An adaptive sorting and tracing system and method for fragrance detection samples

Abstract

The application discloses a kind of essence detection sample's adaptive sorting traceability system and method, comprising: sample receiving execution module, for receiving the detection completed sample of essence detection instrument output and executing conveying, sorting and storage action;Data acquisition module is used to collect detection instrument output data, RFID reading data, image acquisition data, position sensing data, conveying state data and storage confirmation data;Fusion verification module is used to carry out identity consistency verification to current sample based on the above data;Predictive control module is used to determine sorting instruction issuing time according to sample position, conveying speed, target storage unit position, communication delay and sorting execution mechanism response delay;Sorting verification module is used to confirm whether sample enters target storage unit;Trusted traceability module is used to generate sample traceability chain containing detection, identification, sorting, storage and exception handling event;Exception closed loop processing module is used to execute exception review processing when verification fails, sorting exception or storage exception.

Description

Technical Field

[0001] This invention relates to the field of fragrance and flavor testing, and specifically to an adaptive sorting and traceability system and method for fragrance testing samples. Background Technology

[0002] Fragrances and flavorings are widely used in food, daily chemicals, tobacco, and pharmaceutical industries. Their color, concentration, and batch consistency typically require testing using specialized instruments. After testing, samples need to be categorized and collected according to the results, and information such as sample number, test result, sorting time, and storage location must be retained for subsequent re-inspection, quality traceability, and anomaly investigation. Therefore, the automated sorting, accurate storage, and reliable traceability of samples after testing are crucial aspects of the fragrance testing process that affect the standardization of sample management.

[0003] Current methods for collecting fragrance test samples mostly rely on manual sorting or use ordinary conveyor mechanisms in conjunction with RFID readers and image acquisition modules for basic information recording. These methods typically only achieve sample transport, storage, or single-identity identification, making it difficult to coordinate and verify data from testing instruments, RFID reads, image acquisition, sample location, and storage confirmation. When batches of samples are continuously collected, issues such as missed RFID reads, mislabeling, sample queue misalignment, fluctuations in conveyor speed, or lag in the sorting mechanism can easily lead to distortion in the correspondence between test results and the actual samples.

[0004] Therefore, the existing technology has at least the following technical problems: in the process of batch continuous sampling, sorting and storage of fragrance testing samples, there is a lack of technical means to jointly verify the consistency of sample identity, the timing of sorting and the actual storage results, which makes it easy for samples to be missorted, missed or not sorted, or the traceability data to be unreliable, making it difficult to meet the needs of automated management and quality traceability of fragrance testing samples. Summary of the Invention

[0005] To achieve the above and other related objectives, this invention discloses an adaptive sorting and traceability system for fragrance testing samples, comprising: The sample receiving module is used to receive the completed samples output by the fragrance testing instrument and perform conveying, sorting and storage actions. The data acquisition module is used to collect data from the testing instrument output, RFID reading data, image acquisition data, position sensing data, transport status data, and storage confirmation data. The fusion verification module is used to verify the identity consistency of the current sample based on the above data. The predictive control module is used to determine the timing of issuing sorting instructions based on sample position, conveying speed, target storage unit position, communication delay, and sorting execution mechanism response delay. The sorting and verification module is used to confirm whether the sample has entered the target storage unit; The trusted traceability module is used to generate a sample traceability chain that includes detection, identification, sorting, storage, and anomaly handling events; The exception handling module is used to perform exception review processing when verification fails, sorting is abnormal, or storage is abnormal.

[0006] Preferably, the sample receiving execution module includes a collection frame, a conveyor track, a sorting execution mechanism, and multiple sorting and storage units; The input end of the conveyor track is connected to the automatic sampling module of the fragrance testing instrument. The sorting execution mechanism is set at the position of the corresponding classification and storage unit on the conveyor track. The classification and storage unit includes at least a qualified storage unit, a slightly different storage unit, a significantly different storage unit, and an abnormal verification storage unit. The data acquisition module includes an RFID reader / writer, an image acquisition module, a positioning sensor, a conveying speed acquisition unit, an actuator feedback unit, a storage inlet sensor, and a storage box capacity sensor.

