Detection method and system of photoelectric detector based on different working conditions

Abstract

The invention discloses a detection method and system of a photoelectric detector based on different working conditions, and relates to the technical field of photoelectric detectors, and the method comprises the steps: carrying out the simulation of different working conditions of a primary twinborn body, so as to output a multistage working condition twinborn body, and triggering the photoelectric detector to carry out the online detection of a detection object, according to the method, multiple pieces of spectral data are input into the multi-stage working condition twin, and the corresponding detection event is output, so that conversion between the primary twin and the multi-stage working condition twin is realized, and the accuracy of the detection event is improved. And triggering mode optimization of the photoelectric detector according to the detection influence content and the current working condition of the detection area so as to output a final detection mode of the photoelectric detector, and determining a current detection result according to the plurality of detection data, the corresponding detection object and the corresponding environment parameter combination. And the dynamic change of the current detection result is triggered along the updating of the working condition change content, so that the efficient detection of the photoelectric detector based on different working conditions is realized.

Description

Technical Field

[0001] This invention relates to the technical field of photodetectors, and more particularly to a detection method and system for photodetectors based on different operating conditions. Background Technology

[0002] As a core device that converts light signals into electrical signals, photodetectors play a vital role in fields such as industrial automation, environmental monitoring, and security surveillance. With the development of materials science, heterojunction materials, represented by two-dimensional transition metal chalcogenides, are gradually being applied to the development of new high-performance photodetectors due to their wide spectral response and high sensitivity.

[0003] In practical applications, the detection environment is often complex and variable. Existing technologies mostly adopt a linear logic of "real-time acquisition-direct processing," which fails to construct a primary twin of the detection area and cannot simulate different operating conditions in advance based on combinations of environmental parameters. This causes photodetectors to be in a passive receiving state when facing complex environments, affecting the accuracy of detection events and preventing photodetectors from conducting efficient detection based on different operating conditions. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides a detection method and system for photoelectric detectors based on different working conditions.

[0005] This invention provides a detection method for a photodetector based on different operating conditions, including: The detection area of ​​the photodetector is marked, and a primary twin is constructed based on the detection object and the corresponding environmental parameters in the detection area. The primary twin is then simulated under different operating conditions to output a multi-level operating condition twin. When the photodetector performs online detection on the target object, it collects multiple spectral data corresponding to the target object, inputs the multiple spectral data into the multi-level operating condition twin, and outputs the corresponding detection event; Based on the identification of the detection event, the corresponding detection matching content is determined, and the current detection mode and corresponding detection impact content of the photodetector are determined by tracing the detection matching content. Based on the detection impact content and the current operating conditions of the detection area, the mode optimization of the photodetector is triggered to output the final detection mode of the photodetector. In this final detection mode, multiple detection data are determined based on the photoelectric detector. The current detection result is determined based on the combination of multiple detection data, the corresponding detection object, and the corresponding environmental parameters. The dynamic change of the current detection result is triggered as the working condition changes.

[0006] This invention provides a photodetector detection system based on different operating conditions, which is applied to the aforementioned photodetector detection method based on different operating conditions; the photodetector detection system based on different operating conditions includes: The twin module is used to mark the detection area of ​​the photodetector. A primary twin is constructed based on the detection object and the corresponding environmental parameters in the detection area. The primary twin is simulated under different working conditions to output a multi-level working condition twin. The detection event module is used to collect multiple spectral data corresponding to the detection object when the photodetector performs online detection on the detection object, input the multiple spectral data into the multi-level operating condition twin, and output the corresponding detection event; The detection mode optimization module is used to determine the corresponding detection matching content based on the identification of the detection event, and to determine the current detection mode and corresponding detection impact content of the photodetector by tracing the detection matching content. Based on the detection impact content and the current operating conditions of the detection area, the mode optimization of the photodetector is triggered to output the final detection mode of the photodetector. The operating condition change module is used to determine multiple detection data based on the photodetector in the final detection mode, determine the current detection result based on the combination of multiple detection data, the corresponding detection object and the corresponding environmental parameters, and trigger the dynamic change of the current detection result along with the update of the operating condition change content.

[0007] Compared with the prior art, the beneficial effects of the present invention are: (1) A primary twin is constructed based on the detection object and the corresponding environmental parameter combination in the detection area. The primary twin is simulated under different working conditions to output a multi-level working condition twin. The photoelectric detector is triggered to detect the detection object online. Multiple spectral data are input into the multi-level working condition twin and the corresponding detection events are output. The transformation between the primary twin and the multi-level working condition twin is realized, and the multi-level working condition twin is further controlled, which improves the accuracy of the detection events.

[0008] (2) The current detection mode and corresponding detection impact content of the photodetector are determined by tracing the detection matching content. The photodetector mode optimization is triggered according to the detection impact content and the current working condition of the detection area to output the final detection mode of the photodetector. The current detection result is determined according to multiple detection data, the corresponding detection object and the corresponding environmental parameter combination. The dynamic change of the current detection result is triggered by updating the working condition change content. The final detection mode is introduced and the multiple detection data, the corresponding detection object and the corresponding environmental parameter combination are fully considered, which improves the accuracy of the current detection result and dynamically controls the working condition change content, realizing the efficient detection of the photodetector based on different working conditions. Attached Figure Description

[0009] Figure 1 This is a flowchart illustrating the detection method of the photodetector based on different working conditions in the embodiments of the present invention; Figure 2 This is a flowchart illustrating step S11 in the photodetector detection method based on different working conditions in this embodiment of the invention. Figure 3 This is a flowchart illustrating step S12 in the photodetector detection method based on different working conditions in this embodiment of the invention. Figure 4 This is a flowchart illustrating step S13 in the photodetector detection method based on different working conditions in this embodiment of the invention. Figure 5 This is a flowchart illustrating step S14 in the photodetector detection method based on different working conditions in this embodiment of the invention. Figure 6 This is a schematic diagram of the structural composition of the photodetector detection system based on different working conditions in the embodiments of the present invention. Detailed Implementation

[0010] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0011] Please see Figures 1 to 6 A detection method for a photodetector based on different operating conditions, applied to photodetector scenarios; the detection method for a photodetector based on different operating conditions includes: Step S11: Mark the detection area of ​​the photodetector, construct a primary twin based on the detection object and the corresponding environmental parameters of the detection area, simulate the primary twin under different working conditions, and output a multi-level working condition twin. Step S12: When the photodetector performs online detection on the target object, it collects multiple spectral data corresponding to the target object, inputs the multiple spectral data into the multi-level operating condition twin, and outputs the corresponding detection event; Step S13: Based on the identification of the detection event, determine the corresponding detection matching content, and determine the current detection mode and corresponding detection impact content of the photodetector by tracing the detection matching content. Trigger the mode optimization of the photodetector according to the detection impact content and the current working condition of the detection area to output the final detection mode of the photodetector. Step S14: In this final detection mode, multiple detection data are determined based on the photoelectric detector. The current detection result is determined based on the combination of multiple detection data, the corresponding detection object, and the corresponding environmental parameters. The dynamic change of the current detection result is triggered along with the update of the working condition change content.

