CN-121072190-B - SDA spectrum design automation method and system

CN121072190BCN 121072190 BCN121072190 BCN 121072190BCN-121072190-B

Abstract

The invention provides an SDA spectrum design automation method which comprises the steps of setting a database, storing curve data and radiation flux parameters of standard lamps in the database, importing corresponding target curves, generating a plurality of formula groups according to the target curves, wherein each formula group comprises setting parameters of at least one standard lamp, each setting parameter comprises a lamp name, a current value and an occupied proportion, executing curve fitting calculation on each formula group to generate a fitting curve, respectively obtaining absolute amplitude curves of each standard lamp, carrying out area integration superposition on all the absolute amplitude curves involved in fitting to obtain the fitting curve, verifying the fitting curve through a multidimensional verification strategy, obtaining a scoring result, selecting the formula group with the highest score as an optimal formula group, judging whether the optimal formula group accords with expectations, outputting the optimal formula group if the optimal formula group accords with expectations, initializing the formula, and regenerating a new formula group if the optimal formula group does not accord with expectations.

Inventors

CHEN DONG
ZHENG YUN
YU LIANGJUN
LI XI

Assignees

杭州赛美蓝光电科技有限公司

Dates

Publication Date: 20260508
Application Date: 20251104

Claims (7)

An sda spectral design automation method comprising: a curve selection step, namely setting a database, wherein curve data and radiation flux parameters of standard lamps are stored in the database, a corresponding target curve is imported, a plurality of formula groups are generated according to the target curve, each formula group comprises setting parameters of at least one standard lamp, and the setting parameters comprise lamp names, current values and occupation proportions; a curve fitting step, namely executing curve fitting calculation on each formula group to generate a fitting curve, wherein the curve fitting calculation comprises the steps of respectively obtaining absolute amplitude curves of each group of standard lamps through radiation flux parameters and relative spectrum curves, and carrying out area integral superposition on all the absolute amplitude curves participating in fitting to obtain the fitting curve; A curve output step, verifying the fitting curve through a multi-dimensional verification strategy, obtaining a grading result, selecting a formula group with the highest grading as an optimal formula group, judging whether the optimal formula group accords with expectations, outputting the optimal formula group if the optimal formula group accords with expectations, initializing a formula if the optimal formula group does not accord with expectations, and regenerating a new formula group; The multi-dimensional verification strategy is used for acquiring basic information of the standard lamp, acquiring production cost parameters through the basic information of the standard lamp, analyzing performance indexes of the formula set through spectrum characteristics to serve as performance parameters, performing similarity calculation with a target curve to acquire matching parameters, and acquiring a grading result through dynamic weight calculation according to the production cost parameters, the performance parameters and the matching parameters.
2. The SDA spectral design automation method of claim 1, wherein the multi-dimensional verification strategy comprises a cost calculation sub-step for obtaining material cost and production parameters of the standard lamp, and calculating the production cost parameters of the recipe set according to the material cost and the production parameters of the standard lamp.
3. The SDA spectral design automation method of claim 1, wherein the multi-dimensional verification strategy further comprises a performance calculation sub-step, wherein the performance calculation sub-step is used for obtaining core performance parameters of the standard lamp, the core performance parameters comprise radiant flux and spectral distribution, and the performance parameters are generated through a weighting algorithm.
4. The SDA spectral design automation method according to claim 1, wherein the multi-dimensional verification strategy further comprises a similarity calculation sub-step, wherein the similarity calculation sub-step is used for fitting a curve, obtaining a key detection area and a non-key detection area according to a target curve, comparing the target curve with the key detection area of the fitted curve, generating a key similarity value, setting a similarity value threshold, obtaining the non-key similarity value of the non-key detection area when the key similarity value is greater than the similarity value threshold, and obtaining the matching parameters by a weight algorithm with the non-key similarity value and the key similarity value.
5. The SDA spectral design automation method of claim 1, wherein the curve selection step is provided with a recipe group generation step of generating a plurality of recipe groups by setting a plurality of different current values and occupation ratios for standard lamps and test data stored in a database by an operator.
6. The SDA spectral design automation method of claim 1, further comprising a database generation step, the database generation step comprising: A standard lamp testing sub-step, testing the standard lamp by a spectrometer, and naming according to the performance of the lamp beads and the test data; A database creation sub-step of searching whether the standard lamp with the current name exists in the database, if so, completing the current standard lamp database creation, and if not, performing independent database creation; and a database establishment sub-step, wherein an operator judges whether to complete database establishment according to the requirements, and if so, the operator exits the database generation step.
SDA spectral design automation system, characterized by comprising: the curve selection module is provided with a database, wherein the database stores curve data and radiation flux parameters of standard lamps, the curve data and the radiation flux parameters are imported into corresponding target curves, a plurality of formula groups are generated according to the target curves, each formula group comprises setting parameters of at least one standard lamp, and the setting parameters comprise lamp names, current values and occupation ratios; The curve fitting module is used for executing curve fitting calculation on each formula group to generate a fitting curve, wherein the curve fitting calculation comprises the steps of respectively obtaining absolute amplitude curves of each group of standard lamps through radiation flux parameters and relative spectrum curves, and carrying out area integral superposition on all the absolute amplitude curves participating in fitting to obtain the fitting curve; The curve output module is used for verifying the fitting curve through a multi-dimensional verification strategy, obtaining a grading result, selecting a formula group with the highest grading as an optimal formula group, judging whether the optimal formula group accords with the expectation, outputting the optimal formula group if the optimal formula group accords with the expectation, initializing a formula if the optimal formula group does not accord with the expectation, and regenerating a new formula group; The multi-dimensional verification strategy is used for acquiring basic information of the standard lamp, acquiring production cost parameters through the basic information of the standard lamp, analyzing performance indexes of the formula set through spectrum characteristics to serve as performance parameters, performing similarity calculation with a target curve to acquire matching parameters, and acquiring a grading result through dynamic weight calculation according to the production cost parameters, the performance parameters and the matching parameters.

