Research On Performance Monitoring And Evaluation Methods Of High-Power LED Array Light Source

As a new type of artificial light source,high-power LED occupies an important position in the lighting field due to its high brightness,long life,good controllability,energy saving and environmental protection.The photoelectric thermal theory describes the mutual coupling and mutual influence of the photoelectric thermal parameters of a single LED.It shows that the increase in junction temperature will reduce the optical performance of the LED's luminous flux and luminous efficiency.The high-power LED array light source is an extremely complex system composed of multiple different types of LED monomers.It has the characteristics of high nonlinearity,temperature sensitivity,obvious aging characteristics and inconsistencies.These characteristics make the large The measurement and control of nonlinear parameters of power LED array light sources are facing huge challenges,and the optical performance of high-power LED array light sources will degrade with long-term work.Therefore,it can timely and accurately monitor the interior of high-power LED array light sources.The change of photoelectric thermal state parameters,and the diagnosis and positioning of the resulting faults,have important theoretical and application values for ensuring and improving the safety,reliability,and lasting efficiency of high-power LED array light sources.In the photovoltaic field,high-power LED arrays can meet the high-power and high-energy application requirements of solar simulators and the performance requirements of wide dynamic adjustment.Multi-LED solar simulators are used as quality test equipment in the solar cell production process,with real-time performance parameters Monitoring and fault diagnosis and location research are of great significance to safe production,improving production efficiency,and reducing production costs.Therefore,this paper proposes new parameter measurement methods and fault detection methods for the performance monitoring and fault diagnosis of high-power LED array light sources.Aiming at the problem of optical performance degradation of high-power LED array light sources,this paper establishes a photoelectric thermal parameter monitoring system based on the PET theory.At the process level,a state observer is designed to monitor the signal fluctuations of each channel of the external photoelectric sensor,and the thermocouple monitors the ambient temperature and heat sink.With reference to changes in temperature,measuring resistance monitors changes in drive current.And through the data analysis method,the changes in the statistical characteristics of the residual sequence of the observation data of each channel and the ideal state data are analyzed to perform fault detection and location of the light source for evaluation.This article mainly conducts research through the following aspects:(1)Aiming at the problem of dynamic behavior identification and modeling and nonlinear parameter measurement of high-power LED arrays in the operating environment of variable temperature and complex current conditions,first,based on the multi-LED equivalent circuit model,a mathematical model for the description of LED dynamic characteristics is established.The parameters in the mathematical model set sensor monitoring points for the high-power LED array light source and design a data collection scheme for the problem that the light output data of the LED array light source is difficult to measure.The experiment verifies whether the measured data of the data acquisition scheme accurately reflects the mathematical model.An experimental measurement device was built according to the sensor configuration plan,and the TDMA data acquisition method was used to measure the photoelectric thermal parameter data set of a single LED in the array.The experimental results proved that the sensor configuration plan can be used to establish an effective monitoring model.(2)Aiming at the influence of the sensor's measurement noise,ambient light,ambient temperature and other unknown and uncertain interferences during the working process of high-power LEDs,a strong tracking Kalman filter observer is designed.This paper uses the sensor data of each monitoring point of the monitoring model to establish a Kalman filter residual generator:a state observer is designed based on the strong tracking Kalman filter to observe the parameter changes of each observation in each time channel,and form a residual sequence from the sensor measurement data,To provide a data basis for subsequent state monitoring and fault diagnosis,and to evaluate the system by analyzing the size and pattern of the residual error.(3)Aiming at the strong uncertainty of photoelectric thermal parameters and time-varying time-delay nonlinear characteristics in the working process of high-power LED array dynamic light source,this paper adopts dynamic kernel principal component analysis(DKPCA)to conduct online fault diagnosis of high-power LED array light source.This method can effectively capture the non-linear and correlation characteristics of residual sequence data,and realize fault detection based on the principal element characteristics of historical data and the SPE statistic threshold and the statistical characteristics of online data.It is realized by reconstructing the contribution graph method.The separation of the fault.Simulation experiments show that this method can effectively monitor and diagnose the typical sensor and actuator faults of high-power LED array dynamic light sources.Compared with the nuclear principal component analysis method,it is more sensitive to faults and can effectively improve the fault detection rate.