Method, device and equipment for predicting residual life of LED lamp

Abstract

The invention relates to the technical field of LED reliability testing, and provides an LED lamp residual life prediction method, device and equipment, and the method comprises the steps: carrying out the real-time collection of the operation state parameters of an LED lamp, obtaining the operation state data, environment data and work load data, inputting the operation state data into a preset degradation model for analysis, and obtaining the residual life of the LED lamp. Obtaining a real-time degradation trend curve and analyzing a degradation trend interval; and carrying out weighted fusion calculation on the degradation trend interval by utilizing the environment data and the working load data to obtain a residual life prediction value of the LED lamp, and generating a residual life prediction report according to the residual life prediction value of the LED lamp. By collecting operation state parameters, environment data and working load data of the LED lamp in real time and generating a residual life prediction report, life changes of the LED lamp in different environments and working conditions are achieved, and the problem that an existing light attenuation detection method neglects factors of environment and load changes, and life prediction is not accurate is solved.

Description

Technical Field

[0001] The present application relates to the technical field of LED reliability testing, and in particular to a method, device, and apparatus for predicting the remaining life of an LED lamp. Background Art

[0002] In recent years, LED lights have become widely used in various lighting scenarios due to their high efficiency, energy saving, and long life. Whether in home lighting, commercial lighting, or industrial lighting, LED lights have gradually replaced traditional incandescent and fluorescent lamps and become the mainstream light source. With technological advances and reduced production costs, the market demand for LED lights continues to grow.

[0003] Related technologies predict the remaining life of LED lamps primarily by monitoring their operating hours and light output decay. This typically involves measuring the change in LED light output over time to predict the remaining life. This method can accurately reflect the degree of LED lamp aging and thus estimate its lifespan.

[0004] Regarding the above technical solution, although the light decay detection method can more accurately predict the remaining life of the LED lamp, in actual applications, especially when the environmental and load conditions change greatly, the life decay of the LED lamp will be aggravated. The existing light decay detection method ignores the factors of environmental and load changes, and there is a problem of inaccurate life prediction. Summary of the Invention

[0005] In order to improve the problem of LED lamp life degradation in practical applications, especially when the environment and load conditions change greatly, the existing light decay detection method ignores the factors of environmental and load changes, and there is an inaccurate life prediction problem. This application provides a method, device and equipment for predicting the remaining life of LED lamps, which provides a more accurate life prediction by comprehensively considering the operating status, environmental data and workload data of the LED lamp.

[0006] The present invention provides a method for predicting the remaining life of an LED lamp, comprising: collecting operating status parameters of the LED lamp in real time to obtain operating status data, environmental data, and workload data, inputting the operating status data into a preset degradation model for analysis to obtain degradation characteristic parameters and an initial life assessment value of the LED lamp; performing multidimensional correlation analysis on the operating status parameters and the degradation characteristic parameters to obtain an initial life characteristic index, comparing the initial life characteristic index with the initial life assessment value to generate a dynamic degradation factor; performing weighted decomposition analysis on the dynamic degradation factor to obtain a real-time degradation rate and a degradation cumulative value, calibrating the initial life characteristic index using the degradation cumulative value to obtain a remaining life characteristic index; evaluating the real-time degradation rate according to the remaining life characteristic index to obtain a real-time degradation trend curve, and analyzing a plurality of degradation trend intervals using the real-time degradation trend curve; performing weighted fusion calculation on all the degradation trend intervals using the environmental data and the workload data to obtain a remaining life prediction value of the LED lamp, and generating a remaining life prediction report based on the remaining life prediction value of the LED lamp.

[0007] As a preferred solution, the steps of collecting the operating status parameters of the LED lamp in real time to obtain operating status data, environmental data, and workload data, and inputting the operating status data into a preset degradation model for analysis to obtain degradation characteristic parameters and an initial life assessment value of the LED lamp include: synchronously collecting the operating current, operating voltage, ambient temperature, and light output power of the LED lamp to generate raw operating status data; performing data cleaning and classification processing on the raw operating status data to obtain operating status data, environmental data, and workload data; inputting the operating status data into a preset degradation model, analyzing the operating status data in the degradation model based on the physical characteristics of the LED lamp and historical operating data to obtain degradation characteristic parameters and current operating characteristic indicators; and performing distribution fitting and trend evaluation on the current operating characteristic indicators using a multivariate linear regression algorithm to obtain an initial life assessment value.

[0008] As a preferred solution, the steps of performing a multidimensional correlation analysis on the operating state parameters and the degradation characteristic parameters to obtain an initial life characteristic index, comparing the initial life characteristic index with the initial life assessment value, and generating a dynamic degradation factor include: performing a correlation analysis on the operating state parameters and the degradation characteristic parameters using multidimensional regression analysis, extracting an operating state characteristic factor and a degradation characteristic factor based on the analysis results, performing a dynamic pairing analysis on the operating state characteristic factor and the degradation characteristic factor to generate an initial life characteristic index and an operating parameter weight; performing a comparative analysis on the initial life characteristic index and the initial life assessment value to obtain a degradation characteristic deviation, and fusing the degradation characteristic deviation and the operating parameter weight using a weighted least squares method to obtain a degradation deviation matrix; and extracting key characteristics from the degradation deviation matrix using principal component analysis to generate a dynamic degradation factor.

