A stability optimization method of a photoelectric information conversion assembly in a harsh environment

Abstract

The application belongs to the field of electronic information technology, and provides a stability optimization method of photoelectric information conversion assembly in harsh environment, comprising: the scheme is aimed at the stability optimization of photoelectric information conversion assembly in harsh environment, through synchronous collection of temperature, humidity, illumination and other environmental parameters and equipment current data in high temperature and high humidity environment, a multi-dimensional time sequence group is constructed; the coupling degree of each environmental parameter and current sequence is analyzed, after locking the high coupling relationship between illumination and current, it is judged whether there is delay in current response when illumination suddenly changes; if there is delay, the current response sensitivity is further analyzed, if the sensitivity deviates from the expectation, an illumination-current prediction model is constructed, the current adjustment amount is calculated in advance to realize pre-adjustment through predicting the illumination mutation amplitude; finally, the adjusted data is collected to iterate and optimize the model, and the dynamic response performance and stability of the assembly in complex environment are improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of electronic information technology, and particularly relates to a stability optimization method of a photoelectric information conversion assembly in a harsh environment. BACKGROUND

[0002] As a key component for clean energy utilization and information collection, the photoelectric information conversion assembly is long-term placed in a harsh environment with high temperature and high humidity, the environmental temperature often exceeds 40 DEG C, and the humidity is maintained above 80%, which not only accelerates the aging of the photoelectric information conversion assembly and the supporting circuit elements, but also poses a severe challenge to the current response characteristics of the assembly.

[0003] For the current response lag problem of the photoelectric information conversion assembly under light mutation, the existing solutions have obvious limitations: on the one hand, the traditional current regulation strategy relies on real-time feedback control, that is, the regulation is started after the light mutation occurs and causes current deviation, and the high temperature and high humidity environment further amplifies the response delay of the circuit elements, resulting in that the current response lags behind the actual demand; on the other hand, the existing technology is insufficient in analyzing the correlation between light and current, and fails to quantify the coupling degree of the two, and it is difficult to predict the amplitude and timing of light mutation, so it is impossible to realize advance regulation to offset the lag effect; therefore, there is an urgent need for a technical solution that can accurately capture the light-current correlation characteristics and realize advance prediction and dynamic optimization, to solve the current response lag problem under light mutation.

[0004] Therefore, the application provides a stability optimization method of a photoelectric information conversion assembly in a harsh environment. SUMMARY

[0005] In order to make up for the deficiencies of the prior art and solve at least one technical problem proposed in the background art.

[0006] The technical scheme adopted by the application to solve the technical problems is: a stability optimization method of a photoelectric information conversion assembly in a harsh environment, comprising:

[0007] Collecting environmental parameters and photoelectric information conversion assembly device current data; constructing a plurality of single-dimensional sequences of environmental parameters and a current value sequence of current data in the order of collection time, and integrating the plurality of single-dimensional sequences into a multi-dimensional sequence group;

[0008] Analyzing the coupling degree index of the single-dimensional sequence and the current value sequence in the multi-dimensional sequence group, if it is judged by the coupling degree index that the light intensity sequence contained in the single-dimensional sequence and the current value sequence of the current data have a high coupling relationship, then analyzing whether the current response is delayed when the light intensity mutates;

[0009] If the current response exists delay and the current response sensitivity deviates from the expectation, a light-current prediction model is constructed, the light mutation amplitude is predicted, the current adjustment amount is calculated, and the pre-adjustment of the current is realized;

[0010] The light-current prediction model is iteratively optimized through the multiple groups of light intensity and current data that meet the standard after pre-adjustment.

[0011] Further, the process of obtaining the coupling degree index is:

[0012] A plurality of environmental parameters in the environment of the photoelectric information conversion component device are collected, and a single-dimensional sequence corresponding to the plurality of environmental parameters is constructed;

[0013] An association deviation value of the single-dimensional sequence and the current value sequence is obtained, and a coupling degree calculation formula is constructed based on the association deviation value;

[0014] The coupling degree of the single-dimensional sequence and the current value sequence at each collection time point is obtained through the coupling degree calculation formula;

[0015] The coupling degree of each collection point is processed by mean value to obtain the coupling degree index;

[0016] If the coupling degree index is greater than a preset coupling threshold, it indicates that the single-dimensional sequence and the current value sequence are in a preliminary strong coupling relationship;

[0017] The obtained coupling degree index is sorted in descending order, and the single-dimensional sequence corresponding to the maximum value of the coupling degree index is extracted;

[0018] If the single-dimensional sequence and the current value sequence are in a preliminary strong coupling relationship and the single-dimensional sequence is the sequence with the maximum coupling degree index, it indicates that the single-dimensional sequence and the current value sequence are in a strong coupling relationship.

[0019] Further, the manner of obtaining the association deviation value is:

[0020] The single-dimensional sequence is processed by mean value to obtain a single-dimensional sequence average value, and the current value sequence is processed by mean value to obtain a current value sequence average value;

[0021] The maximum value and the minimum value of the single-dimensional sequence are obtained, and the ratio of the difference between the maximum value and the minimum value of the single-dimensional sequence to the single-dimensional sequence average value is processed to obtain a sequence deviation degree of the single-dimensional sequence;

[0022] The maximum value and the minimum value of the current value sequence are obtained, and the ratio of the difference between the maximum value and the minimum value of the current value sequence to the current value sequence average value is processed to obtain a sequence deviation degree of the current value sequence;

[0023] The difference between the sequence deviation degree of the single-dimensional sequence and the sequence deviation degree of the current value sequence is processed to obtain the association deviation value of the single-dimensional sequence and the current value sequence.