[0007] Preferably, the fusion verification module outputs the first... After the test result data package for each sample is generated, a digital twin record is created for that sample. This digital twin record includes the sample number, batch number, test result, set of test instrument parameters, sample morphology, sample delivery sequence number, sample transport status, target storage unit, and sample processing status, and is represented as follows: in, Indicates the first A digital twin record of a sample, Indicates the sample number. Indicates the batch number. This indicates the test result. This represents the set of parameters for the testing instrument. Indicates the sample morphology. Indicates the sample output sequence number. Indicates the sample at time [time]. The delivery status, Indicates the target storage unit. This indicates the sample processing status.

[0008] Preferably, the sample delivery status includes sample position, delivery speed, and position confidence level, and is expressed as: in, Indicates the first Each sample at time The delivery location, Indicates the first Each sample at time The conveying speed, This represents the confidence level of the location determined jointly by location sensor data and sampling order prediction; The fusion verification module continuously updates the correspondence between sample number, delivery location, detection judgment result and target storage unit based on the digital twin record.

[0009] Preferably, the fusion verification module converts the detection instrument output data, RFID read data, image acquisition data, and queue time-series data into detection features, RFID features, visual features, and queue features, respectively, forming multi-source fusion input features: in, Indicates the first Multi-source fusion input features of each sample This indicates the characteristics of the output data from the testing instrument. Indicates the characteristics of RFID read data. Indicates visual characteristics, Represents the temporal characteristics of the queue; The visual features include at least the sample container outline features, label area recognition features, sample appearance features, and sample anomaly detection features.

[0010] Preferably, the fusion verification module outputs the identity consistency confidence level through the consistency verification model: in, Indicates the first Confidence level of identity consistency among individual samples. This represents the Sigmoid activation function. Represents a non-linear activation function. This represents the first weight matrix. This represents the second weight matrix. This represents the first bias vector. Indicates the second bias vector; when Not less than the preset consistency threshold When the sample identities are consistent, Less than When this happens, the sample is marked as an abnormal sample and transferred to the abnormal review process.

[0011] Preferably, when the sample identities are consistent, the prediction control module calculates the estimated time for the sample to reach the target sorting position based on the sample's current position and conveying speed. in, Indicates the first The estimated time for each sample to reach the target sorting location. Indicates the current sampling time. This indicates the sorting location corresponding to the target storage unit. Indicates the first The position of each sample at the current sampling time Indicates the first The transport speed of a sample at the current sampling time. Indicates the minimum speed threshold; The timing of issuing sorting instructions is determined based on communication latency, sorting mechanism response latency, and safety compensation time. in, Indicates the first The time when the sorting instruction for each sample is issued. This indicates the communication delay between the control unit and the sorting actuator. This indicates the response time required for the sorting actuator to reach the valid push position from receiving the instruction. This indicates the time for safety compensation.

[0012] Preferably, when the deviation between the actual arrival time and the expected arrival time of the sample exceeds a preset deviation threshold, the predictive control module corrects the conveying speed. in, This indicates the corrected conveyor speed. This indicates the conveying speed before the correction. This represents the speed correction factor. This indicates the actual location determined by location sensor data. Indicates the predicted location. Indicates the time interval between two consecutive samples; If the arrival time deviation of multiple consecutive samples exceeds the preset deviation threshold, the delivery speed will be reduced or the sample output of the fragrance detection instrument will be paused.

[0013] Preferably, after the sorting action is completed, the sorting verification module generates a post-sorting verification result based on the target storage unit inlet sensor signal, the target storage unit capacity change, the sorting execution mechanism action feedback signal, and the target storage area image recognition result. The trusted traceability module breaks down the processing of each sample into detection events, sample release events, RFID identification events, image acquisition events, consistency verification events, sorting instruction events, execution feedback events, storage confirmation events, and anomaly handling events. It then generates the hash value of the current event based on the hash value of the previous traceability event for the same sample and the hash value of the current traceability event. in, Indicates the first The first sample The hash value of a source event, This represents the hash value of a previous tracing event for the same sample. Indicates the current source tracing event, Indicates the sample number. This indicates a concatenation operation. This represents the hash operation function.

