Photoelectric detection system

Abstract

The invention discloses a photoelectric detection system, which comprises an optical path module, a transmission mechanism, a load configuration circuit, a signal acquisition circuit and a main control computer, and is characterized in that the main control computer is connected with the optical path module, the transmission mechanism, the load configuration circuit and the signal acquisition circuit; the light path module is used for outputting a stable excitation light beam used for exciting a to-be-detected photoelectric device, the transmission mechanism is used for achieving accurate movement of the light path module in a two-dimensional matrix area, and the load configuration circuit is used for executing loading path switching according to an instruction of the main control computer. The signal acquisition circuit is used for acquiring and converting output photoelectric response signals of the to-be-detected photoelectric device in different loading states, and the main control computer is used for performing performance analysis on the to-be-detected photoelectric device according to the photoelectric response signals. According to the invention, photoelectric device detection is expanded from single-point detection to comprehensive detection, and the detection range and flexibility are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of photoelectric conversion and measurement, in particular to a photoelectric detection system. BACKGROUND

[0002] Perovskite solar cells have become a research hotspot in the photovoltaic field due to their high theoretical efficiency (>26%) and low cost, but their industrialization faces two major challenges: ion migration-induced hysteresis effect, spatial non-uniformity defects, etc. These defects lead to a decrease in the internal carrier transport efficiency of the battery and an increase in the recombination rate, and are manifested as significant differences in the current-voltage curve in forward and reverse scans, which seriously restricts the accurate evaluation of efficiency and long-term stability. Current characterization methods are difficult to analyze the spatiotemporal evolution process of carrier dynamics simultaneously.

[0003] Therefore, it is urgent to develop a characterization technique with high spatiotemporal resolution to meet the comprehensive detection needs of defect distribution, ion migration path and carrier transport dynamics of large-area perovskite cells, and to provide accurate guidance for material design and process optimization. SUMMARY

[0004] The purpose of the present application is to provide a photoelectric detection system that extends the measurement range of photoelectric devices from a single point on the surface to the entire measurement surface, breaking through the one-sided limitations of single-point detection and ensuring that the detection results can reliably evaluate the overall performance of the device.

[0005] To achieve the above purpose, the present application provides the following solutions: A photoelectric detection system, comprising: an optical path module, a transmission mechanism, a load configuration circuit, a signal acquisition circuit and a host computer, wherein the host computer is connected with the optical path module, the transmission mechanism, the load configuration circuit and the signal acquisition circuit respectively; The optical path module is used to output stable excitation light beams for exciting the measured photoelectric device, the transmission mechanism is used to realize the precise movement of the optical path module in a two-dimensional matrix area, the load configuration circuit is used to perform load path switching according to the instructions of the host computer, the signal acquisition circuit is used to collect and convert the output photoelectric response signals of the measured photoelectric device under different load states, and the host computer is used to analyze the performance of the measured photoelectric device according to the photoelectric response signals.

[0006] Optionally, the transmission mechanism comprises an X-direction transmission assembly and a Y-direction transmission assembly arranged orthogonally to each other.

[0007] Optionally, the X-axis transmission assembly includes an X-axis lead screw and an X-axis slide for driving the optical path module to move along the X direction; the Y-axis transmission assembly includes a Y-axis lead screw and a Y-axis slide for driving the optical path module to move along the Y direction; the X-axis lead screw and the Y-axis lead screw are orthogonally arranged to each other; the Y-axis slide is fixed on the base, and the X-axis slide is installed above the Y-axis slide, the two forming a slide structure that realizes two-dimensional linear movement.

[0008] Optionally, the laser beam output by the optical path module forms a flat-top beam.

[0009] Optionally, the optical path module includes a light-emitting subunit, an optical fiber, and a beam shaper.

[0010] The light-emitting subunit includes a transient pulse light source, a steady-state light source system, and a Y-type fiber combiner; The Y-type fiber combiner combines transient pulsed laser and steady-state light into a single beam, which is then transmitted via fiber to the beam shaper. The beam shaper converts the input beam into a flat-top beam and orthogonally incident it onto the detection surface of the optoelectronic device under test.

