to the microprocessor, the temperature sensor is used to collect the ambient temperature value of the optical fiber light source and transmit it to the microprocessor, the microprocessor is used to process the collected ambient temperature value of the optical fiber light source, and then calculate the voltage value used to drive the laser driving circuit according to the ambient temperature value of the optical fiber light source after the processing. The optical signal acquisition circuit is also connected to the C end of the monitoring detector, and the B end of the monitoring detector is the output end of the light source. The optical signal acquisition circuit is used to collect the photovoltage of the C end of the monitoring detector and transmit it to the microprocessor for processing, so that the processed photovoltage is regarded as the light intensity of the optical fiber light source, and the ambient temperature value or photovoltage of the optical fiber light source is the optical fiber light source parameter;

[0109] Modules running on a microprocessor include:

[0110] A recognition module, which is used by the microprocessor to obtain the parameter regression line of the collected and transmitted optical fiber light source parameters, and to recognize the peak and valley points in the parameter regression line and the corresponding parameter values ​​of the peak and valley points;

[0111] The decrement module is used to take the amount obtained by subtracting the parameter value of the peak and valley points from the mean of the peak and valley points as decrement one, and determine the mean of the deviation amplitude of the peak and valley points;

[0112] The interference module is used to take the amount obtained by subtracting the parameter value of the peak and valley points from the average of the deviation amplitude as the decrement two, and to determine the amplitude of the interference value of each peak and valley point;

[0113] The key module is used to obtain the fluctuation degree of each peak and valley point and determine the key amplitude of each peak and valley point;

[0114] A calculation module, which is used to calculate the deletion probability of each peak and valley point;

[0115] The processing module is used to delete the peak and valley points whose deletion probability is higher than the preset critical amount to obtain the target parameter regression line, and the RLMD method is used to remove the nonlinear error of the target parameter regression line. Then, the result after removing the linear error is reconstructed into a new regression line using the Hilbert transform, and the regression line equation corresponding to the new regression line is used to calculate at each collection time to obtain the processed fiber light source parameters.

[0116] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include:

[0117] When processing the collected fiber optic light source parameters, the interference value involved has a significant interference effect on the removal effect of nonlinear errors, which will eventually be detrimental to the accuracy of the fiber optic light source parameters. Based on this, the basic change points in the parameters are found through the obtained peak and valley points, and the interference value amplitude and the critical amplitude of the peak and valley points are calculated, and finally the peak and valley points with not low interference value amplitude and not high critical amplitude are identified as interference value points to be deleted. Therefore, the combination of the interference value amplitude and the critical amplitude can greatly improve the accuracy of deleting interference value points, and finally improve the accuracy of the collected optical intensity parameters of the fiber optic light source or the ambient temperature value of the fiber optic light source.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not deviate from the spirit and scope of the present invention should be covered within the protection space of the claims of the present invention.