[0012] refer to Figure 2 In step S11, the specific steps are as follows: S111: Perform primary detection on the photodetector, determine the detection area of ​​the photodetector during the detection process, mark the combination of environmental parameters of the detection area, determine the detection object of the detection area based on the traversal of the detection area, perform multi-level training on the position of the detection object and the corresponding combination of environmental parameters, and construct the corresponding primary twin during the training. S112: Based on the matching of the primary twin with the corresponding virtual space, and the matching of the virtual space with the corresponding working condition database, the simulation data of each working condition is determined. The simulation data of each working condition is clustered and analyzed, and standard working condition twins, disturbance working condition twins and extreme working condition twins are gradually formed. The standard working condition twins, disturbance working condition twins and extreme working condition twins are integrated and multi-level working condition twins are output.

[0013] In the embodiments of this application, a primary detection is performed on the photodetector, and the detection area of ​​the photodetector is determined during the detection process. The combination of environmental parameters of the detection area is marked. At the same time, the detection object of the detection area is determined based on the traversal of the detection area. The position of the detection object and the corresponding combination of environmental parameters are trained in multiple levels, and a corresponding primary twin is constructed during the training. This approach takes into account the overall consideration of the traversal of the detection area and ensures the accuracy of the detection object in the detection area.

[0014] At this point, a standardized preliminary test is performed on the photodetector. The device's broad spectral response characteristics in the visible to infrared region are measured using a spectrometer, and key indicators such as dark current, responsivity, and specific detectivity are tested to ensure that the device performance meets design expectations. During the test, the effective monitoring range of the detector is precisely defined based on the detector's field of view (FOV) and effective detection distance, combined with the spatial structure of the actual application scenario, thereby determining the detection area. This process maps the monitoring range in physical space to the effective boundary of the detector's signal response.

[0015] After determining the detection area, the environmental background within the area needs to be digitally mapped. At this time, a multi-dimensional sensor network is deployed to collect environmental parameters in the detection area in real time, including ambient light intensity, ambient temperature, relative humidity, and specific gas concentrations (such as dust or volatile organic compounds in a warehouse). These heterogeneous data are synchronized in time and registered spatially to construct environmental parameter combinations and perform feature labeling. At the same time, a detector or auxiliary vision system is used to perform a full-area scan of the detection area. Foreground targets are separated through background modeling and differential methods to determine the specific detection objects (such as goods, personnel, or vehicles in a warehouse) within the detection area and record their initial state information.

[0016] The identified detection object is treated as a physical unit, and its position coordinates are fused with the aforementioned marked environmental parameters. This data is then input into the training model for multi-level training. The first level of training focuses on the geometric features and spatial location correlation of the object, while the second level focuses on the spectral response characteristics and state change patterns of the object under different combinations of environmental parameters. Through this multi-level training, the system learns and masters the behavior patterns and physical properties of the detection object in a specific environment, and constructs a primary twin of the detection area. This primary twin not only includes the geometric topology of the detection area but also integrates the feature vectors of the detection object and the dynamic distribution model of the environmental parameters, forming a precise mapping of physical space in digital space and providing a data foundation for subsequent operational simulation.

[0017] Specifically, for warehouse environmental monitoring scenarios using photoelectric detectors, technicians used a standard blackbody radiation source and a monochromator to test the detector's responsivity in the visible to infrared bands to ensure it meets high-performance detection specifications. Based on the detector's installation height and wide-angle lens parameters, they calculated its projected coverage area on the ground and designated "shelf area A" and "main aisle" in the warehouse as specific detection areas.

[0018] The system calls the pre-installed temperature and humidity sensors and light sensors in the warehouse to record the current environment as "temperature 25℃, humidity 45%, and lighting brightness 300Lux", and marks it as the environmental parameter combination for the current time period; the detector starts the scanning mode to traverse "shelf A area" and uses image recognition technology to confirm that the current detection objects are "stacked cardboard boxes" and "a forklift in operation".

[0019] The system binds the forklift's position coordinates (X, Y) with the aforementioned environmental parameters, inputs them into a deep neural network for training, and learns the infrared thermal radiation characteristics and motion trajectory patterns of the forklift under different light intensities. Based on this training data, the system constructs a primary digital twin of warehouse area A in virtual space. This twin accurately replicates the warehouse's shelving layout and assigns digital attributes to the "cardboard boxes" and the "forklift" in an environment of "25℃, 300Lux". This completes the construction of a primary digital twin model containing physical entity information and environmental background features, laying the foundation for subsequent simulation of detection processes under different working conditions.

[0020] Furthermore, based on the matching of the primary twin with the corresponding virtual space, and the matching between the virtual space and the corresponding working condition database, the simulation data for each working condition is determined. Cluster analysis is performed on the simulation data for each working condition, and standard working condition twins, disturbance working condition twins, and extreme working condition twins are gradually formed. The standard working condition twins, disturbance working condition twins, and extreme working condition twins are integrated, and multi-level working condition twins are output. This approach takes into account the overall consideration of matching the virtual space and the corresponding working condition database, ensuring the accuracy of the simulation data for each working condition.

[0021] At this point, based on the primary twin constructed in step S111, the system uses a 3D rendering engine and a physical simulation platform to load it into the corresponding virtual space. This process not only requires proportional replication of geometric dimensions, but also calibration of optical properties within the virtual space. For example, based on the spectral response characteristics of two-dimensional TMDs materials, the corresponding light source wavelength, radiation intensity, and background noise parameters are set in the virtual environment to ensure that the virtual space can truly reflect the response characteristics of the photodetector in the visible to infrared light region, thereby completing the accurate mapping of the physical entity to the digital space.

[0022] After completing the virtual space matching, the system calls the pre-set operating condition database for deep correlation. The operating condition database stores historical environmental monitoring data, typical interference modes, and extreme environment records. The system uses the combination of environmental parameters (such as reference temperature, humidity, and illuminance) in the primary twin as an index to search for the operating condition template with the highest similarity in the database. Using methods such as Monte Carlo simulation, dynamic simulation data under various operating conditions are generated in the virtual space. These data cover the changes in spectral reflectance of the probe object under different operating conditions, fluctuations in environmental background noise, and changes in light intensity distribution within the detector's field of view, providing multi-dimensional data support for subsequent analysis.