[0011] Optionally, the load configuration circuit includes a high-resistance potentiometer, a low-resistance potentiometer, and a MOSFET array switch. The loaded state of the optoelectronic device under test includes a low-resistance on-state and / or a high-resistance off-state. When it is necessary to acquire transient photocurrent signals, the main controller controls the corresponding MOSFET to turn on, connecting the low-resistance resistor to the circuit of the optoelectronic device under test to establish a low-resistance on-state. When it is necessary to acquire transient photovoltage signals, the main controller controls another MOSFET to turn on, connecting the high-resistance resistor in series with the optoelectronic device under test to establish a high-resistance off-state.

[0012] Optionally, the photoelectric detection system further includes a carrying platform and a detection platform. The carrying platform is located below the optical path module, and the plane of the carrying platform is parallel to the plane of the detection platform. The detection platform is used to place the photoelectric device to be tested.

[0013] Optionally, the signal acquisition circuit includes a signal conditioning subunit and an analog-to-digital conversion subunit. The signal conditioning subunit is used to convert the original photoelectric voltage or photocurrent signal output by the optoelectronic device under test into a standard voltage signal, and the analog-to-digital conversion subunit is used to convert the conditioned analog voltage signal into a digital signal.

[0014] Optionally, the main control unit is based on a convolutional neural network. By training and fitting the input dataset, it automatically extracts key feature parameters that reflect changes in the photoelectric characteristics of the optoelectronic device under test, and establishes a mapping relationship between the features and the essential performance parameters of the optoelectronic device. It uses the trained weights to achieve rapid recursive calculation of newly acquired sample data, thereby achieving efficient performance prediction in transient photoelectric testing. The key feature parameters include: rise / fall time of the transient response curve, peak voltage / current, decay time constant, charge storage capacity, and capacitance relaxation characteristics.

[0015] Optionally, the convolutional neural network includes an input layer, a data preprocessing module, a feature extraction module, a parameter inversion module, and a training and deployment module; wherein, The input layer is used to receive waveform signal sequence data and spatial position information data of each point acquired by the data acquisition circuit at each spatial point in the matrix scan. The data preprocessing module performs noise filtering, baseline correction, and normalization on the acquired waveform signals. The feature extraction module is used to extract local temporal features of the waveform and model the dependency between different time points within a single site waveform to capture the influence of early decay on the tail signal. The parameter inversion module outputs local carrier lifetime, mobility, trap density, interface recombination parameters, and built-in electric field strength and its two-dimensional distribution map through a fully connected regression layer.

[0016] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: This photoelectric detection system overcomes the limitations of single-point detection, extending the test area of ​​photoelectric devices from a single point on the surface to the entire test surface. This improvement overcomes the limitations of single-point detection, ensuring that the test results can reliably evaluate the overall performance of the device. It not only broadens the coverage of photoelectric device performance testing but also effectively adapts to the photoelectric performance testing requirements of large-area photovoltaic cells such as crystalline silicon and perovskite. Simultaneously, based on the spatial detection of global photoelectric characteristic parameters, it significantly optimizes the testing reliability and data accuracy of heterogeneous structure devices. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of an embodiment of the photoelectric detection system provided by the present invention; Figure 2 This is a plan view showing the positions of the transmission mechanism and the loading platform. Reference numerals in the attached figures: 1. Steady-state light source system; 2. Transient pulse light source; 3. Y-type fiber optic combiner; 4. Fiber optic cable; 5. Beam shaper; 6. Y-axis transmission assembly; 7. X-axis transmission assembly; 8. Optoelectronic device under test; 9. Load configuration circuit; 10. Signal acquisition circuit; 11. Main controller; 12. Detection bracket; 13. Detection platform; 14. Loading platform. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] This invention proposes a photoelectric detection system. When testing a photoelectric device under test, a transmission mechanism performs a matrix scan of the device. By detecting parameters such as transient photovoltage changes, transient photocurrent changes, additional photovoltage, additional charge, and interface capacitance at high-precision two-dimensional matrix sites, the system can analyze the photoelectric properties of the device, such as current, conductivity, and carrier lifetime, and reflect its spatial distribution characteristics. Therefore, the test results of the device have high spatiotemporal resolution. By utilizing a neural network to rapidly recursively determine the intrinsic properties of the photoelectric device, it can achieve rapid and accurate measurement of the response characteristics of the device at different spatial locations. The transient signal acquisition capability supports nanosecond or even picosecond-level time resolution, making it suitable for parameter testing and process evaluation of devices such as photosensitive elements, photodetectors, and image sensors.