[0014] Secondly, the present invention discloses an adaptive sorting and traceability method for fragrance testing samples, applied to the adaptive sorting and traceability system for fragrance testing samples, the method comprising: Receive sample test result data packets output by the fragrance testing instrument; Establish digital twin records for the samples after testing; Collect data from the testing instrument output, RFID reading data, image acquisition data, and queue time sequence data; verify the identity consistency of the current sample; When the identity consistency verification is successful, the sorting instruction issuance time is determined based on the sample location, conveying speed, target storage unit location, communication delay, and sorting execution mechanism response delay. The drive sorting mechanism sorts the samples to the target storage unit; The actual sample collection results were sorted and verified. Generate a sample traceability chain based on detection events, identification events, sorting events, storage confirmation events, and anomaly handling events; If the identity consistency verification fails, the post-sorting verification fails, or the storage is abnormal, the sample will be transferred to the abnormal review process.

[0015] By adopting the above technical solution, this invention establishes a digital twin record for the tested fragrance samples and integrates the output data of the testing instrument, RFID reading data, image acquisition data, and queue timing data for identity consistency verification. At the same time, it determines the time of sorting instruction issuance based on sample position, conveying speed, communication delay, and sorting execution mechanism response delay. After sorting, it combines the inlet sensor signal, capacity change, execution mechanism feedback, and storage area image for actual storage verification. The detection, identification, sorting, storage, and anomaly handling events are written into a trusted traceability chain. This solves the problems of easy mismatch between test results and actual samples, difficulty in confirming sorting results, and insufficient credibility of traceability data in batch fragrance testing, thereby improving the accuracy of sample sorting. Attached Figure Description

[0016] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. The drawings are provided for a better understanding of the invention and are not intended to limit the scope of this disclosure. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein: Figure 1 This is a system block diagram according to an embodiment of the present invention. Detailed Implementation

[0017] 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.

[0018] Reference Figure 1 This invention provides an adaptive sorting and traceability system for fragrance testing samples. It is suitable for use with fragrance testing instruments to automatically transport, verify, classify, sort, store, and trace fragrance samples after testing. The fragrance testing instrument can be a device used to detect differences in fragrance color, concentration, batch consistency, or other quality indicators. After testing, it can output data such as sample number, batch number, testing time, testing results, testing parameters, and sample output order.

[0019] The system of the present invention includes a sample receiving execution module, a data acquisition module, a fusion verification module, a predictive control module, a sorting and verification module, a reliable traceability module, and an anomaly handling module.

[0020] The sample receiving module is used to receive the completed samples output by the fragrance testing instrument and to perform conveying, sorting and storage actions.

[0021] The data acquisition module is used to collect output data from testing instruments, RFID reading data, image acquisition data, location sensing data, transport status data, and storage confirmation data.

[0022] The fusion verification module is used to verify the identity consistency of the current sample based on the data obtained by the data acquisition module.

[0023] The predictive control module is used to calculate the timing of sorting instruction issuance and control the actions of the sorting execution mechanism.

[0024] The sorting and verification module is used to confirm whether the sample has actually entered the target storage unit.

[0025] The Trusted Traceability module is used to generate a sample traceability chain.

[0026] The exception handling module is used to perform exception verification when there are exceptions in sample identification, sorting, or storage.