[0023] After acquiring massive amounts of simulation data, the system uses clustering analysis methods (such as K-Means or DBSCAN) to perform feature mining and classification on the data; based on the steady-state nature of environmental parameters, the intensity of interference sources, and the degree of distortion of the detection signal, the simulation data is clustered into three core categories: Standard operating condition twin: generated by clustering data with stable environmental parameters and minimal interference factors, reflecting the ideal detection environment of the detection area under normal stable conditions, corresponding to the standard response mode of the detector; Interference twin: generated by clustering data containing random noise, sudden changes in illumination or interference from non-target objects, simulating complex environments such as personnel movement and light flickering in a warehouse, used to train the detector's anti-interference capabilities. Extreme condition twins: generated by clustering data with environmental parameters close to the detector threshold and extremely low signal-to-noise ratio, simulating extreme conditions such as complete darkness, high temperature and humidity, or strong electromagnetic interference, to verify the detector's extreme performance; based on the clustering results, the system generates the above three types of twins respectively, realizing the transformation from data to typical scenario models.

[0024] The system organically integrates standard, disturbance, and extreme condition twins through multi-source fusion technology; it constructs a hierarchical multi-level condition twin architecture, in which the standard condition serves as the baseline layer, the disturbance condition as the dynamic layer, and the extreme condition as the boundary layer. The system packages and outputs these three types of twins through a unified data interface, forming a comprehensive and hierarchical multi-level condition twin. This twin not only includes the static environmental background but also embeds dynamic condition evolution logic, enabling it to automatically switch to the corresponding twin model based on the input real-time condition information, providing a comprehensive digital foundation for event matching and pattern optimization in subsequent online detection.

[0025] Specifically, the primary twin of "Shelf Area A" constructed in S111 is loaded into the high-fidelity virtual simulation platform; based on the wide spectral response characteristics (visible to infrared) of the photodetector based on two-dimensional TMDs heterojunction material, the optical properties of the warehouse, such as the light spectrum distribution and wall reflectivity, are set in the virtual space to ensure that the optical environment of the virtual warehouse is highly consistent with the real environment.

[0026] The system uses the currently marked parameters "temperature 25℃, humidity 45%" as an index to retrieve historical data from the same period in the working condition database; it simulates and generates a series of scene data streams, such as "light fluctuation data during normal daytime operation", "thermal radiation data under nighttime infrared monitoring mode", and "shading and reflection data when forklifts pass by".

[0027] The above data streams are clustered: data with stable lighting and no personnel movement are clustered to generate a standard working condition twin, representing the warehouse's quiet nighttime state; data containing light flickering and infrared heat sources generated by personnel movement are clustered to generate an interference working condition twin, simulating the warehouse's busy daytime operating environment; and data simulating smoke, strong light shining directly on the detector lens, or in a completely dark environment are clustered to generate an extreme working condition twin, simulating fire warning or lighting failure scenarios.

[0028] The system integrates the above three types of twins into a comprehensive multi-level working condition twin for "shelf area A". When the detector is actually running, if a sudden change in ambient light is detected, the system can immediately call the "interference working condition twin" for comparative analysis. If a smoke signal is detected, it will automatically switch to the "extreme working condition twin" for verification, thereby achieving comprehensive digital coverage and dynamic adaptation to the complex environment of the warehouse.

[0029] refer to Figure 3 In step S12, the specific steps are as follows: S121: Real-time monitoring of the photodetector's detection of the object, and triggering the photodetector to conduct online detection of the object to collect the corresponding data cluster, and determining the multiple spectral data corresponding to the object based on the traversal of the data cluster; S122: Multiple spectral data are input to the multi-level operating condition twin along the corresponding data channels, and the multi-level operating condition twin is triggered to match the multiple spectral data to output the corresponding matching results. Based on the identification of the matching results, the twin that best fits the current environmental parameters is determined. Based on the matched twin and excluding environmental noise and background interference, the corresponding detection event is generated. The detection event includes the feature identification of the detection object and the current matching process.

[0030] In the embodiments of this application, the photodetector is monitored in real time to detect the object and is triggered to conduct online detection of the object to collect the corresponding data cluster. Based on the traversal of the data cluster, multiple spectral data corresponding to the object are determined, which is compatible with the overall consideration of the traversal of the data cluster and ensures the accuracy of the multiple spectral data corresponding to the object.

[0031] At this time, the system monitors the operating status of the photodetector in real time, focusing on its operating temperature, bias voltage stability, and dark current level to ensure that the device is in the optimal linear response range. After confirming that the detector is in normal condition, the system sends a trigger command to start the online detection mode. At this time, the detector actively scans the target object in the detection area according to the preset sampling frequency and integration time. This process marks the transition of the detection system from standby state to high-load data acquisition state, providing raw signal input for subsequent spectral analysis.

[0032] For online detection mode, the photodetector continuously observes the target object based on its wide spectral response characteristics (from visible light to infrared region). The data collected by the system not only includes the photocurrent response signal of the detector, but also synchronously correlates the spatial position coordinates, velocity vector of the target object, and current environmental background parameters. These heterogeneous data are aligned and packaged in the time dimension to form a multi-dimensional data cluster. This data cluster not only reflects the optical characteristics of the target object, but also embeds environmental background noise information, providing comprehensive data support for extracting effective signals under complex working conditions, reflecting the "data-driven" detection concept.

[0033] After acquiring the data cluster, the system performs a traversal to deeply analyze the time-series signals within the cluster. The system maps the original photocurrent response signal to the electromagnetic spectrum domain and converts the response intensity at different time slices into reflectance or radiative intensity data for the corresponding wavelengths (such as the visible light band and the near-infrared band) through an inversion method. This process is essentially stripping away environmental interference from the composite signal and accurately extracting the unique spectral fingerprint information of the detected object. After the traversal is completed, the system outputs a set of discrete, physically meaningful spectral data sequences, which fully characterize the optical properties of the detected object at that moment.

[0034] Specifically, the system performs real-time status monitoring and online detection triggering of the detectors; the control center monitors that the current operating temperature of the photoelectric detector installed on the top of the warehouse is stable at 25℃, which meets the high-performance detection standard; at this time, the system recognizes that a forklift has entered the preset "shelf A zone" detection area, and then triggers the online detection command, and the detector activates the high-frequency scanning mode to prepare to capture target information.