[0024] Further, the process of analyzing whether there is a delay in the current response is:

[0025] Obtain the light intensity sequence and the light change value and the current change value of each acquisition time point of the current value sequence;

[0026] Integrate the light change value and the current change value into the light change sequence and the current change sequence;

[0027] And obtain the normal fluctuation value of the light change sequence and the current significant change threshold value respectively;

[0028] The light change value is processed by ratio to obtain the light mutation index;

[0029] The current change value is processed by ratio to obtain the current mutation index;

[0030] According to the obtained light mutation index and current mutation index, the mutation analysis is carried out to mark the acquisition points as light mutation points and current mutation points;

[0031] Integrate the light mutation points and the current mutation points into the light mutation sequence and the current mutation sequence according to the acquisition time;

[0032] Based on the light mutation sequence and the current mutation sequence, obtain the time delay sequence and calculate the coefficient of variation of the time delay sequence;

[0033] According to the time delay sequence and the coefficient of variation of the time delay sequence, it is judged whether there is obvious lag in the current response.

[0034] Further, the process of obtaining the normal fluctuation value and the current significant change threshold value is:

[0035] Based on the light change sequence, the mean and standard deviation of the light change sequence are calculated;

[0036] According to the 3σ principle, and the mean and standard deviation of the light change sequence, the normal fluctuation value is defined;

[0037] Based on the current change sequence, the mean and standard deviation of the current change sequence are calculated;

[0038] According to the 3σ principle, and the mean and standard deviation of the current change sequence, the current significant change threshold value is defined.

[0039] Further, the process of analyzing the current response sensitivity is:

[0040] Based on the time delay sequence, the mean value is processed to obtain the average delay time of the current response , and the standard deviation of the delay time is calculated according to the average delay time;

[0041] Integrating the light change values of all light mutation points to get the light average variable between the light mutation points;

[0042] Integrating the current change values of all current mutation points to get the current average variable between the current mutation points;

[0043] The light current ratio is obtained by ratio processing of the light average variable and the current average variable, and the current response rate is obtained by ratio processing of the light current ratio and the average delay time of the current response.

[0044] The current response sensitivity is obtained by ratio processing of the current response rate and the delay time standard deviation.

[0045] Further, the process of predicting the light mutation amplitude is:

[0046] Based on the average delay time of the current response , the sliding window is divided into ;

[0047] Obtain the change slope, cumulative change before mutation and fluctuation value in the sliding window; and construct a prediction light mutation amplitude equation;

[0048] The change slope, cumulative change before mutation and fluctuation value in the sliding window are input into the prediction light mutation amplitude equation to obtain the predicted light intensity mutation value.

[0049] Further, the way to obtain the change slope, cumulative change before mutation and fluctuation value in the sliding window is:

[0050] Change slope: calculate the linear fitting slope of the light intensity of the first points in the sliding window;

[0051] Cumulative change before mutation: obtain the sum of all values of the light change amount greater than Q times the normal fluctuation value in the sliding window;

[0052] Wherein, Q is a preset multiple;

[0053] Fluctuation value: calculate the ratio of the standard deviation and the mean of the light mutation index of each collection time point in the first points in the sliding window, and obtain the fluctuation value by averaging the ratio of each collection time point.

[0054] Further, the process of calculating the current adjustment amount and realizing pre-adjustment is:

[0055] Extract the mean of the light change sequence and the mean of the current change sequence;

[0056] Make the mean of the light change sequence and the mean of the current change sequence ratio processing to obtain the basic adjustment coefficient;

[0057] The current adjustment amount prediction formula is constructed based on the predicted light intensity mutation amplitude and the basic adjustment coefficient.

[0058] The predicted light intensity mutation value and the basic adjustment coefficient are input into the current adjustment amount prediction formula to obtain the current adjustment amount.

[0059] According to the predicted light intensity mutation value and the current adjustment amount, a targeted adjustment strategy is formulated.

[0060] Further, the iteration optimization process of the light-current prediction model is as follows:

[0061] After the pre-adjustment is implemented, a plurality of groups of light intensity and current data are collected, and the current response sensitivity is calculated respectively, and if the current response sensitivity of each group of light intensity and current data meets the requirement, it is indicated that the current sensitivity meets the requirement.

[0062] Based on the plurality of groups of light intensity and current data, the basic adjustment coefficient of the calibrated light-current prediction model is recalculated using the mean value of light change and the mean value of current change.

[0063] The beneficial effects of the present application are as follows:

[0064] 1. In a harsh environment of high temperature and high humidity, the environmental parameters and the current data of the photoelectric information conversion assembly equipment are synchronously collected; a plurality of single-dimensional sequences of environmental parameters and a current value sequence of current data are constructed according to the collection time sequence, and the environmental parameter sequence is integrated into a multi-dimensional sequence group; the dynamic relationship between the high temperature and high humidity environmental parameters and the current of the photoelectric information conversion assembly is facilitated, and multi-dimensional basic data matched in time sequence for analyzing the performance of the equipment in the harsh environment is provided; based on the multi-dimensional sequence group, the coupling degree index of the sequence in the multi-dimensional sequence group and the current value sequence is analyzed; if the light intensity sequence and the current value sequence are in a high coupling relationship, it is judged whether the current response is delayed when the light intensity mutates; the key influencing factor of light intensity can be locked, and the dynamic response performance of the equipment under the action of the factor is effectively evaluated by judging the current response delay when the factor mutates.