[0009] As a preferred solution, the steps of performing correlation analysis on the operating state parameters and the degradation characteristic parameters using multidimensional regression analysis, extracting operating state characteristic factors and degradation characteristic factors, dynamically pairing the operating state characteristic factors with the degradation characteristic factors, and generating initial life characteristic indicators and operating parameter weights include: screening the operating state parameters and the degradation characteristic parameters using a stepwise regression method to obtain an influencing variable set and a correlation coefficient matrix; comparing and analyzing the influencing variable set with the correlation coefficient matrix to obtain a preliminary characteristic cluster, adjusting and reconstructing the preliminary characteristic cluster using a genetic algorithm to generate an optimized characteristic factor set; performing feature extraction on the optimized characteristic factor set according to factor analysis to obtain operating state characteristic factors and degradation characteristic factors, dynamically pairing the operating state characteristic factors with the degradation characteristic factors using a canonical correlation analysis to generate a synergy matrix and a feature combination vector; parsing the synergy matrix using a factor analysis method to generate an initial life characteristic indicator, weighting the feature combination vector using an entropy method to generate operating parameter weights.

[0010] As a preferred solution, the steps of performing weighted decomposition analysis on the dynamic degradation factor to obtain a real-time degradation rate and a degradation accumulation value, and using the degradation accumulation value to calibrate the initial life characteristic index to obtain a remaining life characteristic index include: obtaining a historical degradation curve of the LED lamp, assigning a weight to each time period of the dynamic degradation factor based on the historical degradation curve using an exponential decay method to obtain a time weighting factor, and adjusting the proportions between the dynamic degradation factors using the time weighting factor to obtain a dynamic degradation factor with adjusted proportions; decomposing the dynamic degradation factor with adjusted proportions into different frequency components through Fourier transform to obtain a real-time degradation rate; and fitting and analyzing the real-time degradation rate with the initial life characteristic index through a method based on partial least squares regression to obtain a degradation accumulation value, and using the degradation accumulation value to calibrate the initial life characteristic index to generate a remaining life characteristic index.

[0011] As a preferred solution, the steps of evaluating the real-time degradation rate according to the remaining life characteristic indicator to obtain a real-time degradation trend curve, and analyzing a number of degradation trend intervals using the real-time degradation trend curve include: performing trend extraction on the remaining life characteristic indicator using a support vector regression algorithm to obtain trend distribution characteristics, life change nodes and time interval factors, using the life change nodes to screen characteristic points of the trend distribution characteristics to obtain a key life node sequence; dividing the key life node sequence according to a dynamic time warping algorithm to obtain a number of degradation trends, dividing the range of each degradation trend by the time interval factor to obtain a number of degradation trend intervals; and using a trend fitting algorithm to match and analyze the several degradation trend intervals in the divided range with the real-time degradation rate to generate a real-time degradation trend curve.

[0012] As a preferred solution, the steps of using the environmental data and the workload data to perform weighted fusion calculation on all the degradation trend intervals to obtain the remaining life prediction value of the LED lamp, and generating a remaining life prediction report based on the remaining life prediction value of the LED lamp include: quantifying the environmental data through a hierarchical analysis method to obtain environmental impact weights, performing distribution analysis on the workload data using a Kalman filter to generate workload impact weights; weighted accumulation of all the degradation trend intervals according to the environmental impact weights and the workload impact weights to obtain a comprehensive degradation index; performing parameter optimization on the comprehensive degradation index through Bayesian prediction to obtain the remaining life prediction value of the LED lamp, inputting the remaining life prediction value of the LED lamp into a preset life analysis model, and outputting a remaining life prediction report.

[0013] The present application also provides a device for predicting the remaining life of an LED lamp, comprising: an acquisition module for real-time acquisition of operating state parameters of the LED lamp to obtain operating state data, environmental data, and workload data, inputting the operating state data into a preset degradation model for analysis to obtain degradation characteristic parameters and an initial life evaluation value of the LED lamp; an analysis module for performing multidimensional correlation analysis on the operating state parameters and the degradation characteristic parameters to obtain an initial life characteristic index, comparing the initial life characteristic index with the initial life evaluation value to generate a dynamic degradation factor; and a weighting module for weighting the dynamic degradation factor. The system comprises a weighted decomposition analysis module, a real-time degradation rate and a degradation cumulative value are obtained by weighted decomposition analysis of the factors, and the initial life characteristic index is calibrated by using the degradation cumulative value to obtain the remaining life characteristic index; an evaluation module is used to evaluate the real-time degradation rate according to the remaining life characteristic index to obtain a real-time degradation trend curve, and a plurality of degradation trend intervals are analyzed by using the real-time degradation trend curve; a generation module is used to perform weighted fusion calculation on all the degradation trend intervals by using the environmental data and the workload data to obtain the remaining life prediction value of the LED lamp, and generate a remaining life prediction report according to the remaining life prediction value of the LED lamp.

[0014] The present application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the method for predicting the remaining life of an LED lamp described above is implemented.

[0015] Compared with the existing technology, this application has the following beneficial effects: comprehensive analysis and accurate prediction. By collecting the operating status parameters, environmental data, and workload data of LED lamps in real time and inputting these data into a preset degradation model for analysis, degradation characteristic parameters and initial life assessment values ​​can be obtained. Combined with a data analysis algorithm, a multidimensional correlation analysis is performed on the operating status parameters and degradation characteristic parameters to generate an initial life characteristic index and a dynamic degradation factor. After weighted decomposition analysis of the dynamic degradation factor, a real-time degradation rate and degradation accumulation value are calculated. These values ​​are used to calibrate the initial life characteristic index to obtain a new remaining life characteristic index. Based on these indicators, the real-time degradation rate is evaluated, a real-time degradation trend curve is generated, and several degradation trend intervals are divided. Combined with environmental data and workload data, a weighted fusion calculation is performed on the degradation trend intervals to obtain a more accurate remaining life prediction value for the LED lamp and generate a remaining life prediction report. This method realizes the life changes of LED lamps under different environments and operating conditions and can adjust the prediction results in real time, improving the accuracy and reliability of the prediction. In practical applications, especially when the environmental and load conditions vary greatly, the life degradation of LED lamps will be aggravated. Existing light decay detection methods ignore the factors of environmental and load changes, resulting in inaccurate life prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] The structures, proportions, sizes, etc. depicted in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not intended to limit the conditions under which the present invention can be implemented and therefore have no substantive technical significance. Any structural modifications, changes in proportional relationships, or adjustments in size should still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and objectives that can be achieved by the present invention.