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] This invention proposes a photoelectric detection system, such as Figure 1 and 2 As shown, it includes an optical path module, a transmission mechanism, a signal acquisition circuit 10, a load configuration circuit 9, a main control unit 11, a detection bracket 12, and a loading platform 13. The detection bracket 12 is used to fix and support the optical path module and the optoelectronic device 8 under test.

[0023] The optical path module outputs high-precision excitation and bias light sources, providing signal excitation and a quasi-steady-state effect of thermal equilibrium. The laser beam output from the optical path module forms a flat-top beam, providing a uniform energy distribution. All optical components within the optical path module move collinearly and synchronously during operation, avoiding relative displacement and ensuring stable beam output and good consistency of the flat-top beam.

[0024] The transmission mechanism uses a ball screw to enable precise movement of the optical path module within a two-dimensional matrix area.

[0025] The load configuration circuit 9 is used to perform load path switching according to the instructions of the main controller 11.

[0026] The signal acquisition circuit 10 is used to acquire and convert the output photoelectric response signals of the photoelectric device under test 8 under different loading states.

[0027] When scanning the optoelectronic device 8 under test, the main controller 11 can adjust the optical path module to perform matrix-based position movement, so as to use the detection beam generated by the optical path module to perform matrix-based surface scanning of the optoelectronic device 8 under test. During the matrix-based surface scanning process, the main controller 11 collects and records the coordinate parameters of each moving position and the corresponding photoelectric signal. Then, based on the position parameters of the incident position of the detection beam on the detection plane of the optoelectronic device 8 under test, and the photoelectric signal corresponding to the position, the main controller 11 simultaneously analyzes the transient carrier dynamics behavior and spatial distribution characteristics of the optoelectronic device.

[0028] In some embodiments, the optical path module includes a light-emitting subunit, a beam transmission subunit, and a beam output subunit.

[0029] The light-emitting subunit includes a transient pulse light source 2, a steady-state light source system 1, and a beam combining unit. The beam combining unit is preferably a Y-type fiber beam combiner 3, which is configured to combine the laser beam from the transient pulse light source 2 with the steady-state beam from the steady-state light source system 1 to form a unified output path, which is then transmitted to the beam shaper 5 through the fiber 4.

[0030] Specifically, the transient pulsed light source 2 is a pulsed laser with a wavelength of 532 nm, a pulse width ≥ 25 ns, a frequency range of 0.1 Hz-1 GHz, and a maximum single-pulse output energy of 377 mJ; the steady-state light source system 1 uses a high-brightness LED with an irradiance range of 0.01 J / s / cm²-0.07 J / s / cm² and a power density of 0.1 J / s / cm²-0.7 J / s / cm².

[0031] The beam transmission subunit includes at least one section of optical fiber 4, used to output the combined beam from the light source module to the beam shaping unit. The beam shaping unit shapes the received combined beam into a flat-top beam with a uniform intensity distribution and outputs it to the detection surface of the solar cell device under test. Preferably, the beam shaper 5 causes the output beam to be incident on the detection surface along the normal direction (i.e., the orthogonal direction) to improve illumination uniformity and excitation stability.

[0032] The beam output subunit includes a beam shaper 5. The Y-type fiber combiner 3 combines the transient pulsed laser and the steady-state light, and transmits the combined beam to the beam shaper 5 via fiber 4. The beam shaper 5 converts the input beam into a high-quality flat-top beam and then incident it onto the detection surface in the normal direction (orthogonal incident) to improve the uniformity of illumination and the stability of excitation.