[0035] The detector utilizes its wide spectral response advantage based on two-dimensional TMDs heterojunction to begin recording the reflected light signal from the forklift. The system synchronously packages the detector's photocurrent output with the forklift's position coordinates (obtained through the positioning system), travel speed (approximately 1.5 m / s), and the warehouse's current illumination intensity (300 Lux) and dust concentration parameters, forming a composite data cluster containing optical signals, motion information, and environmental background.

[0036] The system performs frame-by-frame traversal analysis on the collected data clusters; using spectral inversion, it converts the differences in electrical signals recorded by the detector at different times into spectral features; for example, it identifies the high reflectivity peak of the yellow forklift body near the wavelength of 580nm, and the specific absorption characteristics of its rubber tires in the infrared band (such as 950nm); the traversal process effectively filters out interference signals caused by the flickering of warehouse lighting, and finally outputs multiple spectral data that can uniquely identify the "yellow forklift", laying the foundation for subsequent event matching in the twin.

[0037] Furthermore, multiple spectral data are input to the multi-level operating condition twin along their corresponding data channels, triggering the multi-level operating condition twin to match the multiple spectral data and output the corresponding matching results. Based on the identification of these matching results, the twin that best fits the current environmental parameters is determined. Based on the matched twin and after removing environmental noise and background interference, the corresponding detection event is generated. The detection event includes feature identification of the detection object and the current matching process, incorporating overall considerations for the identification of matching results. This ensures the accuracy of the twin that best fits the current environmental parameters. At the same time, it realizes the conversion between the primary twin and the multi-level operating condition twin, and further manages the multi-level operating condition twin, improving the accuracy of the detection event.

[0038] At this point, based on the multiple spectral data extracted in step S121, the system transmits these discrete data streams containing specific wavelength information to the multi-level operating condition twin system in real time through a preset data channel; the system then triggers a matching mechanism to compare the input spectral data feature vector (such as reflectivity and absorption peak position in a specific band) with the feature thresholds in the multi-level operating condition twin library. This process utilizes the wide spectral response characteristics to ensure that the input data can cover feature information from the visible light to the infrared region, thereby activating the corresponding data processing module in the virtual space.

[0039] After initial matching, the system incorporates real-time monitored environmental parameters (such as ambient light intensity, temperature, and humidity background values) as weighting factors to refine the matching results. The system calculates the similarity distance (such as Euclidean distance or cosine similarity) between the input spectral data and twins for standard, interference, and extreme conditions. Based on the principle of minimizing distance, the system identifies the twin model that best fits the current environmental parameters, effectively solving the misjudgment problem that may exist in single spectral matching. This ensures that the digital model relied upon for subsequent calculations is highly consistent with the actual working conditions in the physical world, achieving precise positioning of "virtual-real mapping".

[0040] Based on the determined optimal twin model, the system performs in-depth analysis of the input spectral data; using the pre-set background interference model in the twin (such as the thermal radiation background of the warehouse wall and the flicker noise of the lighting source), the system performs background subtraction and denoising on the original spectral data, thereby removing environmental noise interference; the system performs feature recognition on the purified spectral features and extracts the key attributes of the detected object (such as material, outline, and state); the system encapsulates the identified object features and matching process (i.e., the complete path from data input to model determination) and outputs the structured detection event, completing the transformation from the original spectral signal to high-dimensional semantic information.

[0041] Specifically, in step S121, the photodetector extracts multiple spectral data about the "yellow forklift" (including strong reflection signals in the 580nm visible light band and thermal radiation signals in the 950nm infrared band). These data are transmitted in real time to the multi-level working condition twin system through an independent data channel. After receiving the data, the system immediately triggers a matching procedure to perform a preliminary comparison of these spectral features with the vehicle spectral templates pre-stored in the twin library.

[0042] The environmental sensors inside the warehouse showed a current illuminance of 300 Lux, accompanied by intermittent light flickering (characteristics of interference conditions), rather than a silent nighttime state (standard operating conditions). After comprehensive evaluation, the system found that the noise distribution in the input spectral data highly matched the background noise model in the "interference condition twin". Therefore, the system determined that the "interference condition twin" was the most fitting digital model and locked it as the basis for subsequent calculations.

[0043] Based on the selected "interference condition twin," the system performs background subtraction on the input spectral data, effectively filtering out the interference ripples generated by the warehouse LED lights in specific bands. The purified spectral data clearly shows the reflection peak of the forklift's yellow coating and the absorption peak of the rubber tires. Based on this, the system identifies the detection object as a "forklift" and records the complete process of "successful matching based on the interference condition model." The system outputs a detection event containing information such as "identified object: forklift," "location: A-zone channel," and "matching path: interference condition > spectral feature comparison > noise reduction and identification," thus completing the accurate qualitative analysis of this detection task.

[0044] refer to Figure 4 In step S13, the specific steps are as follows: S131: Dynamically identify the detection event, determine multiple detection features during the identification process, determine the corresponding detection matching content based on the multiple detection features and the current data of the detection object, mark the associated route of the detection matching content in the detection matching content, and trace along the associated route to determine the current detection mode of the photodetector; S132: Follow the associated route of the detected matching content again, and introduce the dynamic changes of the detection area during the tracing process to determine the corresponding detection impact content. At the same time, collect the current working conditions of the detection area. S133: Determine the corresponding mode optimization content based on the detection impact content and the current operating conditions of the detection area, and further trigger the mode optimization of the photodetector to monitor the optimization results of the current detection mode of the photodetector in real time, and determine the final detection mode of the photodetector as the optimization results change.

[0045] In the embodiments of this application, the detection event is dynamically identified, and multiple detection features are determined during the identification process. Based on the current data of the multiple detection features and the detection object, the corresponding detection matching content is determined. In the detection matching content, the associated route of the detection matching content is marked, and the tracing is carried out along the associated route to determine the current detection mode of the photodetector. This approach takes into account the overall consideration of multiple detection features and the current data of the detection object, ensuring the accuracy of the corresponding detection matching content.

[0046] At this point, based on the selected "interference condition twin," the system performs background subtraction on the input spectral data, effectively filtering out the interference ripples generated by the warehouse LED lights in specific bands. The purified spectral data clearly shows the reflection peak of the forklift's yellow coating and the absorption peak of the rubber tires. Based on this, the system identifies the detection object as a "forklift" and records the complete process of "successful matching based on the interference condition model." The system outputs a detection event containing information such as "identified object: forklift," "location: A-zone channel," and "matching path: interference condition > spectral feature comparison > noise reduction and identification," thus completing the accurate qualitative analysis of this detection task.