[0018] Figure 1 1 is a flow chart of a method for predicting the remaining life of an LED lamp provided by an embodiment of the present invention; Figure 2 This is a schematic block diagram of the structure of a device for predicting the remaining life of an LED lamp provided by an embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0019] Description of reference numerals: 10. Prediction device for remaining life of LED lamp; 11. Acquisition module; 12. Analysis module; 13. Weighting module; 14. Evaluation module; 15. Generation module; 20. Electronic device; 21. Memory; 22. Processor. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0021] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0022] It should also be understood that the terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit the present application. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0023] It should be further understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0024] The technical solution of the present invention will be further described below with reference to the accompanying drawings and through specific implementation methods.

[0025] Example 1: like Figure 1 As shown, the method for predicting the remaining life of an LED lamp provided in the embodiment of the present application includes steps S100 to S500.

[0026] Step S100: collect the operating status parameters of the LED lamp in real time to obtain operating status data, environmental data, and workload data; input the operating status data into a preset degradation model for analysis to obtain degradation characteristic parameters and an initial life assessment value of the LED lamp.

[0027] In this step, the operating status parameters of the LED lamp, such as current, voltage, and temperature, as well as environmental data such as ambient temperature and humidity and the workload data borne by the LED lamp, are collected in real time through sensors and data acquisition modules. Specifically, these real-time collected data are input into a preset degradation model for processing and analysis to generate corresponding degradation characteristic parameters and initial life assessment values.

[0028] For example, the collected current, voltage, and temperature data can be used to analyze the thermal degradation characteristics of LED lamps and estimate their initial lifespan.

[0029] Step S200 , performing multidimensional correlation analysis on the operating state parameters and the degradation characteristic parameters to obtain an initial life characteristic index, comparing the initial life characteristic index with the initial life assessment value to generate a dynamic degradation factor.

[0030] In this step, a data analysis algorithm is used in combination with the operating status parameters and degradation characteristic parameters in the previous step. Through multidimensional correlation analysis, the key indicators affecting the life of the LED lamp are extracted and the initial life characteristic indicators are generated. Specifically, the initial life characteristic indicators are compared with the initial life assessment values, the differences are identified, and a dynamic degradation factor is generated.

[0031] For example, by analyzing the impact of temperature on LED light output, a dynamic degradation factor can be generated to reflect the life span changes under actual usage conditions.

[0032] Step S300: Perform weighted decomposition analysis on the dynamic degradation factor to obtain a real-time degradation rate and a degradation accumulation value, and use the degradation accumulation value to calibrate the initial life characteristic index to obtain a remaining life characteristic index.

[0033] In this step, the dynamic degradation factor is weighted and decomposed to calculate the real-time degradation rate and degradation cumulative value of the LED lamp under different operating conditions. Specifically, by accumulating these real-time degradation rates, the overall degradation cumulative value is obtained, and this value is used to calibrate the initial life characteristic index to generate a new remaining life characteristic index.

[0034] For example, by monitoring the current changes of LED lamps in real time, its real-time degradation rate can be calculated and accumulated to obtain the degradation cumulative value, which is used to calibrate the life prediction.

[0035] Step S400: Evaluate the real-time degradation rate according to the remaining life characteristic index to obtain a real-time degradation trend curve, and analyze a number of degradation trend intervals using the real-time degradation trend curve.

[0036] In this step, the real-time degradation rate of the LED lamp is evaluated based on the calibrated remaining life characteristic index, and a real-time degradation trend curve is generated. Specifically, by analyzing the trend curve, different degradation trend intervals can be divided to reflect the degradation characteristics of the LED lamp at different usage stages.

[0037] For example, by analyzing the temperature and light output changes over a period of time, the thermal degradation trend range of LED lights can be obtained.

[0038] Step S500: Perform weighted fusion calculation on all degradation trend intervals using environmental data and workload data to obtain a predicted value of the remaining life of the LED lamp, and generate a remaining life prediction report based on the predicted value of the remaining life of the LED lamp.

[0039] In this step, the degradation trend intervals divided in the previous step are weighted and fused together to obtain the remaining life prediction value of the LED lamp, combining environmental data and workload data. Specifically, these prediction values ​​are summarized to generate a detailed remaining life prediction report for user reference.

[0040] For example, a more accurate remaining life prediction value can be calculated by combining the workload of the LED lamp and the changes in ambient temperature.

[0041] In this embodiment, operating state parameters, environmental data, and workload data of the LED lamp are collected in real time and input into a preset degradation model for analysis to obtain degradation characteristic parameters and an initial life assessment value. Then, a data analysis algorithm is used to perform a multidimensional correlation analysis on the operating state parameters and the degradation characteristic parameters to obtain an initial life characteristic index and generate a dynamic degradation factor. Subsequently, a weighted decomposition analysis is performed on the dynamic degradation factor to calculate a real-time degradation rate and a degradation accumulation value, which are used to calibrate the initial life characteristic index to generate a new remaining life characteristic index. Then, the real-time degradation rate is evaluated based on the remaining life characteristic index to generate a real-time degradation trend curve and divide it into several degradation trend intervals. Finally, a weighted fusion calculation is performed on the degradation trend intervals in combination with the environmental data and workload data to obtain a more accurate remaining life prediction value of the LED lamp and generate a remaining life prediction report.

