A spectral matching method for a multi-light-source solar simulator

Abstract

The present application relates to a method for matching AM1.5 standard sunlight spectrum by using multiple light sources, and the simulated spectrum wavelength range is the full spectrum of visible-short wave infrared sunlight spectrum radiation. The present application uses metal halide lamp, halogen lamp and multiple different kinds of high-power LED light sources as the alternative sub-light sources of the solar simulator, and the selected light sources are actually tested to obtain the spectral data of each light source as the alternative light source library. According to the superposition principle of the spectrum, the genetic algorithm is used to seek the best configuration of the light source type and quantity under the condition of meeting the spectral matching degree requirement. The present application overcomes the problems of spectral mismatching deviation and wavelength range limitation in the traditional solar simulation technology, and can realize the full spectrum solar simulation in the range of 350-2500nm at low cost, which is suitable for the research of solar spectrum radiation in the optical remote sensing physical simulation system.

Description

Technical Field

[0001] The present invention relates to the technical field of solar simulators, and in particular to a technical method for matching sunlight with a visible-shortwave infrared multi-light source, which is suitable for research on full-band solar spectrum radiation simulation in optical remote sensing physical simulation systems. Background Art

[0002] A solar simulator is an important test device that simulates solar radiation characteristics indoors. It overcomes the fact that natural sunlight is easily affected by unknown and uncontrollable factors such as time, location, and weather conditions. It plays an increasingly important role in solar energy utilization, aerospace, new material development, healthcare, plant cultivation, and other aspects.

[0003] Traditional solar simulators mostly use xenon lamps as light sources, but they have disadvantages such as high heat generation, many auxiliary devices during operation, poor economy, short life and non-adjustable spectrum, making them unacceptable in the design of full-spectrum solar simulators.

[0004] The emergence of light-emitting diodes (LEDs) has further promoted the development of solar simulators. Compared with traditional light sources, LEDs offer advantages such as high luminous efficiency, compact size, long life, environmental friendliness, and flexible control. Using LEDs to simulate sunlight has become a research hotspot in recent years. However, LED solar simulators primarily simulate the AM1.5 standard solar spectrum, with the majority of the spectral band concentrated between 350 and 1100 nm. However, these solar simulators have certain limitations in experiments requiring high spectral quality and a wider spectral range, such as indoor hardware-in-the-loop simulations in remote sensing technology.

[0005] Due to the inherent limitations of the radiation characteristics of various artificial light sources, no single source currently accurately reproduces the true solar spectrum. These sources can only approximately simulate spectral curves within a specific wavelength range, significantly different from the standard solar spectrum curve. Traditional multi-light source solar simulation methods primarily rely on empirical trial-and-error or exhaustive methods to determine the light source combination, making it difficult to achieve the optimal configuration with low cost and high performance.

[0006] In order to overcome the shortcomings of existing spectral matching methods, the present invention provides a simulation method of the AM1.5 standard solar spectrum in the visible-shortwave infrared range based on multiple light sources. A metal halide lamp whose spectral energy distribution is relatively close to the solar spectrum is used as the base light source to ensure the illumination intensity of the entire simulator; the advantage of the halogen lamp in the relatively strong infrared radiation intensity is used to make up for the problem of insufficient infrared energy of the metal halide lamp; finally, different types of high-power LEDs are used to supplement and adjust the energy in some local bands of the visible-near infrared. This not only makes up for the difference between the spectral distribution of metal halide lamps and halogen lamps and the target solar spectrum distribution, but also improves the shortcoming of insufficient light intensity of solar simulators made of pure LEDs. The present invention can obtain the optimal configuration of the number of different types of light sources through the spectral matching method, effectively reducing costs while ensuring high-performance solar spectrum simulation, and has good practical value. Summary of the Invention

[0007] The purpose of the present invention is to address the problems of wavelength range limitation and insufficient light intensity in existing LED solar simulators, and to propose a solar simulation method based on high spectral matching of multiple light sources. The method can simulate the solar spectrum characteristics in the range of 350-2500nm, and serve the experimental needs of simulating sunlight illumination in the process of optical remote sensing physics simulation.