[0047] After acquiring multiple detection features, the system deeply integrates them with the current real-time data of the detected object (such as location coordinates, movement speed, and temperature information). Through feature vector space calculation, the system retrieves the content with the highest similarity to the current feature combination from a pre-set database. The system prioritizes matching templates that have significant response features in specific bands. The final detection matching content includes not only the category label of the detected object but also its current state description (such as "normal movement", "abnormal stagnation", or "feature distortion"), thereby achieving a qualitative judgment of the detection scene.

[0048] Based on the determined detection and matching content, the system uses knowledge graph technology to mark the associated routes of this content. This associated route is essentially a logical chain connecting the underlying physical signals and the top-level detection decisions, recording the complete path from "spectral data input" to "operating condition model matching" and then to "feature recognition". The system traces back along this associated route to analyze the parameter configuration and mode logic adopted by the photodetector when processing the event, thereby determining the current detection mode of the photodetector (e.g., "high sensitivity dynamic tracking mode" or "strong light background suppression mode"), making the implicit mode logic explicit, and providing clear intervention nodes for subsequent mode optimization.

[0049] Specifically, for the "forklift recognition" detection event generated in S122, the system dynamically analyzes it and extracts several specific features: a strong reflection peak is detected in the visible light band (about 580nm), a corresponding thermal radiation distribution is detected in the infrared band (about 950nm), and the photocurrent signal shows periodic fluctuations that are proportional to the forklift's moving speed; based on the detector's performance parameters, the system converts these raw signals into standardized feature vectors.

[0050] The system compares the above spectral features with the forklift's current real-time data (current speed 1.5 m / s, located in the middle of the A-zone aisle); based on the advantage of the photodetector's wide spectral response, the system confirms that the feature combination highly matches the "heavy handling equipment - normal operation" template in the database; therefore, the detected matching content is determined to be "yellow forklift in normal operation", ruling out the possibility of "unmanned transport vehicle" or "stationary obstacle".

[0051] The system marks the associated route in the matching results: starting from "multi-channel spectral input", after "interference condition model denoising" processing, it finally reaches "dynamic target feature matching"; along this route, the system traces back and finds that in order to clearly capture forklift signals against the complex lighting background of the warehouse, the detector has automatically enabled "wide-spectrum dynamic enhancement mode". This mode ensures stable locking of moving targets by adjusting the internal gain and background subtraction method; at this point, the system has clarified the current detection mode and its effective logical path.

[0052] Furthermore, the correlation route of the detected matching content is traced again, and the dynamic changes of the detection area are introduced during the tracing process to determine the corresponding detection impact content. At the same time, the current operating conditions of the detection area are collected, which takes into account the overall consideration of the dynamic changes of the detection area introduced during the tracing process, and ensures the accuracy of the corresponding detection impact content.

[0053] At this point, the system traces the associated route of the detected matching content again. This time, the tracing is not a simple path playback, but combines time series analysis. During the tracing process, the system actively introduces the dynamic changes in the detection area, that is, compares the environmental differences between the current time and the initial modeling time (such as changes in illumination gradient and fluctuations in background thermal radiation). The system maps these dynamic variables to each node of the associated route, analyzes the potential impact of environmental changes on the signal transmission link, and thus evaluates the timeliness and accuracy of the detection logic.

[0054] After introducing dynamic factors, the system calculates the specific impact of these changes on the detection results, thereby determining the content of the detection impact and quantifying the interference of environmental parameter drift (such as the increase in dark current caused by temperature rise and the saturation of photogenerated carriers caused by background light enhancement) on the detector performance indicators. The system generates a detection impact report, clearly indicating which components in the current detection signal are generated by target features and which components are deviations introduced by dynamic environmental changes, providing accurate negative feedback basis for subsequent mode optimization.

[0055] To verify the validity of the tracing results and lock in the current state, the system synchronously triggered the acquisition task of the current operating conditions of the detection area. Utilizing the multi-dimensional sensor network deployed in the detection area, the system captures the current combination of environmental parameters in real time (including ambient temperature, relative humidity, instantaneous light intensity, and electromagnetic background noise, etc.), ensuring that the operating condition model in the digital twin system can maintain a high degree of synchronization with the real-time state of the physical world, and providing accurate boundary conditions for subsequent mode optimization of the photodetector.

[0056] Specifically, for the "forklift operation" event being monitored, the system traces back along the previously determined associated route (from spectral input to feature matching); the system retrieves the environmental change log of the area over the past 10 minutes and finds that the ambient light intensity in the warehouse gradually increased from the initial 300 Lux to 450 Lux, which is due to the opening of the unloading gate and the introduction of natural light; the system incorporates this dynamic light change into the tracing logic to check whether it interferes with the spectral recognition of the "forklift".

[0057] Based on the introduced dynamic illumination changes, the system analysis concluded that the influx of natural light led to a significant increase in background noise in the visible light band, which may have masked the characteristic signals of the forklift body in specific bands (such as the red reflection peak). The system determined that the detection impact was "the decrease in target contrast caused by strong background light interference" and calculated that the background noise level of photogenerated carriers increased by about 15%, which will affect the detector's response rate and signal-to-noise ratio.

[0058] To accurately correct this deviation, the system instructs the sensor network to perform instantaneous sampling of "shelf area A". The sampling results show that the current ambient temperature is 26°C, humidity is 50%, light intensity is 450 Lux, and there is occasional electromagnetic interference (from a nearby running conveyor belt motor). This real-time operating data is immediately uploaded as a key input parameter for subsequent adjustments to the photodetector's operating mode (such as switching to infrared-dominated mode or increasing the gain threshold).

[0059] Therefore, based on the detection impact and the current operating conditions of the detection area, the corresponding mode optimization content is determined, and the mode optimization of the photodetector is further triggered to monitor the optimization results of the current detection mode of the photodetector in real time. As the optimization results change, the final detection mode of the photodetector is determined, which takes into account both the detection impact and the current operating conditions of the detection area, and ensures the accuracy of the corresponding mode optimization content.

[0060] At this point, the system will perform multi-parameter coupling analysis on the detected influencing factors (such as background noise intensity and signal attenuation rate) and the current operating conditions of the detection area (such as real-time temperature and humidity and illuminance). The system will construct an optimization objective function and calculate the optimal parameter combination that can offset the environmental influence. The specific optimization of the mode includes adjusting the detector's operating bias voltage, changing the signal integration time, switching the spectral response weight, or enabling the corresponding background filtering mode, so as to ensure that the detector can still maintain a high signal-to-noise ratio and high response rate under the current complex operating conditions.