[0042] By real-time collection and analysis of LED lamp operating status data, environmental data and workload data, a dynamic and accurate prediction of LED lamp life is achieved. Compared with traditional light decay detection methods, it can not only more accurately reflect the life changes of LED lamps under different environments and working conditions, but also adjust the prediction results in real time, improving the accuracy and reliability of the prediction and improving practical applications. In particular, when the environment and load conditions change greatly, the life decay of LED lamps will be aggravated. Existing light decay detection methods ignore the factors of environmental and load changes, and there is a problem of inaccurate life prediction.

[0043] Example 2: In step S100, the operating current, operating voltage, ambient temperature and light output power of the LED lamp are synchronously collected to generate raw operating status data.

[0044] By adopting high-precision sensors and data acquisition modules, the operating current, operating voltage, ambient temperature and light output power of the LED lamp are collected synchronously in real time. Specifically, a current sensor is used to measure the operating current of the LED lamp, a voltage sensor is used to monitor the operating voltage, an environmental sensor records the ambient temperature, and an optical sensor is used to measure the light output power, ensuring that all data is collected synchronously and recorded with high precision.

[0045] For example, in actual applications, multiple sensors installed in the LED lamp system can monitor current, voltage, temperature and light output power in real time, generating accurate raw data on the operating status for subsequent analysis.

[0046] The original operating status data is cleaned and classified to obtain operating status data, environmental data, and workload data.

[0047] The raw data of the operating status is preprocessed by applying a data cleaning algorithm, including removing outliers, noise, and redundant data. Specifically, the data is classified according to different data types, and the cleaned data is divided into operating status data, environmental data, and workload data, and stored in the corresponding databases respectively.

[0048] For example, the cleaning process involves deleting abnormal current and voltage readings caused by interference or equipment failure during the acquisition process, and then classifying and saving data within the normal range for further processing and analysis.

[0049] The operating status data is input into a preset degradation model, in which the operating status data is analyzed based on the physical characteristics of the LED lamp and historical operating data to obtain degradation characteristic parameters and current operating characteristic indicators.

[0050] The processed operating status data is analyzed by using a preset degradation model; specifically, the model performs a multi-dimensional analysis of the current operating status data based on the physical characteristics and historical operating data of the LED lamp to extract degradation characteristic parameters and current operating characteristic indicators.

[0051] For example, the current, voltage and temperature data of the LED lamp are input into the degradation model, and by comparing with historical data, the characteristic parameters of the current LED lamp degradation rate and performance degradation are extracted.

[0052] The distribution fitting and trend evaluation of the current operating characteristic indicators are performed using a multiple linear regression algorithm to obtain the initial life assessment value.

[0053] The multivariate linear regression algorithm is applied to perform data fitting and trend analysis on the current operating characteristic indicators; specifically, a regression model is established, and the historical data and current characteristic data are used for fitting to predict the initial life assessment value of the LED lamp.

[0054] For example, regression analysis can be used to model current and light output data to predict the start of an LED lamp's life under current operating conditions and estimate its potential lifetime.

[0055] In step S200, a correlation analysis is performed on the operating state parameters and the degradation characteristic parameters using multidimensional regression analysis. Based on the analysis results, the operating state characteristic factors and the degradation characteristic factors are extracted. The operating state characteristic factors and the degradation characteristic factors are dynamically paired and analyzed to generate initial life characteristic indicators and operating parameter weights.

[0056] The relationship between operating status parameters and degradation characteristic parameters is evaluated by using a multidimensional regression analysis method. Specifically, key operating status characteristic factors and degradation characteristic factors are extracted, and these factors are dynamically paired to generate initial life characteristic indicators and corresponding operating parameter weights.

[0057] For example, by analyzing the relationship between ambient temperature and light output, the temperature characteristic factor and light output degradation factor are extracted, and paired analysis is performed to obtain the initial life characteristic index.

[0058] The initial life characteristic index is compared with the initial life assessment value to obtain the degradation characteristic deviation. The degradation characteristic deviation and the operating parameter weight are fused using the weighted least squares method to obtain the degradation deviation matrix.

[0059] The degradation characteristic deviation is calculated by performing a detailed comparative analysis of the initial life characteristic index and the initial life assessment value. Specifically, the degradation characteristic deviation and the operating parameter weight are weighted using the weighted least squares method to generate a degradation deviation matrix.

[0060] For example, by comparing the actual light output degradation value with the predicted value, the light output degradation characteristic deviation is calculated and fused with the weights of current and temperature to form a degradation deviation matrix.

[0061] Principal component analysis is used to extract key features of the degradation deviation matrix and generate dynamic degradation factors.

[0062] By using the principal component analysis method, the degradation deviation matrix is ​​reduced in dimension and features are extracted; specifically, the main components in the degradation deviation matrix are analyzed, key characteristic data are extracted, and a dynamic degradation factor is generated.

[0063] For example, the main components of light output and temperature are extracted from the degradation deviation matrix to form a dynamic degradation factor representing the degradation state of the LED lamp.

[0064] Among them, the steps of using multidimensional regression analysis to perform correlation analysis on operating status parameters and degradation characteristic parameters, extracting operating status characteristic factors and degradation characteristic factors, dynamically pairing the operating status characteristic factors with the degradation characteristic factors, and generating initial life characteristic indicators and operating parameter weights include: screening the operating status parameters and degradation characteristic parameters through a stepwise regression method to obtain an influencing variable set and a correlation coefficient matrix.

[0065] All operating status parameters and degradation characteristic parameters are screened by the stepwise regression method; specifically, the variables that have a significant impact on life prediction are screened out, and a correlation coefficient matrix is ​​generated for further analysis and processing.