[0065] 2. If the current response exists delay, the sensitivity response analysis is carried out to the current response, and the current response sensitivity is obtained; if the current response sensitivity deviates from the expectation, the illumination-current prediction model is constructed, the current adjustment amount is calculated in advance through the prediction of illumination mutation amplitude, and the pre-adjustment of the current is realized; the current response delay problem can be solved in a targeted manner, the adjustment amount is calculated in advance through the illumination-current prediction model when the sensitivity deviates from the expectation, the pre-adjustment is realized, the equipment current stability and operation performance are effectively guaranteed; a plurality of groups of data after adjustment are collected and analyzed, and the illumination-current prediction model is iteratively optimized according to the analysis result; the illumination-current prediction model is more fitted to the actual adjustment data through iterative optimization, the pre-adjustment precision is continuously improved, and the stable operation of the equipment is guaranteed. BRIEF DESCRIPTION OF DRAWINGS

[0066] The application will be further described below with reference to the drawings.

[0067] Figure 1 is a step flow chart of an optimization method of a photoelectric information conversion assembly in a harsh environment according to an embodiment of the application;

[0068] Figure 2 is a logic thinking diagram of a stability optimization method of a photoelectric information conversion assembly in a harsh environment according to an embodiment of the application;

[0069] Figure 3 is a module diagram of a stability optimization system of a photoelectric information conversion assembly in a harsh environment according to an embodiment of the application. DETAILED DESCRIPTION

[0070] In order to make the technical means, creative features, purposes and effects realized by the application easy to understand, the application will be further described below with reference to the specific embodiments.

[0071] Embodiment 1: please refer to Figure 1 The stability optimization method of the photoelectric information conversion assembly in the harsh environment according to the embodiment of the application includes:

[0072] S1: in a high-temperature and high-humidity harsh environment, environment parameters and photoelectric information conversion assembly equipment current data are synchronously collected; a plurality of groups of single-dimensional sequences of the environment parameters and a current value sequence of the current data are constructed in the order of collection time, and the environment parameter sequences are integrated into a multi-dimensional sequence group;

[0073] In the high-temperature and high-humidity harsh environment, the process of synchronously collecting the environment parameters and the photoelectric information conversion assembly equipment current data is as follows:

[0074] Select temperature and humidity data of the harsh environment where the photoelectric information conversion component equipment is located by using temperature and humidity resistant industrial sensors; the temperature range of the sensor should cover 0-120℃, the accuracy is ±0.1℃, the humidity sensor supports 5-100%RH measurement, the accuracy is ±3%RH;

[0075] Collect the light intensity in the harsh environment where the equipment is located by using a spectral sensor; the spectral sensor measures 0-65535 lux, supports visible and infrared spectrum;

[0076] Collect the current value of the photoelectric information conversion component through the current transformer, and adjust the range according to the rated current of the equipment;

[0077] The temperature data, humidity data and light intensity data are collectively referred to as environmental parameters, and the current size is marked as equipment running data;

[0078] All sensors are synchronized by hardware trigger or NTP time synchronization protocol to ensure that the time error of each collection time point is less than 10 milliseconds;

[0079] After ensuring that the environmental parameters and equipment running data are collected in time alignment, the environmental parameters and equipment running data are sampled once per second;

[0080] The collected environmental parameters and equipment running data are preprocessed;

[0081] It can be understood that the preprocessing process is: first, eliminate abnormal values caused by sensor instantaneous drift, communication interruption or electromagnetic interference; then, the environmental parameters and equipment running data are processed by moving average to suppress high frequency noise and ensure the validity of the collected data;

[0082] Among them, the process of constructing multiple single-dimensional sequence groups of environmental parameters and current value sequence of current data according to the collection time sequence, and integrating the environmental parameter sequence into multiple sequence groups is:

[0083] The preprocessed environmental parameters and equipment running data are summarized in the order of collection time points to obtain temperature sequence, humidity sequence, light intensity sequence, which are collectively referred to as single-dimensional sequence, and current value sequence;

[0084] Integrate the single-dimensional sequence into a multi-dimensional sequence group;

[0085] S2: Based on the multi-dimensional sequence group, analyze the coupling degree index of the sequence and the current value sequence in the multi-dimensional sequence group; if the light intensity sequence and the current value sequence have a high coupling relationship, determine whether the current response exists delay when the light intensity suddenly changes;

[0086] Among them, the process of analyzing the coupling degree index of the sequence and the current value sequence in the multi-dimensional sequence group based on the multi-dimensional sequence group is:

[0087] The coupling index is used to quantify the correlation between the single-dimensional sequence and the current value sequence, and the calculation process is as follows:

[0088] It can be understood that the single-dimensional sequence represents any one sequence in the multi-dimensional sequence group;

[0089] The single-dimensional sequence is [x1, x2,..., x i ], and the current value sequence is [y1, y2,..., y i ];

[0090] Wherein, i is the acquisition time point;

[0091] The single-dimensional sequence is subjected to mean value processing to obtain a single-dimensional sequence average value, and the current value sequence is subjected to mean value processing to obtain a current value sequence average value;