[0061] After determining the optimization content, the system sends a command to the photodetector through the control interface to trigger mode optimization. This process is a dynamic closed loop. The system monitors the detector's response to the command and changes in key performance indicators (such as dark current level and photocurrent gain) in real time. The system compares the actual feedback data with the theoretical prediction model in real time. If the optimization effect is found to be unsatisfactory, the system will fine-tune the parameters until the detector performance indicators converge to the optimal range, ensuring the smoothness and effectiveness of mode switching.

[0062] As the optimization results stabilize, the system locks in all operating parameters of the detector. The system comprehensively evaluates key indicators such as the optimized response rate, detectivity, and signal-to-noise ratio to confirm that it has adapted to the current working conditions and meets the detection requirements. At this point, the system officially outputs the final detection mode of the photodetector. This mode not only includes the hardware operating parameter configuration but also defines the logical thresholds for subsequent data processing methods, marking that the detector has transitioned from an adaptive adjustment state to a stable final detection state, laying the foundation for subsequent high-precision data acquisition.

[0063] Specifically, regarding the detection impact identified in step S132 as "natural light influx leading to enhanced background noise in the visible light band," and considering the current operating conditions (illuminance 450 Lux), the system, based on the broad spectral response characteristics of the two-dimensional TMDs material, decides to implement an optimization strategy of "infrared enhancement and visible light suppression." The system calculates that the gain weight of the detector in the infrared band (around 950 nm) needs to be increased, while simultaneously activating dynamic background subtraction to suppress clutter in the visible light band, generating a specific mode optimization parameter package.

[0064] The optimized parameters are sent to the photodetector to adjust the bias voltage of its internal circuit and update the signal processing firmware. The system monitors the detector's output signal in real time and finds that the infrared characteristic peak in the spectral response curve is significantly enhanced, while the background noise in the visible light region is reduced by about 20%. The monitoring data shows that the detector is smoothly transitioning to a state resistant to strong light interference, and the carrier mobility remains at a high efficiency level.

[0065] When the detector's signal-to-noise ratio stabilizes above 25dB and the infrared thermal outline of the forklift can be clearly distinguished, the system determines that the optimization is successful. The system then locks the current "infrared-dominated anti-interference mode" as the final detection mode. In this mode, the detector will focus on capturing the radiation characteristics of objects in the infrared band, effectively avoiding natural light interference caused by the opening of the warehouse roller shutter door, and ensuring continuous and accurate tracking and monitoring of the forklift.

[0066] refer to Figure 5 In step S14, the specific steps are as follows: S141: In the final detection mode, multiple detection data of the photodetector are continuously acquired, including real-time photocurrent intensity, response time, and quantum efficiency; a corresponding detection framework is constructed based on the multiple detection data and the corresponding detection object, and a deep detection of the combination of environmental parameters is triggered based on the detection framework to output the current detection result. The current detection result is not a static value, but a comprehensive information package containing real-time spectral characteristics, target attributes, and environmental background. S142: As the detection process continues, the operating conditions change and the changes in operating conditions are updated in real time to trigger a dynamic calibration mechanism for the detection results. In the dynamic calibration mechanism for the detection results, a multi-level operating condition twin is used to predict the trend of the results at the next moment, so as to realize the dynamic change of the current detection results.

[0067] In the embodiments of this application, in the final detection mode, multiple detection data from the photodetector are continuously acquired. These multiple detection data include real-time photocurrent intensity, response time, and quantum efficiency. A corresponding detection framework is constructed based on the multiple detection data and the corresponding detection object. Based on this detection framework, in-depth detection of the combination of environmental parameters is triggered to output the current detection result. This current detection result is not a static value, but a comprehensive information package containing real-time spectral characteristics, target attributes, and environmental background. It takes into account the overall consideration of multiple detection data and the corresponding detection object, ensuring the accuracy of the corresponding detection framework.

[0068] At this point, in the final detection mode, the photodetector continuously outputs multiple key detection data based on its high-performance photoelectric response characteristics. The system focuses on acquiring real-time photocurrent intensity, which directly reflects the photogenerated carrier generation efficiency of the detector in a specific wavelength band, limited by the band structure of the material. At the same time, it acquires response time to evaluate the detector's transient response capability to changes in optical signal, as well as quantum efficiency to measure the efficiency of incident photons converting into electrons. These data constitute the basic flow of underlying physical sensing, ensuring that the data source for subsequent analysis has high signal-to-noise ratio and wide spectral response characteristics.

[0069] Based on the collected multidimensional detection data, the system constructs a dynamic detection framework by combining the known characteristics of the detected object (such as its position and trajectory). This framework is essentially an analysis model that integrates physical signals and logical attributes, and it spatiotemporally correlates the fluctuations in photocurrent intensity with the actions of the detected object. Based on this framework, the system triggers in-depth detection of the combination of environmental parameters. Unlike the macroscopic environmental perception in primary detection, in-depth detection utilizes the high sensitivity of the detector to capture deep parameters such as the minute flicker of ambient light and subtle changes in air refractive index. The system aligns and analyzes these environmental parameters with the detection data, removing the interference of the environmental background on the target signal, thereby restoring the true physical state of the detected object in the current environment.

[0070] After in-depth detection and data fusion, the system outputs the current detection result. This result is no longer limited to a simple "present / absent" judgment or a simple numerical reading, but is encapsulated into a comprehensive information package containing multi-dimensional information. This information package includes real-time spectral features, recording the spectral fingerprint of the detected object in the visible to infrared light region; target attributes, clarifying the material, temperature and motion state of the detected object; and environmental background, describing the specific impact of current illumination, temperature and humidity on the detection process. This structured output method provides comprehensive data support for subsequent intelligent decision-making, demonstrating the advantages of the new photodetector based on two-dimensional TMDs heterojunction materials in terms of information acquisition depth.

[0071] Specifically, in the "infrared-dominated anti-interference mode", the photodetector continuously tracks the forklift in "shelf area A". The system reads the output signal of the detector in real time, and obtains that the photocurrent intensity is stable at the milliampere level and the response time is maintained at the millisecond level to capture the instantaneous changes in the movement of the forklift. The quantum efficiency data shows that the photoelectric conversion efficiency of the detector in the infrared band is at its peak. These data streams are continuously input to the processing unit, reflecting the real-time optical characteristics of the forklift.