[0066] For example, during the screening process, current fluctuation and ambient temperature change are determined to be the parameters that have the greatest impact on life, and a correlation coefficient matrix of these parameters is generated.

[0067] The influencing variable set and the correlation coefficient matrix are compared and analyzed to obtain preliminary characteristic clusters, which are then adjusted and reconstructed through genetic algorithms to generate an optimized characteristic factor set.

[0068] By comparing and analyzing the influencing variable set with the correlation coefficient matrix, a preliminary characteristic cluster is formed; specifically, a genetic algorithm is applied to adjust and optimize the preliminary characteristic cluster, and finally an efficient characteristic factor set is generated.

[0069] For example, a more optimized characteristic factor set can be obtained by optimizing variables in the characteristic factor set through a genetic algorithm, such as adjusting the ratio of current and temperature characteristics.

[0070] Based on factor analysis, the optimized characteristic factor set is extracted to obtain the operating state characteristic factor and the degradation characteristic factor. The operating state characteristic factor and the degradation characteristic factor are dynamically paired using canonical correlation analysis to generate a synergy matrix and characteristic combination vector.

[0071] The optimized characteristic factor set is subjected to detailed feature extraction through factor analysis. Specifically, the canonical correlation analysis method is applied to dynamically pair the operating status characteristic factors with the degradation characteristic factors to generate a synergy matrix and characteristic combination vector for further analysis.

[0072] For example, current characteristic factors and light output degradation factors are extracted and dynamic pairing analysis is performed to generate a synergy matrix for lifetime prediction.

[0073] The factor analysis method is used to analyze the synergy matrix to generate the initial life characteristic index. The entropy method is used to evaluate the weight of the characteristic combination vector to generate the operating parameter weight.

[0074] The synergy matrix is ​​deeply analyzed through factor analysis to extract the initial life characteristic indicators; specifically, the entropy method is used to evaluate the weight of the characteristic combination vector, calculate the influence weight of each parameter, and generate the operating parameter weight.

[0075] For example, the weights of current and temperature characteristic factors are evaluated to derive their relative importance to life prediction and form accurate operating parameter weights.

[0076] In step S300, a historical degradation curve of the LED lamp is obtained, and a weight is assigned to each time period of the dynamic degradation factor based on the historical degradation curve using an exponential decay method to obtain a time weighting factor. The time weighting factor is used to adjust the proportion between the dynamic degradation factors to obtain the dynamic degradation factor after the adjustment.

[0077] By collecting and analyzing the historical degradation data of LED lights, a detailed degradation curve is obtained. Specifically, using the exponential decay method, a corresponding time weight is assigned to each dynamic degradation factor based on the degradation data of each time period to form a time weighting factor, and the proportion of each dynamic degradation factor is adjusted.

[0078] For example, based on light attenuation data from past years, weights are assigned and dynamic degradation factors are adjusted to accurately reflect degradation at different stages of use.

[0079] The dynamic degradation factor after weight adjustment is decomposed into different frequency components through Fourier transform to obtain the real-time degradation rate.

[0080] By applying Fourier transform technology, the dynamic degradation factor after adjusting the proportion is decomposed into different frequency components; specifically, the changes in each frequency component are analyzed, and the real-time degradation rate is extracted to reflect the real-time degradation of the LED lamp.

[0081] For example, the adjusted light output data is Fourier transformed and decomposed into multiple frequency bands to obtain the real-time degradation rate at different frequencies.

[0082] The real-time degradation rate is fitted with the initial life characteristic index by the partial least squares regression method to obtain the degradation accumulation value, which is then used to calibrate the initial life characteristic index and generate the remaining life characteristic index.

[0083] By applying the partial least squares regression method, a fitting analysis is performed on the real-time degradation rate and the initial life characteristic index; specifically, the initial life characteristic index is calibrated using the degradation cumulative value to generate a more accurate remaining life characteristic index.

[0084] For example, the real-time degradation rate data is fitted with the initial light output life characteristic index, and the degradation accumulation value is calculated through the partial least squares regression method. This value is used to calibrate the life characteristic index to generate the remaining life characteristic index of the LED lamp.

[0085] In step S400, the trend of the remaining life characteristic index is extracted by the support vector regression algorithm to obtain the trend distribution characteristics, life change nodes and time interval factors. The life change nodes are used to screen the characteristic points of the trend distribution characteristics to obtain the key life node sequence.

[0086] By applying the support vector regression algorithm, the trend of the remaining life characteristic indicators is extracted in detail; specifically, the trend distribution of the life characteristic indicators is analyzed, the life change nodes and time interval factors are identified, and characteristic points are screened to generate a key life node sequence.

[0087] For example, by analyzing the changing trends of LED lamp life characteristic indicators, we can identify the time points when the light output drops significantly, form a life change node sequence, and further refine the key nodes of the life characteristics.

[0088] The key life node sequence is divided according to the dynamic time warping algorithm to obtain several degradation trends. The range of each degradation trend is divided by the time interval factor to obtain several degradation trend intervals.

[0089] By adopting the dynamic time warping algorithm, the key life node sequence is divided into subdivisions; specifically, the time interval factor is used to determine the range of each degradation trend, thereby generating multiple degradation trend intervals.

[0090] For example, based on historical life data, the life change nodes in different time periods are divided into several independent degradation trend intervals, and each interval reflects the degradation characteristics of different stages.

[0091] The trend fitting algorithm is used to match and analyze several degradation trend intervals and real-time degradation rates to generate a real-time degradation trend curve.