[0033] In some embodiments, the transmission mechanism may include an X-axis transmission assembly 7 and a Y-axis transmission assembly 6 arranged orthogonally to each other. The X-axis transmission assembly 7 includes an X-axis lead screw, an X-axis motor, an X-axis nut, and an X-axis slide, used to drive the optical path module to move along the X direction. The Y-axis transmission assembly 6 includes a Y-axis lead screw, a Y-axis motor, a Y-axis nut, and a Y-axis slide, used to drive the optical path module to move along the Y direction. The X-axis and Y-axis lead screws are orthogonally arranged to form a slide structure capable of two-dimensional linear movement within the detection plane. Specifically, the optical path module is fixed to a loading platform 14 on top of the X-axis slide. The main controller 11 controls the X and Y axis motors to drive the lead screws to rotate, realizing two-dimensional matrix scanning movement of the optical output module within the detection plane.

[0034] In some embodiments, such as Figure 1 As shown, a detection platform 13 is configured inside the detection bracket 12. The photoelectric device under test 8, the main controller 11, and the load configuration circuit 9 are all mounted on the detection platform 13. The loading platform 14 is located below the optical path module, and the plane of the loading platform 14 is parallel to the plane of the detection platform 13 (i.e., the detection plane).

[0035] Specifically, the transmission mechanism may include a horizontal lead screw, a vertical lead screw, a corresponding nut, a slide, and a servo drive device. The horizontal and vertical lead screws are set inside the detection bracket, perpendicular to each other, forming a matrix scanning platform. The optical path module is mounted on the slide, and the nut is fixed to the lead screw. The X-axis servo motor controls the horizontal lead screw to drive the slide to move in the X direction, and the Y-axis servo motor controls the vertical lead screw to drive the slide to move in the Y direction, thereby realizing two-dimensional matrix surface scanning. This design supports a minimum step accuracy better than 0.1 mm, a scanning area range of up to 210 mm × 210 mm, and is flexibly adjustable, and is also suitable for the detection of crystalline silicon solar cells.

[0036] In some embodiments, a load configuration circuit 9 is further included, which is used to switch the load resistor connected in series with the optoelectronic device under test 8 in different test modes according to the control instructions of the main controller 11, thereby realizing automatic configuration of different load states. The load states of the optoelectronic device under test 8 include a short-circuit state (low-resistance on-state) and / or an open-circuit state (high-resistance off-state). In this document, "short-circuit state" refers to the low-resistance state with an on-resistance ≤ 1Ω, and "open-circuit state" refers to the high-resistance state with a leakage current ≤ 1nA, both of which are ideal states that can be achieved in engineering.

[0037] The photoelectric signals include the transient photovoltage and transient photocurrent of the photoelectric device 8 under test. When detecting the transient photocurrent of the photoelectric device 8 under test, the data acquisition circuit sets the loaded state of the photoelectric device 8 under test to the short-circuit state. When detecting the transient photovoltage of the photoelectric device 8 under test, the data acquisition circuit configures the loaded state of the photoelectric device 8 under test to the open-circuit state.

[0038] The load configuration circuit 9 may include a set of digitally programmable resistor arrays and MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor) array switches. The digitally programmable resistor array includes: a high-resistance potentiometer with a resistance range of 100MΩ to 500MΩ for simulating an open-circuit state; and a low-resistance potentiometer with a resistance range of 10Ω to 50Ω for simulating a short-circuit state. The MOSFET array switches include multiple N-channel or P-channel metal-oxide-semiconductor field-effect transistors (MOSFETs), whose sources are connected to the different resistance channels mentioned above, whose drains are connected to the optoelectronic device under test 8, and whose gates are controlled by the main controller 11.

[0039] When it is necessary to collect transient photocurrent signals, the main controller 11 controls the corresponding MOSFET to turn on, and connects the low-resistance resistor to the circuit of the optoelectronic device under test 8 to establish a short-circuit state; when it is necessary to collect transient photovoltage signals, it controls another MOSFET to turn on, so that the high-resistance resistor is connected in series with the device under test to establish an open-circuit state.

[0040] In some embodiments, the signal acquisition circuit 10 is used to acquire and convert the output photoelectric signals of the photoelectric device 8 under different loading states with high precision and high time resolution.