[0092] The maximum value and the minimum value of the single-dimensional sequence are obtained, and the difference between the maximum value and the minimum value of the single-dimensional sequence is subjected to ratio processing with the single-dimensional sequence average value to obtain a sequence deviation degree of the single-dimensional sequence;

[0093] The maximum value and the minimum value of the current value sequence are obtained synchronously, and the difference between the maximum value and the minimum value of the current value sequence is subjected to ratio processing with the current value sequence average value to obtain a sequence deviation degree of the current value sequence;

[0094] The sequence deviation degree of the single-dimensional sequence is subjected to difference processing with the sequence deviation degree of the current value sequence to obtain a correlation deviation value of the single-dimensional sequence and the current value sequence ;

[0095] It should be noted that the physical meaning of the correlation deviation value is that the correlation deviation value is calculated by the sequence deviation degree of the single-dimensional sequence and the sequence deviation degree of the current value sequence; the sequence deviation degree of the single-dimensional sequence reflects the discrete degree of the single-dimensional sequence, and the sequence deviation degree of the current value sequence reflects the fluctuation degree of the current value sequence; specifically, the correlation deviation value quantifies the synchronous consistency of the single-dimensional sequence fluctuation and the current value sequence fluctuation, and the smaller the correlation deviation value, the closer the fluctuation amplitudes of the single-dimensional sequence and the current value sequence, and the stronger the synchronization; on the contrary, the difference between the two sequences is significant, and the synchronization is weaker;

[0096] Based on the obtained correlation deviation value, the coupling degree of the single-dimensional sequence and the current value sequence at each acquisition time point is calculated; wherein, the coupling degree calculation formula is:

[0097] ;

[0098] Wherein, represents the value of the single-dimensional sequence at the acquisition time point i, and the current value of the current value sequence at the acquisition time point i. This indicates the coupling degree between the one-dimensional sequence and the current value sequence at acquisition time point i;

[0099] The coupling degree between the one-dimensional sequence and the current value sequence at each acquisition time point is averaged to obtain the coupling degree index between the one-dimensional sequence and the current value sequence.

[0100] Based on the above calculation process of coupling degree index, the coupling degree index of temperature sequence and current value sequence, the coupling degree index of humidity sequence and current value sequence, and the coupling degree index of light intensity sequence and current value sequence are obtained in sequence.

[0101] The obtained coupling index is compared with the preset coupling threshold;

[0102] If the coupling index is greater than the preset coupling threshold, it indicates that the one-dimensional sequence and the current value sequence have a preliminary strong coupling relationship.

[0103] If the coupling index is less than or equal to the preset coupling threshold, it indicates that the one-dimensional sequence and the current value sequence are weakly coupled.

[0104] Sort the obtained coupling indexes by size and take the maximum value;

[0105] If the one-dimensional sequence and the current value sequence have a preliminary strong coupling relationship and the coupling degree index is at its maximum value, then it indicates that the one-dimensional sequence and the current value sequence have a strong coupling relationship.

[0106] If the light intensity sequence and the current value sequence are highly coupled, then it is determined whether there is a delay in the current response when the light intensity changes abruptly.

[0107] The light intensity sequence is [L1, L2, ... L i ]; where L i This represents the light intensity value at the i-th sampling time point, where i is the sampling time point;

[0108] First, calculate the change in illumination between two adjacent acquisition time points in the illumination intensity sequence; then integrate the illumination change values ​​into an illumination change sequence ΔL according to the time sequence.

[0109] ΔL=[ΔL1,ΔL2,...,ΔL i ];

[0110] Where, ΔL i This represents the absolute change in light intensity between the i-th acquisition time point and the (i-1)-th acquisition time point;

[0111] It is understandable that taking absolute values ​​is to uniformly measure the magnitude of changes in light intensity, whether it increases or decreases.

[0112] The mean μΔL and standard deviation σΔL of the light intensity variation sequence were calculated; where the mean μΔL reflects the average variation range of light intensity, and the standard deviation σΔL reflects the dispersion of light intensity variation.

[0113] According to the 3σ principle, the normal fluctuation value is defined as T = μΔL + 3 × σΔL;

[0114] It should be noted that the specific meaning of the 3σ principle is that most data will fall within the range of the mean ± 3 standard deviations;

[0115] The illumination change value at each acquisition time point is compared with the normal fluctuation value to obtain the illumination mutation index (MI) at each acquisition time point. i ;

[0116] If MI i When the value is greater than 1, it indicates that the change in illumination at the sampling time Δ point has exceeded the normal fluctuation range, and the sampling time point is determined to be a point of sudden change in illumination.

[0117] If MI i When the value is less than or equal to 1, it indicates that the change in illumination at that time point is within the normal fluctuation range, and the time point is determined to be a non-abrupt point.

[0118] Calculate the current mutation index MI at each acquisition time point of the current sequence. i Mark the points where the current value changes abruptly, and the calculation process is as follows:

[0119] Calculate the current change value at adjacent acquisition time points, and integrate the current change value into a current change sequence ΔY according to the time sequence;

[0120] ΔY=[ΔY1,ΔY2,...,ΔY i ];

[0121] The mean μΔY and standard deviation σΔY of the current change sequence ΔY were calculated based on the current change sequence.