Claims

1. A method for controlling an optical fiber light source based on parameter optimization, characterized in that: include: The temperature sensor collects the ambient temperature value of the optical fiber light source and transmits it to the microprocessor. The microprocessor processes the collected ambient temperature value of the optical fiber light source, and then calculates the voltage value used to drive the laser driving circuit based on the processed ambient temperature value of the optical fiber light source. The optical signal acquisition circuit collects the photovoltage at the C end of the monitoring detector and transmits it to the microprocessor for processing, so that the processed photovoltage is regarded as the light intensity of the optical fiber light source. The ambient temperature value or photovoltage of the optical fiber light source is the parameter of the optical fiber light source. The microprocessor receives the fiber optic light source parameters to perform the processing method, including: Step 1, the microprocessor obtains the parameter regression line of the collected and transmitted optical fiber light source parameters, and identifies the peak and valley points in the parameter regression line and the corresponding parameter values ​​of the peak and valley points; Step 2, take the amount obtained by subtracting the parameter value of the peak and valley point from the mean of the peak and valley point as the deduction one, and determine the mean of the deviation amplitude of the peak and valley point; Step 3, the amount obtained by subtracting the parameter value of the peak and valley point from the mean of the deviation amplitude is regarded as the reduction amount 2, and the interference value amplitude of each peak and valley point is determined; Step 4, obtain the fluctuation degree of each peak and valley point, and determine the important amplitude of each peak and valley point; Step 5, calculate the deletion probability of each peak and valley point; Step 6, the peak and valley points whose deletion probability is higher than the preset critical value are deleted to obtain the target parameter regression line, and the RLMD method is used to remove the nonlinear error of the target parameter regression line, and then the result after removing the linear error is reconstructed into a new regression line by using the Hilbert transform, and the regression line equation corresponding to the new regression line is used to calculate at each acquisition time to obtain the processed fiber light source parameters; The calculation equation for determining the interference value amplitude of the peak and valley points is: ； In the equation, Representative The interference value amplitude of the peak and valley points, Representative The parameter value of the peak and valley point, Representative The parameter value of the peak and valley point, represents the mean of peak and valley points, Represents the number of peak and valley points; In Step 4, if the peak-to-valley point is a peak point, when obtaining the fluctuation degree of the peak point, the average of the difference between the peak point and the valley points adjacent to it on both sides is obtained. The calculation equation is: ； In the equation, Representative The fluctuation of the peak point, Representative The parameter value of the peak point, Represents the parameter value of the valley point adjacent to the peak point. Represents the parameter value of the peak point next to the valley point; If the peak-to-valley point is a valley point, when obtaining the fluctuation degree of the valley point, it is to obtain the average of the difference between the valley point and the peak points adjacent to it on both sides, and the difference is the difference obtained by subtracting them; In Step 4, the calculation equation for determining the critical amplitude of the peak and valley points is: ； In the equation, Representative The critical amplitude of the peak and valley points, Representative The fluctuation of the peaks and valleys, Representative When the fluctuation degree of the peak and valley points is regressed The regression value of the fluctuation degree of the peak and valley points, Representative When the fluctuation degree of the peak and valley points is not regressed, The regression value of the fluctuation degree of the peak and valley points, Represents the number of peak and valley points, Representative The fluctuation degree of each peak-to-valley point is the same as that of the The amount obtained by subtracting the fluctuations of the peaks and valleys; In Step 5, the calculation equation for the probability of deleting peak and valley points is: ； In the equation, Representative The probability of deleting peak and valley points, Representative The interference value amplitude of the peak and valley points, Representative The critical amplitude of the peak and valley points, Represents the use of Z-score method to Perform standardized operations; In Step 6, the pre-set critical amount is defined as eighty-one percent.

2. The optical fiber light source control method based on parameter optimization according to claim 1, characterized in that: The temperature sensor or the optical signal acquisition circuit acquires the optical fiber light source parameters at a speed of five times per second.

3. The optical fiber light source control method based on parameter optimization according to claim 2, characterized in that: In Step 1, after the microprocessor obtains the collected and transmitted fiber optic light source parameters, it can construct a parameter regression line of the fiber optic light source parameters on a rectangular coordinate system through a RANSAC algorithm, and obtain peak and valley points in the parameter regression line.

4. The optical fiber light source control method based on parameter optimization according to claim 3, characterized in that: In Step 1, the parameter regression line of the fiber optic light source parameters is obtained. When the peak and valley points in the parameter regression line and the corresponding parameter values ​​of the peak and valley points are identified, the slice span is set in advance. After the parameter regression line of the fiber optic light source parameters is obtained, the parameter regression line is sliced ​​according to the slice span to obtain several sub-regression lines, and the peak and valley points of each sub-regression line in the parameter regression line and the corresponding parameter values ​​of the peak and valley points are identified.

5. The optical fiber light source control method based on parameter optimization according to claim 4, characterized in that: In Step 4, after determining the fluctuation of each peak and valley point, the important amplitude of each peak and valley point is determined according to the fluctuation: When identifying the critical amplitude of each peak-valley point, take any peak-valley point in the parameter regression line as the target peak-valley point, take the interference value amplitude of the peak-valley point as the regression weight to perform regression on the fluctuation degree, and obtain the fluctuation degree regression value of each peak-valley point; Under the condition that the fluctuation degree of the target peak and valley point is regressed, the total amount of residual errors of the fluctuation degree of all peak and valley points outside the target peak and valley point and the fluctuation degree regression value is obtained, which is used as regression attribute one; under the condition that the fluctuation degree of the target peak and valley point is not regressed, the total amount of residual errors of the fluctuation degree of all peak and valley points outside the target peak and valley point and the fluctuation degree regression value is obtained, which is used as regression attribute two; the important amplitude of each peak and valley point is determined.

Images