[0041] Specifically, the signal acquisition circuit 10 includes a signal conditioning subunit and an analog-to-digital conversion subunit. The signal conditioning subunit is used to convert the raw photoelectric voltage or photocurrent signal output by the photoelectric device under test 8 into a standard voltage signal. It includes a sampling resistor, a low-noise buffer operational amplifier, and an optional filtering module. The sampling resistor is used to convert the photocurrent output by the photoelectric device into a voltage signal in current detection mode. The resistance value of the sampling resistor can be selected according to the test range. The low-noise buffer operational amplifier is used to isolate the signal source, buffer and amplify the voltage signal, and improve signal stability and anti-interference capability. A high input impedance and low-noise operational amplifier chip is preferred. The optional filtering module can connect a low-pass filter in series at the output of the buffer circuit to suppress high-frequency noise and improve sampling accuracy.

[0042] The analog-to-digital conversion subunit is used to convert the conditioned analog voltage signal into a digital signal, and it may include a high-precision analog-to-digital converter.

[0043] The signal acquisition circuit 10 can use a high-precision oscilloscope (such as a Tektronix series) to acquire the device response curves (such as voltage and current) under the action of laser pulses in real time, and automatically save the data for post-processing analysis by the main control computer 16. The acquired content includes the coordinate values ​​of each scanning point and the corresponding transient response data, which facilitates subsequent two-dimensional spatial visualization plotting.

[0044] In some embodiments, the main controller 11 connects to the optical path module, the transmission mechanism, and the load configuration circuit 9, and controls the oscilloscope to acquire transient signals via synchronous triggering. Specifically, the main controller 11 is used to control and schedule the optical path module, the load configuration circuit 9, and the signal acquisition circuit 10, including: controlling the optical path module to perform time control and site scanning of the excitation light; configuring the load configuration circuit 9 to switch the loaded state of the optoelectronic device 8 under test; setting the sampling parameters of the signal acquisition circuit 11 and triggering signal acquisition after completing the state switching; and storing, analyzing, or uploading the sampled data.

[0045] The main controller 11 uses a convolutional recognition scheme to perform calculations and recognition on the sampled input information, and recursively identifies the most matching (matching degree > 90%) internal attributes of the device, thereby identifying the essential problems caused by the device's working state.

[0046] Specifically, the main control unit 11, based on a convolutional neural network architecture, rapidly processes and analyzes multi-point spatially distributed data collected by the global photoelectric detection module. Through training and fitting the input dataset, the network automatically extracts key feature parameters reflecting changes in the device's photoelectric properties and establishes a mapping relationship between these features and the device's intrinsic performance parameters. During operation, the convolutional neural network utilizes pre-trained weights to perform rapid recursive calculations on newly acquired sample data, thereby achieving efficient performance prediction in transient photoelectric testing.

[0047] Key characteristic parameters refer to physically meaningful quantitative indicators extracted from transient photovoltage, transient photocurrent, additional photovoltage, interface capacitance, and charge response curves, including but not limited to: rise / fall time, peak voltage / current, decay time constant, charge storage capacity, and capacitance relaxation characteristics of transient response curves. These parameters can characterize intrinsic properties of optoelectronic devices such as photogenerated carrier dynamics, interface charge transfer efficiency, and defect state density distribution at various spatial locations.

[0048] The mapping relationship refers to the nonlinear functional relationship from the input feature space to the target performance parameter space constructed by a convolutional neural network. Specifically, the network extracts spatial and temporal features from the input signal through multiple layers of convolution and pooling operations. These extracted high-dimensional features are then mapped to target performance parameters, such as carrier lifetime (τ), interface recombination rate, photoelectric conversion efficiency, and dark current density, via fully connected layers. During the training phase, this mapping relationship is learned using backpropagation and gradient descent optimization methods, aiming to minimize the error between the predicted values ​​and the true labels, thereby establishing a reliable predictive model from multi-channel transient response data to the intrinsic performance of the device.

[0049] Among them, performance prediction refers to using a trained convolutional neural network to quantitatively estimate the photoelectric performance parameters of an unknown sample, such as predicting the carrier diffusion length, quantum efficiency, responsivity, etc. at a specific location.