[0008] The technical solution adopted by the present invention to solve this technical problem is: providing a method for simulating the AM1.5 standard solar spectrum in the synthetic visible-shortwave infrared range. To simulate the solar spectrum, it is necessary to form a light source array with a small error between the spectral characteristics and the sunlight spectral distribution, good consistency in lighting performance such as radiance and radiation intensity directional distribution, and long life. Therefore, it is necessary to first select a suitable light source, measure the actual spectral energy distribution data of the light source using a spectrometer, and use this data as a light source library for subsequent spectrum fitting. Then, according to the relevant principles of spectral matching, an optimization algorithm is used to continuously superimpose and fit the actual spectrum of the light source, so that the synthetic spectrum continuously approaches the standard solar spectrum, and an evaluation function is set to evaluate the effect of spectral matching.

[0009] The light source combination array includes metal halide lamps, halogen lamps and a series of high-power LED lamps of different types. Because metal halide lamps have a spectral energy distribution that is relatively close to the solar spectrum, high light efficiency, low cost, simple operation and high safety, metal halide lamps are first considered as the base light source to ensure the basic spectral energy distribution and the illumination intensity of the entire simulator. The radiation intensity of the infrared spectrum of halogen lamps is higher than that of other light sources, so halogen lamps are selected to supplement the spectral energy of the short-wave infrared. LED light sources currently have mature products with multiple wavelengths. They are safe, stable and controllable, and LEDs have good monochromaticity. They can be used to supplement the energy in certain specific bands within the visible-near infrared part. The base light source can ensure the light intensity requirements of the solar spectrum simulator, while different combinations of LEDs and halogen lamps can ensure the required output spectral distribution. In addition, LEDs can also improve the spectral tunability of the solar simulator. For example, when the luminous spectrum of the base light source changes, the matching spectrum can be corrected by adjusting the LED combination.

[0010] The principle of spectral matching is as follows: spectral matching is the process of matching the target spectrum characteristics using light sources with different spectra given a known target spectrum. The AM1.5 standard energy distribution curve shows that the solar spectrum curve to be fitted is a continuous curve. According to the principle of spectral superposition, the light emitted by a light source is usually composed of various wavelengths, each wavelength has a different spectral radiation flux, and the total radiation flux of the light source is the sum of the radiation fluxes of each wavelength. Therefore, in theory, if the solar spectrum is regarded as a linear superposition of different types of continuous spectra, the simulation of solar spectral radiation can be achieved by combining multiple light sources with different spectra.

[0011] The optimization algorithm described is a genetic algorithm. Because different combinations of light sources, both in terms of type and quantity, inevitably produce different solutions and results, the key to this design approach is finding the combination that best matches the target spectrum. To achieve the optimal light source ratio that matches the target spectrum, spectral matching can be viewed as an optimization problem, solved using a genetic algorithm. This algorithm continuously performs genetic operations until the spectral match is optimal.

[0012] The evaluation functions of the spectrum matching effect are the Pearson correlation coefficient and the root mean square error. The closer the correlation coefficient is to 1 and the smaller the root mean square error is, the higher the matching degree between the fitted spectrum and the standard spectrum is.

[0013] Beneficial effects of the present invention:

[0014] (1) The design method described in the present invention fully utilizes the spectral radiation characteristics of various light sources by combining three types of light sources: metal halide lamps, halogen lamps, and LED lamps, and realizes the design of a solar simulator with high-quality spectrum in the visible-shortwave infrared range. Compared with traditional xenon lamp solar simulators or LED solar simulators, the design method eliminates the need for developing filters and complex beam shaping systems, and has the advantages of high safety, low cost, and a wide spectral range.

[0015] (2) The spectral matching method proposed in the present invention is suitable for solving the problem of simulating solar spectra with various types of light sources. By iteratively calculating the optimal configuration of different types of light sources, the solar spectrum radiation simulation under AM1.5 atmospheric quality conditions is achieved in the visible-shortwave infrared (350-2500nm) range, and the spectral correlation coefficient is above 0.91. The same method can be extended to the solar spectrum simulation under different atmospheric conditions and is universal.