[0072] Based on the periodic fluctuations in photocurrent intensity, the system constructs a detection framework for "moving forklifts". Guided by this framework, the system triggers in-depth detection of the surrounding environment, focusing on analyzing the infrared reflection characteristics of the forklift's metal surface under the current illumination conditions (450 Lux natural light background). By comparing the coupling relationship between the photocurrent signal and the ambient illumination parameters, the system accurately isolates the background noise caused by changes in natural light at the unloading port and locks in the infrared characteristic signal generated by the forklift's engine heat source.

[0073] The system generates a current detection result information package, which includes: real-time spectral characteristics (yellow body reflection peak of the forklift in the 580nm band and thermal radiation peak in the 950nm band), target attributes (category: heavy forklift, status: traveling at a speed of 1.5 m / s, engine temperature: 85℃), and environmental background (background light intensity: 450 Lux, ambient temperature: 26℃). This comprehensive information package is transmitted to the warehouse management center in real time, which not only informs the management personnel that a vehicle has passed by, but also provides detailed operating status of the vehicle and environmental background data, realizing a deep perception of the warehouse scene.

[0074] Furthermore, as the detection process continues, the operating conditions change, and the changes in operating conditions are updated in real time to trigger a dynamic calibration mechanism for the detection results. In this dynamic calibration mechanism, a multi-level operating condition twin is used to predict the trend of the results at the next moment, thereby realizing the dynamic change of the current detection results. This introduces the dynamic change of the current detection results. At the same time, a final detection mode is introduced, and multiple detection data, corresponding detection objects, and corresponding environmental parameter combinations are fully considered, which improves the accuracy of the current detection results and dynamically manages the changes in operating conditions, enabling the photoelectric detector to perform efficient detection based on different operating conditions.

[0075] At this time, as the detection time continues, the external environment is not static. The system uses a sensor network to continuously monitor the state of the detection area. Based on its sensitivity to environmental factors (such as temperature and light), the system captures subtle shifts in operating conditions in real time, such as gradual changes in ambient light, optical path disturbances caused by atmospheric turbulence, or the accumulation of background thermal radiation. The system calculates the difference between the operating parameters at the current moment and the previous moment and updates the changes in operating conditions in real time. This process ensures that the system always keeps abreast of the latest physical environment state, providing a basis for judging whether the detector performance needs dynamic compensation.

[0076] Because changes in operating conditions can cause key parameters of the detector, such as dark current and responsivity, to drift (e.g., increased temperature may lead to increased dark current), the system needs to correct the current detection results in real time, mapping changes in operating conditions into correction factors for detector performance indicators. The system uses built-in methods to counteract non-target signal interference caused by environmental changes, ensuring the accuracy and consistency of the output results and preventing false alarms or missed alarms caused by environmental fluctuations.

[0077] Based on the current trend of changing operating conditions (such as continuous increase in light intensity), the twin performs rapid simulations in virtual space to predict the possible changes in detector response characteristics and target spectral features at the next moment. The system compares the predicted trend with the real-time acquired data to form a closed-loop feedback. The current detection result is no longer an isolated static value, but rather, as the operating conditions evolve and time progresses, it continuously integrates predicted information and calibration data to achieve dynamic changes. This dynamic result truly reflects the real-time state of the detected object in a continuously changing environment, demonstrating the intelligent adaptive capability of the detection system.

[0078] Specifically, during the continuous monitoring of the forklifts in "Rack Area A", as time went on, the clouds outside the warehouse gradually dispersed, and more natural light shone in through the unloading port, causing the light intensity in the detection area to gradually increase from 450 Lux to 600 Lux, and the ambient temperature rose slightly accordingly; the system captured this change in real time, updated the changes in operating conditions, and marked it as "increased trend of strong light interference".

[0079] Given that a surge in light intensity may cause the detector to saturate or decrease in signal-to-noise ratio in the visible light band, the system immediately triggers dynamic calibration. The system automatically adjusts the detector's signal processing threshold and introduces a negative feedback factor to offset the increase in background photocurrent. This operation ensures that the infrared feature signal of the forklift can still be effectively extracted even in strong light, maintaining the stability of the detection results.

[0080] Given that a surge in light intensity may cause the detector to saturate or decrease in signal-to-noise ratio in the visible light band, the system immediately triggers dynamic calibration. The system automatically adjusts the detector's signal processing threshold and introduces a negative feedback factor to offset the increase in background photocurrent. This operation ensures that the infrared feature signal of the forklift can still be effectively extracted even in strong light, maintaining the stability of the detection results.

[0081] Please see Figure 6 , Figure 6 This is a schematic diagram of the structural composition of the photodetector detection system based on different operating conditions in an embodiment of the present invention; the photodetector detection system based on different operating conditions is applied to the above-mentioned photodetector detection method based on different operating conditions; the photodetector detection system based on different operating conditions includes: The twin module 21 is used to mark the detection area of ​​the photodetector. A primary twin is constructed based on the detection object and the corresponding environmental parameters in the detection area. The primary twin is simulated under different working conditions to output a multi-level working condition twin. The detection event module 22 is used to collect multiple spectral data corresponding to the detection object when the photoelectric detector performs online detection on the detection object, input the multiple spectral data into the multi-level operating condition twin, and output the corresponding detection event; The detection mode optimization module 23 is used to determine the corresponding detection matching content based on the identification of the detection event, and to determine the current detection mode and corresponding detection impact content of the photodetector by tracing the detection matching content. Based on the detection impact content and the current working condition of the detection area, the mode optimization of the photodetector is triggered to output the final detection mode of the photodetector. The operating condition change module 24 is used to determine multiple detection data based on the photodetector in the final detection mode, determine the current detection result based on the combination of multiple detection data, the corresponding detection object and the corresponding environmental parameters, and trigger the dynamic change of the current detection result along with the update of the operating condition change content.

[0082] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A detection method for a photoelectric detector based on different operating conditions, characterized in that, include: The detection area of ​​the photodetector is marked, and a primary twin is constructed based on the detection object and the corresponding environmental parameters in the detection area. The primary twin is then simulated under different operating conditions to output a multi-level operating condition twin. When the photodetector performs online detection on the target object, it collects multiple spectral data corresponding to the target object, inputs the multiple spectral data into the multi-level operating condition twin, and outputs the corresponding detection event; Based on the identification of the detection event, the corresponding detection matching content is determined, and the current detection mode and corresponding detection impact content of the photodetector are determined by tracing the detection matching content. Based on the detection impact content and the current operating conditions of the detection area, the mode optimization of the photodetector is triggered to output the final detection mode of the photodetector. In this final detection mode, multiple detection data are determined based on the photoelectric detector. The current detection result is determined based on the combination of multiple detection data, the corresponding detection object, and the corresponding environmental parameters. The dynamic change of the current detection result is triggered as the working condition changes.

