Comparison and validation of neural network models to estimate LED spectral power distribution

The spectral power distribution (SPD) is the true fingerprint of a light source and is mainly dependent on electrical and thermal loading. Both the photometric and colorimetric quantities are originally extracted from SPD. Therefore, the dynamic prediction of SPD for LED has become an important aspect to evaluate the performance of LED during its time of operation. Generally, the statistical models are used to predict SPD. However, the statistical model with more than two input makes the system complex and time demanding. Artificial Neural Network (ANN) models, on the other hand, can help with this problem. The major goal of this research is to improve the utility of ANN in lighting applications. This is demonstrated by various neural network (NN) structures referred as models 1, 2 and 3 with combinations of varied neurons and hidden layers (HLs) to forecast SPD for various electrical and thermal stress levels at zero hours. The results are compared and based on absolute prediction error (APE) set to 5%, model 1 is considered as the best model for the SPD prediction. In addition, the time-based SPD prediction with model 1 is investigated using temperature, wavelength and time as input parameters for the LED luminaire and is validated.

Automated summary

The spectral power distribution (SPD) is influenced by electrical and thermal loading and plays a significant role in evaluating the performance of LED lighting. Statistical models are commonly used for SPD prediction, but become complex and time-consuming with more than two inputs. Artificial Neural Network (ANN) models offer a solution to this problem. This study aims to enhance the practicality of ANN in lighting applications, utilizing different neural network structures (models 1, 2, and 3) to forecast SPD under various electrical and thermal stress levels at zero hours. Model 1 is identified as the best SPD prediction model based on the absolute prediction error (APE) set at 5%, and its time-based SPD prediction for LED luminaires is validated using temperature, wavelength, and time as input parameters.