[0016] The outstanding essential features and remarkable improvements of the present invention are further illustrated in the following specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The measured spectrum curves of the metal halide lamp and halogen lamp used in this invention are

[0018] Figure 2 The measured spectrum curves of 32 high-power LEDs used in this invention are

[0019] In the picture:

[0020]

[0021] Figure 3 This is a structural diagram of the multi-light source solar simulator produced under the guidance of the present invention.

[0022] In the figure: 33. Metal halide lamp 34. Halogen lamp 35. LED light array

[0023] Figure 4 This is a comparison chart of the spectrum of the solar simulator made by the present invention and the spectrum of AM1.5 sunlight DETAILED DESCRIPTION

[0024] The present invention will be further described below with reference to specific embodiments.

[0025] (1) In order to solve the problem of matching the standard solar spectrum of visible-shortwave infrared, the present invention firstly selects the 70W metal halide lamp with a 10° divergence angle of PHILIPS, which is a compact and efficient discharge reflector light source. It can output stable light color and transparent flashing light throughout its life, providing stable lighting with high irradiance. The measured spectral distribution curve is Figure 1 The 50W halogen lamp with a 10° divergence angle from OSRAM is preferred as the supplementary light source. The measured spectral distribution curve is Figure 1 The dashed line indicates that 32 high-power LED light sources are selected to construct a candidate light source library.

[0026] The LEDs initially packaged are Lambertian light sources with a beam divergence angle of 120°. Their light intensity follows a cosine function distribution, resulting in energy dispersion and low energy entering the subsequent optical system. To minimize the light source divergence angle and improve the beam collimation of the entire simulator, an optical lens with a 10° divergence angle was added. The measured spectral curves of the 32 LEDs are shown below. Figure 2 shown.

[0027] (2) To achieve high spectral matching, it is necessary to select the best light source combination from a large number of light source combinations of different types and quantities. This paper uses a genetic algorithm as a spectral matching algorithm. In the process of using the genetic algorithm for spectral fitting, the measured spectral data of three light sources are superimposed in different combinations to continuously approach and match the standard solar spectrum. The correlation coefficient and root mean square error are used as the fitness function, i.e., the termination condition, in the algorithm. The optimal light source combination type and number are obtained through continuous optimization and iteration. After calculation, the optimal light source combination obtained in this description is: 1 70W metal halide lamp, 4 50W halogen lamps, 23 types of LED light sources, totaling 128, namely 5 400nm LEDs, 4 420nm LEDs, 1 450nm LED, 1 480nm LED, 9 500nm LEDs, 1 530nm LED, 1 545nm LED, 1 630nm LED, 2 660nm LEDs, 5 680nm LEDs, 3 700nm LEDs, 11 730nm LEDs, 1 760nm LED, 8 780nm LEDs, 5 800nm ​​LEDs, 1 820nm LED, 6 850nm LEDs, 13 880nm LEDs, 2 925nm LEDs, 6 980nm LEDs, 8 blue LEDs, 29 white LEDs, and 5 green LEDs. After spectrum matching using genetic algorithm, the correlation coefficient between the synthetic spectrum and the standard spectrum is 0.932, and the root mean square error is 0.367. It can be seen that the matching effect is very good.

[0028] In order to verify the feasibility of the preferred embodiment, the present invention uses a physical production method to verify the effect of the lighting system of the preferred embodiment. The production process is as follows:

[0029] Use the selected light source combination to make the physical object. After determining the light source combination, it is necessary to arrange a series of light sources reasonably in space, such as Figure 3 As shown, one metal halide lamp is placed in the center of the entire installation, four halogen lamps are distributed in the four corners, and LEDs are evenly distributed around the metal halide lamp. The LEDs are controlled in series and powered by a 700A constant-current power driver. The metal halide lamp requires an electronic ballast to protect its proper operation, and a matching 12V electronic transformer is required for power. A high-power fan is also installed below the entire light source to minimize the impact of temperature increases on the LED light spectrum.