[0092] By applying a trend fitting algorithm, a matching analysis is performed on multiple degradation trend intervals and real-time degradation rates. Specifically, by combining different degradation trend intervals and real-time degradation rate data, a detailed real-time degradation trend curve is generated to fully reflect the degradation process of LED lamps.

[0093] For example, the data on light output changes over time are matched with each degradation trend interval, and a real-time degradation trend curve is generated through a trend fitting algorithm to show the degradation path of LED lights under different operating conditions.

[0094] In step S500, the environmental data is quantified by the hierarchical analysis method to obtain the environmental impact weight, and the workload data is distributed analyzed by the Kalman filter to generate the workload impact weight.

[0095] The collected environmental data are quantitatively evaluated through the hierarchical analysis method to determine the weight of their impact on the life of LED lamps. Specifically, the workload data are analyzed using the Kalman filtering technology to obtain the workload impact weight, which reflects its impact on the life of LED lamps under different working conditions.

[0096] For example, the impact of changes in ambient temperature and humidity on the life of LED lamps can be quantitatively evaluated, and workload changes can be analyzed using Kalman filtering to generate their respective impact weights.

[0097] All degradation trend intervals are weighted and accumulated according to the environmental impact weight and workload impact weight to obtain the comprehensive degradation index.

[0098] By comprehensively considering the environmental impact weight and workload impact weight, a weighted cumulative calculation is performed on each degradation trend interval; specifically, a comprehensive degradation index is calculated to more accurately reflect the comprehensive degradation characteristics of LED lamps under different environmental and load conditions.

[0099] For example, by combining the weights of temperature, humidity and workload, the data of all degradation trend intervals are weighted and accumulated to generate a comprehensive degradation index of the LED lamp and evaluate its comprehensive degradation situation.

[0100] The parameters of the comprehensive degradation index are optimized through Bayesian prediction to obtain the predicted value of the remaining life of the LED lamp. The predicted value of the remaining life of the LED lamp is input into the preset life analysis model to output the remaining life prediction report.

[0101] By applying the Bayesian prediction method, the parameters of the comprehensive degradation index are optimized to obtain the remaining life prediction value of the LED lamp; specifically, these prediction values ​​are input into the preset life analysis model to generate a detailed remaining life prediction report, providing users with reliable life prediction data.

[0102] For example, based on the comprehensive degradation index, the model parameters are optimized through the Bayesian prediction method to predict the remaining life of LED lamps and generate a remaining life prediction report containing detailed data and analysis.

[0103] In this embodiment, operating status raw data is generated by real-time acquisition of operating status parameters such as the operating current, operating voltage, ambient temperature, and light output power of the LED lamp. Through data cleaning and classification processing, these raw data are divided into operating status data, environmental data, and workload data, and analyzed using a preset degradation model to obtain degradation characteristic parameters and current operating characteristic indicators. Then, the current operating characteristic indicators are subjected to distribution fitting and trend evaluation using a multivariate linear regression algorithm to obtain an initial life assessment value. Multidimensional regression analysis is used to perform correlation analysis on the operating status parameters and degradation characteristic parameters, extract operating status characteristic factors and degradation characteristic factors, and perform dynamic pairing analysis to generate initial life characteristic indicators and operating parameter weights. The initial life characteristic indicators are compared with the initial life assessment value, the degradation characteristic deviation is calculated, and the degradation characteristic deviation and operating parameter weight are fused using the weighted least squares method to obtain a degradation deviation matrix.

[0104] Principal component analysis is used to extract characteristics from the degradation deviation matrix and generate a dynamic degradation factor. The dynamic degradation factor is then weighted using historical degradation curves and the exponential decay method to obtain a time-weighted factor. This factor is then decomposed into different frequency components using a Fourier transform to calculate the real-time degradation rate. The real-time degradation rate is then fitted to the initial life characteristic index using partial least squares regression to obtain a cumulative degradation value. The initial life characteristic index is then calibrated to generate a remaining life characteristic index. A support vector regression algorithm is used to extract trends, identify key life nodes, and divide degradation trend intervals. A trend fitting algorithm is then used to match the real-time degradation rate to generate a real-time degradation trend curve. Finally, the analytic hierarchy process and Kalman filtering are used to quantify the impact weights of environmental and workload data, and weighted accumulation is performed to generate a comprehensive degradation index. Parameters are optimized using a Bayesian prediction method to obtain the predicted remaining life of the LED lamp. This value is then input into a pre-defined life analysis model to generate a detailed remaining life prediction report.

[0105] Example 3: like Figure 2 As shown, the present application also provides a device 10 for predicting the remaining life of an LED lamp, including a collection module 11 , an analysis module 12 , a weighting module 13 , an evaluation module 14 and a generation module 15 .

[0106] The acquisition module 11 is mainly used to collect the operating status parameters of the LED lamp in real time, obtain operating status data, environmental data and workload data, input the operating status data into the preset degradation model for analysis, and obtain the degradation characteristic parameters and initial life assessment value of the LED lamp.

[0107] The analysis module 12 is mainly used to perform multidimensional correlation analysis on the operating state parameters and degradation characteristic parameters to obtain the initial life characteristic index, compare the initial life characteristic index with the initial life assessment value, and generate a dynamic degradation factor.

[0108] The weighting module 13 is mainly used to perform weighted decomposition analysis on the dynamic degradation factor to obtain the real-time degradation rate and degradation accumulation value, and to calibrate the initial life characteristic index using the degradation accumulation value to obtain the remaining life characteristic index.

[0109] The evaluation module 14 is mainly used to evaluate the real-time degradation rate according to the remaining life characteristic index, obtain a real-time degradation trend curve, and analyze a number of degradation trend intervals using the real-time degradation trend curve.