[0027] Preferably, the sample receiving execution module includes a collection frame, a conveyor track, a sorting execution mechanism, and multiple classification and storage units. The input end of the conveyor track is connected to the automatic sample dispensing module of the fragrance testing instrument to receive fragrance samples after testing. The sorting execution mechanism is located at the position corresponding to the classification and storage unit on the conveyor track. The sorting execution mechanism can be a sorting cylinder, a sorting baffle, a lever mechanism, or a motor-driven pushing mechanism. The multiple classification and storage units include at least a qualified storage unit, a slight difference storage unit, a significant difference storage unit, and an anomaly verification storage unit. The qualified storage unit is used to store samples that are determined to have no difference or are qualified; the slight difference storage unit is used to store samples that are determined to have slight differences; the significant difference storage unit is used to store samples that are determined to have significant differences; and the anomaly verification storage unit is used to store samples that fail identity consistency verification, fail post-sorting verification, have abnormal images, or have abnormal RFID reading.

[0028] Preferably, the data acquisition module includes an RFID reader / writer, an image acquisition module, a positioning sensor, a conveyor speed acquisition unit, an actuator feedback unit, a storage inlet sensor, and a storage box capacity sensor. The RFID reader / writer is located in the identification area of ​​the conveyor track and is used to read the sample number, batch number, and tag writing time from the sample RFID tags. The image acquisition module is located in the identification area or sorting area and is used to acquire images of the sample container, tag area, sample appearance, and storage area. The positioning sensor is used to detect the position of the sample on the conveyor track. The conveyor speed acquisition unit is used to obtain the running speed of the sample or the conveyor track. The actuator feedback unit is used to provide feedback on whether the sorting actuator has completed its action. The storage inlet sensor is used to detect whether the sample has entered the corresponding storage unit. The storage box capacity sensor is used to detect the quantity or capacity of samples in each category of storage unit.

[0029] Preferably, the fragrance testing instrument completes the first step. After each sample is tested, a test result data packet is sent to the system. This data packet includes at least the sample number, batch number, test result, set of instrument parameters, sample morphology, and sample release sequence number. Upon receiving this test result data packet, the fusion verification module creates a digital twin record for that sample. in, Indicates the first A digital twin record of a sample, Indicates the sample number, Indicates the batch number. This indicates the test result. This represents the set of parameters for the testing instrument. Indicates the sample morphology. Indicates the sample output sequence number. Indicates the sample at time [time]. The delivery status, Indicates the target storage unit. This indicates the sample processing status.

[0030] Preferred sample morphology Samples can be in liquid or solid form. Liquid samples can be placed in cuvettes, sample vials, or other testing containers, while solid samples can be placed in testing trays or sample boxes. The test results will be used to determine the outcome. This can include classification results such as no difference, slight difference, significant difference, or other results output by the fragrance testing instrument. The system determines the classification based on the test results. Preliminary identification of target storage units and the sample processing status Set to pending recognition status.

[0031] Sample delivery status This includes sample location, transport speed, and location confidence level. in, Indicates the first Each sample at time The delivery location, Indicates the first Each sample at time The conveying speed This represents the confidence level of the location, determined jointly by location sensor data and sample order prediction. Location Confidence Level This is used to characterize the consistency between the actual delivery position of the current sample and the predicted position of its delivery sequence. When the sample sequence, sensor trigger time, and delivery speed are all consistent with expectations, the position confidence is high; when there is sample delay, abnormal sample spacing, or missing sensor trigger, the position confidence is low.

[0032] Once the sample enters the identification area, the data acquisition module simultaneously acquires the output data of the testing instrument, RFID reading data, image acquisition data, and queue timing data. The testing instrument output data includes the sample number, batch number, test result, test time, and sample release sequence number. The RFID reading data includes the sample number, batch number, tag writing time, and tag reading status from the RFID tag. The image acquisition data includes images of the sample container, the tag area, the sample contents, and the sample's appearance. The queue timing data includes the estimated arrival time, actual trigger time, distance between adjacent samples, and changes in transport speed.

[0033] The fusion verification module converts the output data of the detection instrument, RFID reading data, image acquisition data, and queue time-series data into detection features, RFID features, visual features, and queue features, respectively, and forms multi-source fusion input features: in, Indicates the first Multi-source fusion input features of each sample This indicates the characteristics of the output data from the testing instrument. Indicates the characteristics of RFID read data. Indicates visual characteristics, This represents the temporal characteristics of the queue. The symbol ";" indicates feature concatenation.