[0030] To verify that the solar simulator's irradiance spectrum met the design requirements, a spectroradiometer was used to measure the system's spectral irradiance. Calculations showed a correlation coefficient of 0.911 and a root mean square error of 0.416 between the experimental spectrum of the solar simulator and the standard solar spectrum.

[0031] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A spectrum matching method for a multi-light source solar simulator, characterized by: 1) First, different types of light sources are selected to construct a light source database. Metal halide lamps and halogen lamps are selected as the main simulated light sources in the visible-shortwave infrared band, and multiple different types of high-power LEDs are selected as the simulated light sources in the visible light band. The metal halide lamps, halogen lamps, and multiple different types of high-power LEDs together form a light source combination array. The light source array includes metal halide lamps, halogen lamps, and a series of different high-power LED lamps. The spectral energy distribution of metal halide lamps is close to that of the solar spectrum, and they offer high light efficiency, low cost, simple operation, and high safety. Metal halide lamps are used as the base light source to ensure the basic spectral energy distribution and illumination intensity of the entire simulator. Halogen lamps have high infrared radiation intensity and are used to supplement short-wave infrared spectrum energy. LEDs offer multiple wavelengths, are safe, stable, controllable, and have good monochromaticity, and are used to supplement energy within specific wavelengths in the visible and near-infrared regions. The base light source ensures the light intensity requirements of the solar spectrum simulator, and different combinations of LEDs and halogen lamps ensure the desired output spectral distribution. LEDs can also improve the solar simulator's spectral tunability. When the base light source's luminous spectrum changes, the matching spectrum can be corrected by adjusting the LED combination. Different types of high-power LEDs are used to supplement and adjust energy in certain localized wavelengths in the visible and near-infrared regions. 2) Conduct actual tests on the selected light sources to obtain the spectral irradiance distribution data of each light source and construct lighting models for different light sources; 3) Based on the principle of spectral superposition, an optimization algorithm is used to perform superposition fitting on the actual spectral irradiances of different light sources, so that the synthetic spectrum continuously approaches the reference sunlight spectrum, and the light source type and ratio with the optimal spectral matching degree are obtained; In order to obtain the optimal light source combination ratio that matches the target spectrum, spectrum matching is regarded as an optimization combination problem and solved with the help of genetic algorithm. Genetic operation is continuously performed using genetic algorithm until the spectrum matching degree reaches the optimal level. 4) Finally, an evaluation function is set to evaluate the actual effect of simulating the AM1.5 standard sunlight spectrum in the visible-shortwave infrared wavelength range.

2. The spectrum matching method of a multi-light source solar simulator according to claim 1, characterized in that: The selected light sources are metal halide lamps with a divergence angle of 10°, halogen lamps with a divergence angle of 10°, high-power white light, green light, blue light LED light sources and single-wavelength LED light sources.

3. The spectrum matching method of a multi-light source solar simulator according to claim 1, characterized in that: The selected optimization algorithm is a genetic algorithm, which continuously iterates and optimizes the simulated spectrum formed by the superposition of radiation from different types of light sources to minimize the difference with the target spectrum and achieve the predetermined light source spectral power distribution; In order to achieve the requirements of high spectral matching, it is necessary to select the best light source combination from a large number of light source combinations of different types and quantities, and use genetic algorithms as spectral matching algorithms; in the process of using genetic algorithms for spectral fitting, the measured spectral data of three light sources are superimposed in different combinations to make them continuously approach and match the standard solar spectrum. In the algorithm, the correlation coefficient and root mean square error are used as the fitness function, that is, the termination condition, and the optimal light source combination type and number are obtained through continuous optimization and iteration.

4. The spectrum matching method of a multi-light source solar simulator according to claim 1, characterized in that: The evaluation functions of the selected spectrum matching effect are the Pearson correlation coefficient and the root mean square error. The closer the correlation coefficient is to 1 and the smaller the root mean square error is, the higher the matching degree between the fitted spectrum and the standard spectrum is.

Images