2. The photodetector detection method based on different operating conditions according to claim 1, characterized in that, The detection area of ​​the marked photodetector is used to construct a primary twin based on the detection object and corresponding environmental parameters within that area. This primary twin is then used to simulate different operating conditions to output a multi-level operating condition twin, including: A primary detection is performed on the photodetector, and the detection area of ​​the photodetector is determined during the detection process. The combination of environmental parameters in the detection area is marked. At the same time, the detection object in the detection area is determined based on the traversal of the detection area. The location of the detection object and the corresponding combination of environmental parameters are trained in multiple levels, and the corresponding primary twin is constructed during the training.

3. The photodetector detection method based on different operating conditions according to claim 2, characterized in that, The detection area of ​​the marked photodetector is used to construct a primary twin based on the detection object and corresponding environmental parameters within that detection area. This primary twin is then used to simulate different operating conditions to output a multi-level operating condition twin. The system also includes: Based on the matching of the primary twin with the corresponding virtual space, and the matching of the virtual space with the corresponding working condition database, the simulation data of each working condition is determined. Cluster analysis is performed on the simulation data of each working condition, and standard working condition twins, disturbance working condition twins and extreme working condition twins are gradually formed. The standard working condition twins, disturbance working condition twins and extreme working condition twins are integrated and multi-level working condition twins are output.

4. The photodetector detection method based on different operating conditions according to claim 1, characterized in that, When the photodetector performs online detection of the target object, it collects multiple spectral data corresponding to the target object, inputs the multiple spectral data into a multi-level operating condition twin, and outputs the corresponding detection event, including: The photodetector is monitored in real time to detect the object and is triggered to conduct online detection of the object to collect the corresponding data cluster. Based on the traversal of the data cluster, multiple spectral data corresponding to the object are determined.

5. The photodetector detection method based on different operating conditions according to claim 4, characterized in that, The process of collecting multiple spectral data corresponding to the object being detected during online detection by the photodetector, inputting the multiple spectral data into a multi-level operating condition twin, and outputting the corresponding detection event also includes: Multiple spectral data are input to a multi-level operating condition twin along their corresponding data channels, triggering the matching of the multi-level operating condition twin with the multiple spectral data to output the corresponding matching results. Based on the identification of the matching results, the twin that best fits the current environmental parameters is determined. Based on the matched twin and the removal of environmental noise and background interference, the corresponding detection event is generated. The detection event includes the feature identification of the detection object and the current matching process.

6. The photodetector detection method based on different operating conditions according to claim 1, characterized in that, The process of identifying the detection event, determining the corresponding detection matching content, tracing the detection matching content to determine the current detection mode and corresponding detection impact content of the photodetector, and triggering mode optimization of the photodetector based on the detection impact content and the current operating conditions of the detection area to output the final detection mode of the photodetector includes: The detection event is dynamically identified, and multiple detection features are determined during the identification process. Based on the multiple detection features and the current data of the detection object, the corresponding detection matching content is determined. In the detection matching content, the associated route of the detection matching content is marked, and the tracing is carried out along the associated route to determine the current detection mode of the photodetector.

7. The photodetector detection method based on different operating conditions according to claim 6, characterized in that, The process of determining the corresponding detection matching content based on the identification of the detection event, determining the current detection mode and corresponding detection impact content of the photodetector by tracing the detection matching content, triggering mode optimization of the photodetector based on the detection impact content and the current operating conditions of the detection area, and outputting the final detection mode of the photodetector, further includes: The detection and matching content is traced again along the associated route, and the dynamic changes in the detection area are introduced during the tracing process to determine the corresponding detection impact content. At the same time, the current operating conditions of the detection area are collected. Based on the content of the detection impact and the current operating conditions of the detection area, the corresponding mode optimization content is determined, and the mode optimization of the photodetector is further triggered to monitor the optimization results of the current detection mode of the photodetector in real time, and the final detection mode of the photodetector is determined as the optimization results change.

8. The photodetector detection method based on different operating conditions according to claim 1, characterized in that, In this final detection mode, multiple detection data are determined based on a photodetector. The current detection result is determined based on the combination of these multiple detection data, the corresponding detection object, and the corresponding environmental parameters. The dynamic change of this current detection result is triggered along with updates to the operating conditions, including: In the final detection mode, multiple detection data from the photodetector are continuously acquired, including real-time photocurrent intensity, response time, and quantum efficiency. Based on the multiple detection data and the corresponding detection object, a corresponding detection framework is constructed, and based on the detection framework, in-depth detection of the combination of environmental parameters is triggered to output the current detection result. This current detection result is not a static value, but a comprehensive information package containing real-time spectral characteristics, target attributes, and environmental background.

9. The photodetector detection method based on different operating conditions according to claim 8, characterized in that, In this final detection mode, multiple detection data are determined based on a photodetector. The current detection result is determined based on the combination of these multiple detection data, the corresponding detection object, and the corresponding environmental parameters. The dynamic change of this current detection result is triggered as the operating conditions change. The method also includes: As the detection process continues, the operating conditions change and the changes in operating conditions are updated in real time to trigger a dynamic calibration mechanism for the detection results. In the dynamic calibration mechanism for the detection results, a multi-level operating condition twin is used to predict the trend of the results at the next moment, so as to realize the dynamic change of the current detection results.

10. A photoelectric detector-based detection system for different operating conditions, characterized in that, The photodetector based on different operating conditions detection system is applied to the photodetector based on different operating conditions detection method as described in any one of claims 1-9; The photodetector-based detection system for different operating conditions includes: The twin module is used to mark the detection area of ​​the photodetector. A primary twin is constructed based on the detection object and the corresponding environmental parameters in the detection area. The primary twin is simulated under different working conditions to output a multi-level working condition twin. The detection event module is used to collect multiple spectral data corresponding to the detection object when the photodetector performs online detection on the detection object, input the multiple spectral data into the multi-level operating condition twin, and output the corresponding detection event; The detection mode optimization module is used to determine the corresponding detection matching content based on the identification of the detection event, and to determine the current detection mode and corresponding detection impact content of the photodetector by tracing the detection matching content. Based on the detection impact content and the current operating conditions of the detection area, the mode optimization of the photodetector is triggered to output the final detection mode of the photodetector. The operating condition change module is used to determine multiple detection data based on the photodetector in the final detection mode, determine the current detection result based on the combination of multiple detection data, the corresponding detection object and the corresponding environmental parameters, and trigger the dynamic change of the current detection result along with the update of the operating condition change content.

Images