[0110] The generation module 15 is mainly used to perform weighted fusion calculation on all degradation trend intervals using environmental data and workload data to obtain the remaining life prediction value of the LED lamp, and generate a remaining life prediction report based on the remaining life prediction value of the LED lamp.

[0111] In this embodiment, an apparatus comprising an acquisition module 11, an analysis module 12, a weighting module 13, an evaluation module 14, and a generation module 15 is employed to accurately predict the remaining life of an LED lamp. The acquisition module 11 is responsible for real-time acquisition of the LED lamp's operating parameters, including operating current, operating voltage, ambient temperature, and optical output power, generating operating status data, environmental data, and workload data. This data is then input into a preset degradation model for analysis, resulting in degradation characteristic parameters and an initial lifetime estimate for the LED lamp. The analysis module 12 performs a multidimensional correlation analysis on the operating status and degradation characteristic parameters to generate an initial lifetime characteristic index, which is then compared with the initial lifetime estimate to generate a dynamic degradation factor. The weighting module 13 performs a weighted decomposition analysis on the dynamic degradation factor to obtain a real-time degradation rate and a cumulative degradation value. The cumulative degradation value is then used to calibrate the initial lifetime characteristic index to generate a remaining lifetime characteristic index. The evaluation module 14 evaluates the real-time degradation rate based on the remaining lifetime characteristic index, generating a real-time degradation trend curve. This trend curve is then used to analyze several degradation trend intervals. Finally, generation module 15 uses environmental and workload data to perform a weighted fusion calculation on all degradation trend intervals, deriving a predicted remaining lifespan for the LED lamp and generating a detailed remaining lifespan prediction report. This device enables accurate prediction of LED lamp lifespan under various operating conditions, improving prediction accuracy and reliability, extending the lifespan of LED lamps, and reducing maintenance costs.

[0112] It should be noted that, those skilled in the art will clearly understand that, for the sake of convenience and brevity of description, the specific working processes of the above-described device and each module can refer to the corresponding processes in the aforementioned embodiment of the method for predicting the remaining life of an LED lamp, and will not be repeated here.

[0113] Example 4: like Figure 3 As shown, the present application also proposes an electronic device 20, including a memory 21 and a processor 22, wherein the memory 21 stores a computer program that can be run on the processor 22, and when the processor 22 executes the computer program, the method for predicting the remaining life of the LED lamp of Example 1 is implemented.

[0114] In this embodiment, by integrating a memory 21 storing a computer program executable on a processor 22 and a processor 22 into an electronic device 20, an accurate prediction of the remaining life of an LED lamp is achieved. The processor 22 runs the computer program in the memory 21 and executes the remaining life prediction method for an LED lamp described in Example 1. Specifically, the electronic device 20 collects the operating status parameters of the LED lamp in real time, inputs these data into a preset degradation model for analysis, and finally generates a remaining life prediction report by combining steps such as multidimensional correlation analysis, weighted decomposition, and real-time evaluation. In this way, the electronic device 20 can flexibly and efficiently evaluate and predict the remaining life of an LED lamp in actual applications, improve the efficiency of the LED lamp, and ensure the long-term and stable operation of the equipment.

[0115] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the remaining life of an LED lamp, characterized in that: include: Collecting the operating status parameters of the LED lamp in real time to obtain operating status data, environmental data and workload data, inputting the operating status data into a preset degradation model for analysis, and obtaining degradation characteristic parameters and initial life assessment values ​​of the LED lamp; Performing a multidimensional correlation analysis on the operating state parameter and the degradation characteristic parameter to obtain an initial life characteristic index, comparing the initial life characteristic index with the initial life assessment value to generate a dynamic degradation factor; Performing weighted decomposition analysis on the dynamic degradation factor to obtain a real-time degradation rate and a degradation accumulation value, and calibrating the initial life characteristic index using the degradation accumulation value to obtain a remaining life characteristic index; The real-time degradation rate is evaluated according to the remaining life characteristic index to obtain a real-time degradation trend curve, and a plurality of degradation trend intervals are analyzed using the real-time degradation trend curve; The environmental data and the workload data are used to perform weighted fusion calculation on all the degradation trend intervals to obtain a remaining life prediction value of the LED lamp, and a remaining life prediction report is generated based on the remaining life prediction value of the LED lamp.

2. The method for predicting the remaining life of an LED lamp according to claim 1, characterized in that: The step of collecting the operating state parameters of the LED lamp in real time to obtain operating state data, environmental data and workload data, inputting the operating state data into a preset degradation model for analysis, and obtaining the degradation characteristic parameters and initial life assessment value of the LED lamp includes: Synchronously collect the working current, working voltage, ambient temperature and light output power of the LED lamp to generate the original data of the operating status; Performing data cleaning and classification processing on the raw operating status data to obtain operating status data, environment data, and workload data; Inputting the operating status data into a preset degradation model, analyzing the operating status data in the degradation model based on the physical characteristics of the LED lamp and historical operating data to obtain degradation characteristic parameters and current operating characteristic indicators; The current operating characteristic index is subjected to distribution fitting and trend evaluation by a multiple linear regression algorithm to obtain an initial life evaluation value.

3. The method for predicting the remaining life of an LED lamp according to claim 1, characterized in that: The step of performing a multidimensional correlation analysis on the operating state parameter and the degradation characteristic parameter to obtain an initial life characteristic index, comparing the initial life characteristic index with the initial life assessment value, and generating a dynamic degradation factor comprises: Performing a correlation analysis on the operating state parameters and the degradation characteristic parameters by using multidimensional regression analysis, extracting operating state characteristic factors and degradation characteristic factors based on the analysis results, performing a dynamic pairing analysis on the operating state characteristic factors and the degradation characteristic factors, and generating an initial life characteristic index and an operating parameter weight; Comparing and analyzing the initial life characteristic index with the initial life assessment value to obtain a degradation characteristic deviation, and fusing the degradation characteristic deviation with the operating parameter weight using a weighted least square method to obtain a degradation deviation matrix; The principal component analysis is used to extract key characteristics of the degradation deviation matrix and generate dynamic degradation factors.