[0122] The threshold T for significant current change is defined based on the mean μΔY and standard deviation σΔY of the current change sequence ΔY. Y T Y =μΔY+3×σΔY;

[0123] The current change value at each acquisition time point is compared with the current significant change threshold to obtain the current mutation index MI at each acquisition time point. i ;

[0124] If MI i When the value is greater than 1, it indicates that the current change at the sampling time point has exceeded the normal fluctuation range, and the sampling time point is determined to be a current change point.

[0125] If MI i If the current change amount at the collection time point is less than or equal to 1, it is indicated that the current change amount at the collection time point is in a normal fluctuation range, and it is determined that the collection time point is a non-mutation point.

[0126] The collection time points of the light mutation points are arranged in time sequence to obtain a light mutation sequence.

[0127] The collection time points of the current mutation points are arranged in time sequence to obtain a current mutation sequence.

[0128] The light mutation sequence and the current mutation sequence are subjected to difference processing to obtain a time delay sequence.

[0129] The coefficient of variation of the time delay sequence is calculated.

[0130] The coefficient of variation is obtained by calculating the ratio of the mean value to the standard deviation of the time delay sequence.

[0131] If the coefficient of variation of the time delay sequence is below a preset threshold value and the values of the time delay sequence are all negative values.

[0132] Preferably, the preset threshold value is 5%.

[0133] It is indicated that after the mutation of the light intensity, the current value also appears a mutation condition, but the current response has an obvious lag and the difference between each lag time is small, and the fluctuation is not obvious.

[0134] The technical scheme of the embodiment is: in a harsh environment of high temperature and high humidity, environmental parameters and photoelectric information conversion component device current data are synchronously collected; a plurality of single-dimensional sequences of the environmental parameters and a current value sequence of the current data are constructed in the order of collection time, and the environmental parameter sequences are integrated into a multi-dimensional sequence group; the dynamic relationship between the high temperature and high humidity environmental parameters and the photoelectric information conversion component current is facilitated to be associated, multi-dimensional basic data matched in time sequence for analyzing the performance of the device in the harsh environment is provided; based on the multi-dimensional sequence group, a coupling degree index of the sequences in the multi-dimensional sequence group and the current value sequence is analyzed; if the light intensity sequence and the current value sequence have a high coupling relationship, it is determined whether the current response has a delay when the light intensity mutates; the key influencing factor of the light intensity can be locked, and the dynamic response performance of the device under the action of the factor is effectively evaluated by judging the current response delay when the factor mutates.

[0135] Embodiment 2: Please refer to Figure 1 As shown in the figure, the stability optimization method of the photoelectric information conversion component in the harsh environment according to the embodiment of the application comprises:

[0136] S3: if the current response has a delay, sensitivity response analysis is performed on the current response to obtain a current response sensitivity; if the current response sensitivity deviates from the expectation, a light-current prediction model is constructed, the current adjustment amount is calculated in advance by predicting the light mutation amplitude, and the pre-adjustment of the current is realized;

[0137] If the current response has a delay, the process of sensitivity response analysis on the current response to obtain the current response sensitivity is:

[0138] The mean delay time of the current response is obtained by mean value processing based on the delay sequence , and the delay time standard deviation F is calculated according to the mean delay time;

[0139] The light change value of the light mutation point is recorded, and all light change values are integrated for mean value processing to obtain the light mean variable of the light intensity between the light mutation points;

[0140] The current change value of the current mutation point is recorded, and all current change values are integrated for mean value processing to obtain the current mean variable of the current value between the current mutation points;

[0141] The light mean variable and the current mean variable are processed by ratio to obtain the light-current ratio, and the light-current ratio and the mean delay time of the current response are processed by ratio to obtain the current response rate R;

[0142] Based on the current response rate, a sensitivity calculation equation is constructed; the delay time standard deviation and the current response rate are input into the sensitivity calculation equation to obtain the current response sensitivity LY; wherein the sensitivity calculation equation is:

[0143] ;

[0144] Wherein, R is the current response rate, and F is the delay time standard deviation;

[0145] The calculated current response sensitivity is compared with the preset sensitivity threshold;

[0146] As shown in Figure 2 , if the current response sensitivity is greater than the preset sensitivity threshold; it indicates that in the case of light intensity mutation, the sensitivity of the current response is good;

[0147] If the current response sensitivity is less than or equal to the preset sensitivity threshold; it indicates that in the case of light intensity mutation, the sensitivity of the current response is poor;

[0148] It should be noted that the physical meaning of the current response sensitivity is that it comprehensively measures the response performance of the current when the light suddenly changes; the current response sensitivity is calculated by the current response rate and the standard deviation of the delay time, the current response rate reflects the response speed of the current change amplitude relative to the light change amplitude when the light intensity suddenly changes, the standard deviation of the delay time reflects the dispersion degree of the delay time of the current response relative to the average delay time when the light intensity suddenly changes, which reflects the fluctuation size and consistency of each response lag; specifically, the current response sensitivity integrates the amplitude sensitivity, response speed and stability, and becomes the core physical quantity for describing the comprehensive response ability of the current to the light change;

[0149] It should also be noted that the role of obtaining the current response sensitivity is to provide a quantitative basis for performance evaluation and optimization; by comparing with the preset threshold, it can directly judge whether the current response meets the standard: if it exceeds the threshold, the comprehensive performance is good, otherwise there are problems such as slow response, insufficient sensitivity or large fluctuation; at the same time, it can guide the optimization of the device, such as improving the circuit speed for long delay time, reducing interference for large standard deviation. In the scene of fast light control and photovoltaic regulation, it can also evaluate the adaptability of the device to ensure its reliable work in complex light environment;