Description

SDA spectrum design automation method and system Technical Field The invention relates to spectrum design, in particular to an SDA spectrum design automation method and system. Background In the field of spectrum design, in particular to an application scene of standard lamp combination to realize target spectrum, the traditional spectrum design method generally has the problems of low efficiency, insufficient precision and dependence on manual experience. At present, the operation of generating a target spectrum aiming at standard lamp combination in the industry mostly requires technicians to manually screen standard lamp types, adjust current parameters of standard lamps and occupation proportion of each standard lamp by virtue of self professional knowledge, and the whole process lacks systematic database support, so that not only is the curve data and radiation flux parameters of various standard lamps difficult to fully cover, but also the selectable standard lamp combination scheme is single due to the limitation of manual screening, and the design requirement of complex target spectrum cannot be met. In a curve fitting link, a simple linear superposition mode is generally adopted in the traditional method, and the scientificity of accurate calculation and area integral superposition of an absolute amplitude curve of a standard lamp is not fully considered, so that the deviation between a generated fitting curve and a target curve is large, and the requirement of high-precision spectrum design is difficult to reach. Meanwhile, in the verification stage of the fitting curve, a multi-dimensional verification strategy is lacking, the fitting effect is judged only through a single index, the suitability of the fitting curve in different wave bands and different radiation intensity intervals cannot be comprehensively evaluated, and the problems that the stability is poor and the spectral performance in practical application does not reach the standard possibly exist in the finally determined standard lamp formula group are caused. Disclosure of Invention In view of the shortcomings of the prior art, the present invention is directed to an automated SDA spectral design method and system for overcoming the aforementioned shortcomings of the prior art. In order to achieve the above purpose, the present invention provides the following technical solutions: an SDA spectral design automation method comprising: a curve selection step, namely setting a database, wherein curve data and radiation flux parameters of standard lamps are stored in the database, a corresponding target curve is imported, a plurality of formula groups are generated according to the target curve, each formula group comprises setting parameters of at least one standard lamp, and the setting parameters comprise lamp names, current values and occupation proportions; a curve fitting step, namely executing curve fitting calculation on each formula group to generate a fitting curve, wherein the curve fitting calculation comprises the steps of respectively obtaining absolute amplitude curves of each group of standard lamps through radiation flux parameters and relative spectrum curves, and carrying out area integral superposition on all the absolute amplitude curves participating in fitting to obtain the fitting curve; And a curve output step, wherein the fitted curve is verified through a multi-dimensional verification strategy, a scoring result is obtained, a formula group with the highest score is selected as an optimal formula group, whether the optimal formula group meets expectations or not is judged, if so, the optimal formula group is output, and if not, a formula is initialized, and a new formula group is regenerated. Preferably, the multi-dimensional verification strategy is used for acquiring basic information of the standard lamp, acquiring production cost parameters through the basic information of the standard lamp, analyzing performance indexes of the formula set through spectrum characteristics to serve as performance parameters, performing similarity calculation with a target curve to acquire matching parameters, and acquiring a grading result through dynamic weight calculation according to the production cost parameters, the performance parameters and the matching parameters. Preferably, the multi-dimensional verification strategy includes a cost calculation sub-step, wherein the cost calculation sub-step is used for obtaining the material cost and the production parameter of the standard lamp, and calculating and obtaining the production cost parameter of the formula set according to the material cost and the production parameter of the standard lamp. Preferably, the multi-dimensional verification strategy further comprises a performance calculation sub-step, wherein the performance calculation sub-step is used for obtaining core performance parameters of the standard lamp, the core performance parameters comprise radiant flux