4. The method for predicting the remaining life of an LED lamp according to claim 3, characterized in that: The step of performing correlation analysis on the operating state parameters and the degradation characteristic parameters by using multidimensional regression analysis, extracting operating state characteristic factors and degradation characteristic factors, performing dynamic pairing analysis on the operating state characteristic factors and the degradation characteristic factors, and generating initial life characteristic indicators and operating parameter weights includes: The operating state parameters and the degradation characteristic parameters are screened by a stepwise regression method to obtain an influencing variable set and a correlation coefficient matrix; Comparing and analyzing the influencing variable set with the correlation coefficient matrix to obtain a preliminary characteristic cluster, adjusting and reconstructing the preliminary characteristic cluster through a genetic algorithm to generate an optimized characteristic factor set; Performing feature extraction on the optimized characteristic factor set according to factor analysis to obtain operating state characteristic factors and degradation characteristic factors, and performing dynamic pairing analysis on the operating state characteristic factors and the degradation characteristic factors using canonical correlation analysis to generate a synergy matrix and a characteristic combination vector; The synergy matrix is ​​analyzed by factor analysis to generate initial life characteristic indicators, and the characteristic combination vector is weighted by entropy method to generate operating parameter weights.

5. The method for predicting the remaining life of an LED lamp according to claim 1, characterized in that: The step of performing weighted decomposition analysis on the dynamic degradation factor to obtain a real-time degradation rate and a degradation accumulation value, and using the degradation accumulation value to calibrate the initial life characteristic index to obtain a remaining life characteristic index comprises: Acquire a historical degradation curve of the LED lamp, assign a weight to each time period of the dynamic degradation factor based on the historical degradation curve using an exponential decay method to obtain a time weighting factor, and adjust the proportion between the dynamic degradation factors using the time weighting factor to obtain the dynamic degradation factor after the proportion is adjusted; Decomposing the dynamic degradation factor after adjusting the weight into different frequency components through Fourier transform to obtain a real-time degradation rate; The real-time degradation rate and the initial life characteristic index are fitted and analyzed by a method based on partial least squares regression to obtain a degradation accumulation value, and the initial life characteristic index is calibrated using the degradation accumulation value to generate a remaining life characteristic index.

6. The method for predicting the remaining life of an LED lamp according to claim 1, characterized in that: The step of evaluating the real-time degradation rate according to the remaining life characteristic index to obtain a real-time degradation trend curve, and analyzing a plurality of degradation trend intervals using the real-time degradation trend curve comprises: The trend of the remaining life characteristic index is extracted by a support vector regression algorithm to obtain trend distribution characteristics, life change nodes and time interval factors, and the trend distribution characteristics are screened by using the life change nodes to obtain a key life node sequence; Dividing the key life node sequence according to a dynamic time warping algorithm to obtain a plurality of degradation trends, and dividing the range of each degradation trend by the time interval factor to obtain a plurality of degradation trend intervals; A trend fitting algorithm is used to match and analyze the several degradation trend intervals and the real-time degradation rate in the divided ranges to generate a real-time degradation trend curve.

7. The method for predicting the remaining life of an LED lamp according to claim 1, characterized in that: The step of performing weighted fusion calculation on all the degradation trend intervals by using the environmental data and the workload data to obtain a predicted value of the remaining life of the LED lamp, and generating a remaining life prediction report according to the predicted value of the remaining life of the LED lamp comprises: quantifying the environmental data by using a hierarchical analysis method to obtain environmental impact weights, and performing distribution analysis on the workload data by using a Kalman filter to generate workload impact weights; Performing weighted accumulation on all the degradation trend intervals according to the environmental impact weight and the workload impact weight to obtain a comprehensive degradation index; The parameters of the comprehensive degradation index are optimized through Bayesian prediction to obtain a predicted value of the remaining life of the LED lamp, the predicted value of the remaining life of the LED lamp is input into a preset life analysis model, and a remaining life prediction report is output.

8. A device for predicting the remaining life of an LED lamp, characterized in that: include: A collection module is used to collect the operating state parameters of the LED lamp in real time to obtain operating state data, environmental data and workload data, input the operating state data into a preset degradation model for analysis, and obtain the degradation characteristic parameters and initial life evaluation value of the LED lamp; An analysis module, used to perform multi-dimensional correlation analysis on the operating state parameter and the degradation characteristic parameter to obtain an initial life characteristic index, compare the initial life characteristic index with the initial life assessment value, and generate a dynamic degradation factor; A weighting module, used for performing weighted decomposition analysis on the dynamic degradation factor to obtain a real-time degradation rate and a degradation accumulation value, and using the degradation accumulation value to calibrate the initial life characteristic index to obtain a remaining life characteristic index; An evaluation module, used to evaluate the real-time degradation rate according to the remaining life characteristic index to obtain a real-time degradation trend curve, and analyze a plurality of degradation trend intervals using the real-time degradation trend curve; A generation module is used to perform weighted fusion calculation on all the degradation trend intervals using the environmental data and the workload data to obtain a remaining life prediction value of the LED lamp, and generate a remaining life prediction report based on the remaining life prediction value of the LED lamp.

9. An electronic device, characterized in that: The invention comprises a memory and a processor, wherein the memory stores a computer program executable on the processor, and the processor implements the method for predicting the remaining life of an LED lamp as claimed in any one of claims 1 to 7 when executing the computer program.

Images