[0034] Visual features At least the following should be included: sample container outline features, label area recognition features, sample appearance features, and sample anomaly detection features: in, Indicates the first The profile features of the sample container for each sample. Indicates the first Label region identification features of each sample Indicates the first The appearance characteristics of each sample Indicates the first The abnormal detection features corresponding to sample damage, tilting, leakage, blurred label, or tray misalignment.

[0035] Visual features This information can be obtained through traditional image processing methods or through a lightweight convolutional neural network. Using traditional image processing methods, the sample image is converted to grayscale, edge detection is performed, color space conversion is performed, and template matching is done to obtain the sample container outline, label position, sample color state, and container tilt state. Using a lightweight convolutional neural network, historical sample images are used as training data. Training labels include normal samples, samples with blurred labels, tilted containers, leaking samples, tray-off samples, and broken containers. The model outputs the sample visual state category and its corresponding confidence score.

[0036] The fusion verification module outputs the identity consistency confidence level through the consistency verification model: in, Indicates the first Confidence level of identity consistency among individual samples. This represents the Sigmoid activation function. Represents a non-linear activation function. This represents the first weight matrix. This represents the second weight matrix. This represents the first bias vector. This represents the second bias vector. Indicates the first Multi-source fusion input features of each sample.

[0037] when Not less than the preset consistency threshold When the fusion verification module determines that the current sample identity is consistent, the system allows the sample to enter the predictive sorting control process. Less than the preset consistency threshold When the fusion verification module determines that the current sample has an identity conflict or incomplete data, the system marks the sample as an identity anomaly sample, and the anomaly handling module transfers it to anomaly review processing. A preset consistency threshold is set. Settings can be configured based on the batch size, sample type, or historical false positive rate.

[0038] When the sample identities are consistent, the predictive control module determines the outcome based on the detection results. Determine the target storage unit When the test results are determined... When there is no difference or the target storage unit is qualified. The storage unit is deemed qualified; when the test results are... When the difference is slight, the target storage unit This is a storage unit with slight differences; when the test results are... When there is a significant difference, the target storage unit The target storage unit is the one with significant differences; when identity consistency verification fails, the target storage unit... This is an abnormality review and storage unit.

[0039] The predictive control module calculates the estimated time for the sample to reach the target sorting position based on the sample's current position and conveyor speed. in, Indicates the first The estimated time for each sample to reach the target sorting location. Indicates the current sampling time. Indicates the target storage unit The corresponding sorting location Indicates the first Each sample at the current sampling time Location, Indicates the first Each sample at the current sampling time The conveying speed, This represents the minimum speed threshold.

[0040] The predictive control module further determines the timing of issuing sorting instructions based on communication latency, sorting mechanism response latency, and safety compensation time. in, Indicates the first The time when the sorting instruction for each sample is issued. Indicates the first The estimated time for each sample to reach the target sorting location. This indicates the communication delay between the control unit and the sorting actuator. This indicates the response time required for the sorting actuator to reach the valid push position from receiving the instruction. This indicates the time for safety compensation.

[0041] At any moment The system sends action commands to the sorting actuators, ensuring that the sorting baffles, cylinders, or levers are in an effective pushing state when the sample reaches the target sorting position. This pre-compensation method reduces sorting deviations caused by communication delays, actuator response delays, and fluctuations in conveyor speed.

[0042] When the deviation between the actual arrival time and the expected arrival time of the sample exceeds a preset deviation threshold, the predictive control module corrects the conveying speed. in, This indicates the corrected conveyor speed. This indicates the conveying speed before the correction. This represents the speed correction factor. This indicates the actual location determined by location sensor data. Indicates the predicted location. This indicates the time interval between two consecutive samples. If the arrival time deviation of multiple consecutive samples exceeds the preset deviation threshold, the predictive control module will reduce the delivery speed or pause the sample output of the fragrance detection instrument to avoid continuous misclassification of subsequent samples.