[0150] If the current response sensitivity deviates from the expectation, a light-current prediction model is constructed to calculate the current adjustment amount in advance by predicting the light change amplitude, and the process of pre-adjusting the current is as follows:

[0151] The process of constructing the light-current prediction model is as follows:

[0152] The prediction light change amplitude module of the light-current prediction model is as follows:

[0153] Based on the obtained average delay time of the current response , the prediction period is set as acquisition time points;

[0154] For example, the light change trend needs to be predicted seconds in advance to ensure that the current adjustment action can take effect at the same time as the light change occurs, and offset the lag effect;

[0155] Extract the preprocessed light intensity sequence [L1, L2,..., L i ]; divide the sliding window according to the prediction period , the window length is , which contains the light intensity data of the past time points and the to-be-predicted interval of the future time points;

[0156] Calculate the following three types of characteristic values for each sliding window:

[0157] The slope W of the change: within the sliding window The slope of the linear fitting of the light intensity at each acquisition time point;

[0158] Cumulative change ΔL before mutation sum : Sliding window inside front Among these points, the sum of all values ​​where the change in illumination is greater than Q times the normal fluctuation value;

[0159] Preferably, Q=0.5;

[0160] Fluctuation value DX: Calculated within the sliding window. At each data collection time point, the illumination mutation index (MI) is... i The ratio of the standard deviation to the mean is used to average the ratio at each collection time point to obtain the fluctuation value DX.

[0161] If the cumulative change before a certain sliding window abruptly exceeds 0.8T and the fluctuation value is less than 0.3, then the future is determined to be... At each data collection time point, a sudden change in illumination will occur, denoted as "predicted change," and the magnitude of the predicted illumination change ΔL is calculated. pred ;

[0162] Where ΔL is calculated pred The formula is:

[0163] ;

[0164] Where W is the slope of change, ΔL sum为 Cumulative change before the mutation;

[0165] It is understandable that 0.5 is an adjustment coefficient, which was obtained by those skilled in the art through historical data analysis;

[0166] Input the slope of change and the cumulative change before the abrupt change into the formula for predicting the magnitude of the light intensity abrupt change to obtain the predicted light intensity abrupt change value.

[0167] The current regulation module of the illumination-current prediction model is shown below:

[0168] Based on the predicted abrupt change in light intensity, a formula for predicting current regulation is constructed; the formula for predicting current regulation is as follows:

[0169] ;

[0170] Where k is the basic adjustment coefficient, ΔY theo This refers to the current adjustment amount;

[0171] Exemplarily, the mean value μΔL of the illumination change sequence ΔL and the mean value μΔY of the current change sequence ΔY are extracted; the mean value μΔL of the illumination change sequence ΔL and the mean value μΔY of the current change sequence ΔY are subjected to ratio processing to obtain a basic adjustment coefficient k;

[0172] It should be noted that the physical meaning of the basic adjustment coefficient k is that the current mean variable caused by unit illumination intensity change, which is used to quantify the reference correlation ratio between illumination and current change;

[0173] The predicted light intensity mutation value and the basic adjustment coefficient k are input into the current adjustment amount prediction formula, and the current adjustment amount is obtained after calculation;

[0174] If the future light mutation will occur at the future light mutation will occur, the current adjustment module is triggered to predict the current adjustment amount, and the current is adjusted accordingly;

[0175] If the predicted light intensity rises (ΔL pred > 0), the current is increased according to ΔY theo ;

[0176] If the predicted light intensity decreases (ΔL pred < 0), the current is decreased according to ΔY theo ;

[0177] S4: After multiple pre-adjustment, the illumination intensity and current data that meet the standard are used for current response sensitivity analysis, and the illumination-current prediction model is iteratively optimized;

[0178] After the pre-adjustment strategy is implemented, multiple groups of illumination intensity and current data are collected and analyzed, and the illumination-current prediction model is iteratively optimized according to the analysis results;

[0179] After the pre-adjustment strategy is implemented, multiple groups of illumination intensity and current data are collected;

[0180] The collected multiple groups of illumination intensity and current data are preprocessed;

[0181] Based on the preprocessed multiple groups of illumination intensity and current data, the sensitivity calculation equation is uniformly used to calculate the current response sensitivity;

[0182] If the current response sensitivity of each group of illumination intensity and current data is greater than the preset sensitivity threshold; and the standard deviation of the current response sensitivity of the multiple groups of illumination intensity and current data is less than 5%, it is determined that the current sensitivity after the adjustment strategy is implemented meets the standard; otherwise, the current sensitivity after the adjustment strategy is implemented does not meet the standard;

[0183] If the current sensitivity is determined to be up to standard, iterative optimization is performed through accumulation of multiple sets of light intensity and current data to continuously improve stability and accuracy. Specifically:

[0184] Based on multiple sets of light intensity and current data, the calibration base adjustment coefficient k is recalculated using the mean of light change and the mean of current change, so that k is closer to the actual correlation law during long-term operation;

[0185] Using multiple sets of light intensity and current data, the statistical distribution of the characteristic values (including change slope W, accumulated change amount ΔL sum , fluctuation value DX) in the predicted light mutation amplitude module is continuously updated, so that the predicted light mutation amplitude is more consistent with the actual environmental changes;

[0186] If the current sensitivity is determined to be not up to standard, the scheme improvement process is started;