[0050] The neural network of this invention is used to invert local physical parameters of a single site based on transient photoelectric signals. It includes an input layer, a data preprocessing module, a feature extraction module, a parameter inversion module, and a training and deployment module. The input layer receives sequential data such as transient photovoltage attenuation curves, transient photocurrent attenuation curves, and preprocessed charge response curves acquired by the data acquisition unit at each spatial site in a matrix scan, as well as spatial location information data for each site. The preprocessing module performs noise filtering, baseline correction, and normalization on the acquired waveform signals to eliminate measurement noise and baseline drift, thereby improving the signal quality of the waveform at a single site. The feature extraction module includes a one-dimensional convolutional layer and a Transformer module. The one-dimensional convolutional layer extracts local temporal features of the waveform, such as rapid attenuation segments and peak positions. The Transformer module models the dependencies between different time points within the waveform at a single site, capturing the influence of early attenuation on the tail signal. Each spatial site waveform is processed independently, and the attention mechanism does not model across sites. The parameter inversion module outputs local carrier lifetime, mobility, trap density, interface recombination parameters, and the built-in electric field intensity and its two-dimensional distribution map through a fully connected regression layer. During training, the neural network employs physical constraint loss, including exponential decay constraint, charge continuity constraint, and drift-diffusion consistency constraint, to ensure that the predicted physical parameters conform to fundamental physical laws.

[0051] When testing the optoelectronic device 8, the main controller 11 controls the transmission mechanism to drive the optical path module to move relative to the optoelectronic device 8, thereby using the detection beam in the optical path module to scan the surface of the optoelectronic device 8. The main controller 11 records the position information of the detection beam incident on the test surface of the optoelectronic device 8 in real time, and acquires the transient photoelectric signal (voltage / current) corresponding to the position through the signal acquisition circuit 10, and transmits it to the main controller 11 for processing and analysis.

[0052] During the testing process, the optoelectronic device 8 under test is placed at the center of the platform, with its testing surface facing upwards and aligned with the incident direction of the light beam. The main control unit 11 sets the matrix scanning path and spacing parameters, and drives the slide to move point by point along the X / Y direction to complete the scanning of the entire testing surface. At each scanning point, the laser triggers the device to respond, and the oscilloscope records the response signal. The system automatically correlates the response signal with the spatial position information to form a dataset.

[0053] By normalizing, fitting, and statistically analyzing photoelectric response data from multiple spatial points, the distribution of key parameters such as carrier lifetime, charge extraction rate, and conductivity on the device surface can be obtained. This accurately reveals spatial non-uniformity of the device, such as grain boundary effects, local defects, and ion migration paths, effectively supporting the structural optimization and process debugging of perovskite solar cells and other heterogeneous devices.

[0054] For devices under test with relatively uniform photoelectric performance in space, their transient photoelectric response can be characterized by a single-point scan. However, for devices with spatial inhomogeneities, it is necessary to combine the beam positions and corresponding photoelectric signals recorded by multiple single-point scans to perform spatial photoelectric performance analysis, including carrier migration paths, carrier distribution, ion migration processes, and defect distribution. During each single-point scan, the main controller records the current incident position of the beam and acquires the corresponding signal at that position through the photoelectric signal acquisition circuit. By analyzing the signals, parameters such as charge extraction quantity, extraction rate, and carrier recombination lifetime are obtained, thereby achieving a precise evaluation of the spatial transient performance of the optoelectronic device.

[0055] Therefore, this invention expands the detection of optoelectronic devices from a single point to a comprehensive approach, improving the detection range and flexibility, and meeting the measurement needs of large-area photovoltaic devices such as perovskite and dye-sensitized photovoltaics. It can also characterize carrier dynamics and spatial inhomogeneities. Unlike traditional PL / EL area scanning techniques, this system controls the detection beam to scan along a preset path, rather than exciting the entire system, achieving spatiotemporal resolution of carrier dynamics. This enables quantitative analysis of ion migration and regional defect effects in perovskite solar cells, and has wide-ranging applications.

[0056] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0057] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A photoelectric detection system, characterized in that, include: The system includes an optical path module, a transmission mechanism, a load configuration circuit, a signal acquisition circuit, and a main control unit, wherein the main control unit is connected to the optical path module, the transmission mechanism, the load configuration circuit, and the signal acquisition circuit, respectively. The optical path module is used to output a stable excitation beam for exciting the optoelectronic device under test. The transmission mechanism is used to move the optical path module within the two-dimensional matrix area. The load configuration circuit is used to execute load path switching according to the instructions of the main controller. The signal acquisition circuit is used to acquire and convert the output photoelectric response signals of the optoelectronic device under test under different loading states. The main controller is used to perform performance analysis on the optoelectronic device under test based on the photoelectric response signals.