[0043] After the sorting process is completed, the sorting verification module generates the post-sorting verification result based on the target storage unit inlet sensor signal, the target storage unit capacity change, the sorting actuator action feedback signal, and the target storage area image recognition result: in, Indicates the first The results of sorting and verifying each sample. This indicates the inlet sensor signal of the target storage unit. This indicates the change in the capacity of the target storage unit. This indicates the feedback signal of the sorting execution mechanism. This indicates the image recognition result of the target storage area. This represents the verification function after sorting.

[0044] When the target storage unit inlet sensor is triggered, the target storage unit capacity increases, the sorting execution mechanism completes its feedback action, and the target storage area image recognizes the newly added sample, the post-sorting verification function... Output This indicates that the sample has entered the target storage unit. If any of the above key conditions is not met, the post-sorting verification function will be activated. Output This indicates an abnormality in sample sorting. The system generates a sorting abnormality event and transfers it to the abnormality review process.

[0045] The trusted traceability module breaks down the processing of each sample into multiple traceability events. These events include at least detection events, sample release events, RFID identification events, image acquisition events, consistency verification events, sorting instruction events, execution feedback events, storage confirmation events, and anomaly handling events. The first sample Each source tracing event is represented as: in, Indicates the first The first sample A source tracing event, Indicates the time when the event occurred. Indicates the event type, Represents event data, This indicates the device unit or operator that performed the event.

[0046] The trusted tracing module generates the current event hash value based on the hash value of the previous tracing event for the same sample and the current tracing event: in, Indicates the first The first sample The hash value of a source event, This represents the hash value of a previous tracing event for the same sample. Indicates the current source tracing event, Indicates the sample number. This indicates a concatenation operation. This represents the hash function. The hash function can be SHA-256 or other hash functions that can generate a fixed-length digest value.

[0047] Using the event hash generation method described above, the detection, identification, sorting, storage, and anomaly handling events of the same sample form a sequential traceability chain. When any historical event data is modified, the hash value of that event and its subsequent events changes, thereby revealing whether the traceability data has been tampered with or the traceability event chain has been broken.

[0048] After batch sample testing is completed, the trusted traceability module can calculate the batch trusted traceability completeness rate based on the completeness of required fields, the continuity of the traceability event chain, and the post-sorting verification results. in, Indicates the completeness of batch traceability. Indicates the total number of samples in the batch. Indicates the first Are all the required fields for each sample complete? Indicates the first Is the traceability event chain of a single sample continuous? Indicates the first The results of sorting and verifying each sample. This indicates an indicator function that takes the value 1 when all conditions within the parentheses are met, and 0 otherwise.

[0049] The anomaly handling module performs different actions based on different anomaly types. When RFID reading fails but the consistency confidence level between the instrument output data, image acquisition data, and queue time-series data is not less than the supplementary recording threshold, the system allows the sample to enter the target storage unit and generates an RFID reading failure event for subsequent RFID information supplementation. When the RFID reading result is inconsistent with the sample number output by the instrument, the system marks the sample as an identity conflict sample and sorts it to the anomaly verification storage unit. When image acquisition data indicates that the sample has container damage, liquid leakage, blurred labels, tray misalignment, or sample tilting, the system transfers the sample to anomaly verification processing and generates an image anomaly event.

[0050] When the capacity of the target storage unit reaches the warning threshold, the system stops sorting samples into that unit. If a backup storage unit is provided, the system switches samples with the same test results to the backup unit and writes the switching event to the traceability chain. If no backup storage unit is provided, the system pauses sample output from the fragrance testing instrument and triggers an alarm. The results are then verified after sorting. When an anomaly occurs, the system generates a sorting anomaly event, records the target storage unit, the feedback from the actuator, the sensor signal at the storage entrance, the image of the storage area, and the time of the anomaly, and prompts the operator to review it.