[0187] Check the data collection environment: verify the sensor accuracy, determine whether the time synchronization accuracy is still less than 10 milliseconds, and whether the sampling rate is still 1 time / second. If the above problems occur, the sensor needs to be calibrated;

[0188] Light prediction optimization: optimize the predicted light mutation amplitude module, change the sliding window length from a fixed value to a dynamic value, adjust according to the delay time standard deviation F, calculate the Euclidean distance between the sliding window length and the delay time standard deviation, and dynamically adjust according to the Euclidean distance;

[0189] Long-term decay compensation: combined with the device running time data, establish a decay model of current response sensitivity with device aging, preset compensation in advance, and delay performance degradation;

[0190] Regularly (such as every week) evaluate the performance of the device and draw a current response sensitivity trend chart;

[0191] If the sensitivity continuously decreases or fluctuates abnormally, automatically trigger the scheme improvement process for iterative optimization;

[0192] Regularly review the iteration log and summarize the trend of parameter optimization to predict performance degradation risks in advance and achieve proactive optimization;

[0193] The technical scheme of the embodiment is: if the current response exists delay, sensitivity response analysis is performed on the current response to obtain current response sensitivity; if the current response sensitivity deviates from the expectation, a light-current prediction model is constructed, the current adjustment amount is calculated in advance through the predicted light mutation amplitude, and the pre-adjustment of the current is realized; the current response delay problem can be solved in a targeted manner, the adjustment amount is calculated in advance through the light-current prediction model when the sensitivity deviates from the expectation to realize the pre-adjustment, the equipment current stability and operation performance are effectively guaranteed; a plurality of groups of adjusted data are collected and analyzed, and the light-current prediction model is iteratively optimized according to the analysis result; the light-current prediction model is more fitted to the actual adjustment data through iterative optimization, the pre-adjustment accuracy is continuously improved, and the stable operation of the equipment is guaranteed.

[0194] Embodiment 3: see Figure 3 As shown in the figure, the stability optimization system of the photoelectric information conversion assembly in a harsh environment according to the embodiment of the application comprises the following modules:

[0195] The data acquisition module acquires environmental parameters and equipment current data of the photoelectric information conversion assembly; a plurality of groups of single-dimensional sequences of the environmental parameters and a current value sequence of the current data are constructed in the order of acquisition time, and the plurality of groups of single-dimensional sequences are integrated into a multi-dimensional sequence group;

[0196] The change analysis module analyzes the coupling degree index of the single-dimensional sequence and the current value sequence in the multi-dimensional sequence group, if it is judged through the coupling degree index that the light intensity sequence contained in the single-dimensional sequence and the current value sequence of the current data have a high coupling relationship, whether the current response exists delay when the light intensity mutates is analyzed;

[0197] The current adjustment module: if the current response exists delay and the current response sensitivity deviates from the expectation, a light-current prediction model is constructed, the light mutation amplitude is predicted, the current adjustment amount is calculated, and the pre-adjustment of the current is realized;

[0198] The verification and optimization module: the light-current prediction model is iteratively optimized through a plurality of groups of light intensity and current data that meet the standard after pre-adjustment.

[0199] The basic principles, main features and advantages of the application are shown and described above. Those skilled in the art should understand that the application is not limited by the above embodiments, the above embodiments and descriptions in the specification are only to illustrate the principles of the application, and various changes and improvements can be made without departing from the spirit and scope of the application, and these changes and improvements all fall within the scope of the claimed application. The scope of protection of the application is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing the stability of a photoelectric information conversion component under harsh environments, characterized in that: The method comprises the following steps: Collecting environmental parameters and photoelectric information conversion component device current data; Constructing a plurality of single-dimensional sequences of environmental parameters and current value sequences of current data in chronological order, and integrating the plurality of single-dimensional sequences into a multi-dimensional sequence group; Analyzing the coupling degree index of the single-dimensional sequence and the current value sequence in the multi-dimensional sequence group. If it is determined through the coupling degree index that the illumination intensity sequence contained in the single-dimensional sequence and the current value sequence of the current data have a high coupling relationship, then analyzing whether the current response exists a delay when the illumination intensity suddenly changes; If the current response exists a delay and the current response sensitivity deviates from the expectation, then constructing an illumination-current prediction model, predicting the illumination mutation amplitude, calculating the current adjustment amount, and realizing the pre-adjustment of the current; Iterative optimization of the illumination-current prediction model is performed through the pre-adjusted and qualified groups of illumination intensity and current data.

2. The method of claim 1, wherein: The process of obtaining the coupling degree index comprises the following steps: Collecting a plurality of environmental parameters in the environment where the photoelectric information conversion component device is located, and constructing single-dimensional sequences corresponding to the plurality of environmental parameters; Obtaining the correlation deviation value of the single-dimensional sequence and the current value sequence, and constructing a coupling degree calculation formula based on the correlation deviation value; Obtaining the coupling degree of the single-dimensional sequence and the current value sequence at each collection time point through the coupling degree calculation formula; Obtaining the coupling degree index by averaging the coupling degree of each collection point; If the coupling degree index is greater than a preset coupling threshold, it indicates that the single-dimensional sequence and the current value sequence have a preliminary strong coupling relationship; Sorting the obtained coupling degree index in descending order, and extracting the single-dimensional sequence corresponding to the maximum coupling degree index; If the single-dimensional sequence and the current value sequence have a preliminary strong coupling relationship and the single-dimensional sequence is the sequence with the maximum coupling degree index, it indicates that the single-dimensional sequence and the current value sequence have a strong coupling relationship.