2. The photoelectric detection system as described in claim 1, characterized in that, The transmission mechanism includes an X-axis transmission component and a Y-axis transmission component that are orthogonally arranged to each other.

3. The photoelectric detection system as described in claim 2, characterized in that, The X-axis transmission assembly includes an X-axis lead screw and an X-axis slide, used to drive the optical path module to move along the X direction; the Y-axis transmission assembly includes a Y-axis lead screw and a Y-axis slide, used to drive the optical path module to move along the Y direction; the X-axis lead screw and the Y-axis lead screw are orthogonally arranged to each other, and the two constitute a slide structure to realize two-dimensional linear movement.

4. The photoelectric detection system as described in claim 1, characterized in that, The laser beam output by the optical path module forms a flat-top beam.

5. The photoelectric detection system as described in claim 1, characterized in that, The optical path module includes a light-emitting subunit, an optical fiber, and a beam shaper; The light-emitting subunit includes a transient pulse light source, a steady-state light source system, and a Y-type fiber combiner; The Y-type fiber combiner combines transient pulsed laser and steady-state light into a single beam, which is then transmitted via the fiber to the beam shaper. The beam shaper converts the input beam into a flat-top beam and orthogonally incident it onto the detection surface of the optoelectronic device under test.

6. The photoelectric detection system as described in claim 1, characterized in that, The load configuration circuit includes a high-resistance potentiometer, a low-resistance potentiometer, and a MOSFET array switch. The load state of the optoelectronic device under test includes a low-resistance on state and / or a high-resistance off state. When it is necessary to collect transient photocurrent signals, the main controller controls the corresponding MOSFET to turn on and connects the low-resistance resistor to the circuit of the optoelectronic device under test to establish a low-resistance on state. When it is necessary to collect transient photovoltage signals, the main controller controls another MOSFET to turn on, so that the high-resistance resistor is connected in series with the photoelectric device under test to establish a high-resistance off state.

7. The photoelectric detection system as described in claim 1, characterized in that, The photoelectric detection system also includes a carrier platform and a detection platform. The carrier platform is located below the optical path module, and the plane of the carrier platform is parallel to the plane of the detection platform. The detection platform is used to place the photoelectric device to be tested.

8. The photoelectric detection system as described in claim 1, characterized in that, The signal acquisition circuit includes a signal conditioning subunit and an analog-to-digital conversion subunit. The signal conditioning subunit is used to convert the original photoelectric voltage or photocurrent signal output by the photoelectric device under test into a standard voltage signal. The analog-to-digital conversion subunit is used to convert the conditioned analog voltage signal into a digital signal.

9. The photoelectric detection system as described in claim 1, characterized in that, The main control computer is based on a convolutional neural network. By training and fitting the input dataset, it automatically extracts key feature parameters that reflect the changes in the photoelectric characteristics of the optoelectronic device under test, and establishes a mapping relationship between the features and the essential performance parameters of the optoelectronic device. It uses the pre-trained weights to realize rapid recursive calculation of newly acquired sample data, thereby achieving performance prediction in transient photoelectric testing. The key feature parameters include at least: the rise / fall time of the transient response curve, peak voltage / current, decay time constant, charge storage capacity, and capacitance relaxation characteristics.

10. The photoelectric detection system as described in claim 9, characterized in that, The convolutional neural network includes an input layer, a data preprocessing module, a feature extraction module, a parameter inversion module, and a training and deployment module; wherein, The input layer is used to receive waveform signal sequence data and spatial position information data of each point acquired by the data acquisition circuit at each spatial point in the matrix scan. The data preprocessing module performs noise filtering, baseline correction, and normalization on the acquired waveform signals. The feature extraction module is used to extract local temporal features of the waveform and model the dependency between different time points within a single site waveform to capture the influence of early decay on the tail signal. The parameter inversion module outputs local carrier lifetime, mobility, trap density, interface recombination parameters, and built-in electric field strength and its two-dimensional distribution map through a fully connected regression layer.

Images