3. The method of claim 2, wherein the method further comprises: The way to obtain the correlation deviation value is as follows: Performing average processing on the single-dimensional sequence to obtain the average value of the single-dimensional sequence, and performing average processing on the current value sequence to obtain the average value of the current value sequence; Obtaining the maximum value and the minimum value of the single-dimensional sequence, and performing ratio processing on the difference between the maximum value and the minimum value of the single-dimensional sequence and the average value of the single-dimensional sequence to obtain the sequence deviation degree of the single-dimensional sequence; Obtaining the maximum value and the minimum value of the current value sequence, and performing ratio processing on the difference between the maximum value and the minimum value of the current value sequence and the average value of the current value sequence to obtain the sequence deviation degree of the current value sequence; Performing difference processing on the sequence deviation degree of the single-dimensional sequence and the sequence deviation degree of the current value sequence to obtain the correlation deviation value of the single-dimensional sequence and the current value sequence.

4. The method of claim 1, wherein: The process of analyzing whether the current response exists a delay is as follows: Obtaining the illumination change value and the current change value of each collection time point of the illumination intensity sequence and the current value sequence; Integrating the illumination change value and the current change value into an illumination change sequence and a current change sequence; Obtaining the normal fluctuation value of the illumination change sequence and the current significant change threshold value; Performing ratio processing on the illumination change value and the normal fluctuation value of the illumination change sequence to obtain the illumination mutation index; Performing ratio processing on the current change value and the current significant change threshold value of the current change sequence to obtain the current mutation index; According to the obtained illumination mutation index and current mutation index, performing mutation analysis to mark the collection points as illumination mutation points and current mutation points; Integrate the light mutation points and the current mutation points according to the collection time into light mutation sequences and current mutation sequences; Obtain a time delay sequence based on the light mutation sequences and the current mutation sequences, and calculate a variation coefficient of the time delay sequence; Determine whether the current response has obvious lag according to the time delay sequence and the variation coefficient of the time delay sequence.

5. The method of claim 4, wherein: The process of obtaining the normal fluctuation value and the current significant change threshold value is: Calculate the mean value and the standard deviation of the light change sequence based on the light change sequence; Define the normal fluctuation value according to the 3σ principle and the mean value and the standard deviation of the light change sequence; Calculate the mean value and the standard deviation of the current change sequence based on the current change sequence; Define the current significant change threshold value according to the 3σ principle and the mean value and the standard deviation of the current change sequence.

6. The method of claim 1, wherein: The process of performing current response sensitivity analysis is: The mean delay time of the current response is obtained by mean processing based on the delay sequence and the delay time standard deviation is calculated according to the mean delay time; Integrate the light change values of all the light mutation points to obtain the light mean variable between the light mutation points; Integrate the current change values of all the current mutation points to obtain the current mean variable between the current mutation points; Obtain the light current ratio by performing ratio processing on the light mean variable and the current mean variable, obtain the current response rate by performing ratio processing on the light current ratio and the average delay time of the current response, and obtain the current response sensitivity by performing ratio processing on the current response rate and the delay time standard deviation. The process of predicting the light mutation amplitude is:

7. The method of claim 1, wherein the method further comprises: Obtain the change slope, the cumulative change amount before mutation and the fluctuation value in the sliding window, and construct a prediction light mutation amplitude equation; Average delay time based on current response , dividing the sliding window into ; Input the change slope, the cumulative change amount before mutation and the fluctuation value in the sliding window into the prediction light mutation amplitude equation to obtain the predicted light intensity mutation value. The method of obtaining the change slope, the cumulative change amount before mutation and the fluctuation value in the sliding window is:

8. The method of claim 7, wherein the method further comprises: The cumulative change amount before mutation: obtain the sum of all values of the light change amount greater than Q times the normal fluctuation value in the sliding window; change slope: calculate the linear fit slope of the light intensity over the preceding acquisition time points within the sliding window; Wherein, Q is a preset multiple; The fluctuation value: calculate the ratio of the standard deviation to the mean value of the light mutation index at each collection time point of the previous point in the sliding window, and perform mean value processing on the ratio at each collection time point to obtain the fluctuation value. The process of calculating the current regulation amount and realizing pre-regulation is:

9. The method of claim 1, wherein: Extract the mean value of the light change sequence and the mean value of the current change sequence; Perform ratio processing on the mean value of the light change sequence and the mean value of the current change sequence to obtain the basic regulation coefficient; Construct a current regulation amount prediction formula based on the predicted light intensity mutation amplitude and the basic regulation coefficient; Input the predicted light intensity mutation value and the basic regulation coefficient into the current regulation amount prediction formula to obtain the current regulation amount; Formulate a targeted regulation strategy according to the predicted light intensity mutation value and the current regulation amount. The process of iterative optimization of the light-current prediction model is:

10. The method of claim 1, wherein: After the pre-regulation is implemented, a plurality of light intensity and current data are collected, and the current response sensitivity is calculated respectively, if the current response sensitivity of each group of light intensity and current data meets the requirements, it indicates that the current sensitivity meets the requirements; Based on the plurality of light intensity and current data, the basic regulation coefficient of the calibrated light-current prediction model is recalculated using the light change mean value and the current change mean value. ​

Images