[0051] The workflow of this invention is as follows: After the fragrance testing instrument completes the sample testing, it sends the sample testing result data packet to the system; the fusion verification module establishes a digital twin record for the sample and determines the target storage unit based on the testing results; after the sample enters the conveying track, the data acquisition module continuously collects position sensor data and conveying status data; when the sample reaches the identification area, the RFID reader reads the tag information, and the image acquisition module acquires the sample image; the fusion verification module performs identity consistency verification on the output data of the testing instrument, RFID reading data, image acquisition data, and queue timing data; after the identity consistency verification is passed, the prediction control module calculates the expected time for the sample to reach the target sorting position and the time when the sorting instruction is issued; the sorting execution mechanism sorts the sample to the target storage unit according to the control instruction; the sorting verification module confirms whether the sample has actually entered the target storage unit; the reliable traceability module writes the detection, identification, sorting, storage, and abnormal handling events into the sample traceability chain; when an abnormality occurs, the abnormal handling module performs abnormal review, alarm, suspension of conveying, reduction of conveying speed, or switching to a backup storage unit.

[0052] In liquid fragrance sample testing scenarios, samples can be placed in cuvettes or sample vials. The system uses an image acquisition module to identify the cuvette outline, liquid state, and label area, reads the sample number using an RFID reader, and confirms the correspondence between the current sample and the output result of the testing instrument using queue time-series data. After the identity consistency verification is passed, the system sorts the sample into a qualified storage unit, a slightly different storage unit, or a significantly different storage unit based on the test judgment result. If the image acquisition module detects a tilted cuvette, liquid leakage, or a blurred label, the system sorts the sample into an anomaly verification storage unit and writes an image anomaly event into the traceability chain.

[0053] In the scenario of solid fragrance sample testing, the sample can be placed in a testing tray. The system uses an image acquisition module to identify the tray outline, whether the sample has spilled, whether the tray has shifted, and the status of the label area. When the tray position deviates too much from the center line of the conveyor track, the fusion verification module lowers the identity consistency confidence level; if the identity consistency confidence level is lower than a preset consistency threshold, the system transfers the sample to anomaly review processing. After the identity consistency verification is passed, the predictive control module calculates the sorting instruction issuance time based on the current tray position, conveyor speed, and target storage unit position, and controls the sorting execution mechanism to complete the sorting.

[0054] Through the above structure and process, this invention enables closed-loop control of the transportation, identification, sorting, storage and traceability of fragrance test samples after testing. It can improve the reliability of sample identity binding, the accuracy of sorting actions, the verifiability of actual storage and the credibility of traceability data under continuous batch testing conditions.

[0055] Secondly, the present invention discloses an adaptive sorting and traceability method for fragrance testing samples, applied to the adaptive sorting and traceability system for fragrance testing samples, the method comprising: Receive sample test result data packets output by the fragrance testing instrument; Establish digital twin records for the samples after testing; Collect data from the testing instrument output, RFID reading data, image acquisition data, and queue time sequence data; verify the identity consistency of the current sample; When the identity consistency verification is successful, the sorting instruction issuance time is determined based on the sample location, conveying speed, target storage unit location, communication delay, and sorting execution mechanism response delay. The drive sorting mechanism sorts the samples to the target storage unit; The actual sample collection results were sorted and verified. Generate a sample traceability chain based on detection events, identification events, sorting events, storage confirmation events, and anomaly handling events; If the identity consistency verification fails, the post-sorting verification fails, or the storage is abnormal, the sample will be transferred to the abnormal review process.

[0056] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the meaning consistent with their meaning in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined.

[0057] For the sake of simplicity, the method embodiments are described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0058] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0059] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An adaptive sorting and traceability system for fragrance testing samples, characterized in that, include: The sample receiving module is used to receive the completed samples output by the fragrance testing instrument and perform conveying, sorting and storage actions. The data acquisition module is used to collect data from the testing instrument output, RFID reading data, image acquisition data, position sensing data, transport status data, and storage confirmation data. The fusion verification module is used to verify the identity consistency of the current sample based on the above data. The predictive control module is used to determine the timing of sorting instruction issuance for samples that have passed the identity consistency verification, based on sample location, conveying speed, target storage unit location, communication delay, and sorting execution mechanism response delay. The sorting and verification module is used to confirm whether the sample has entered the target storage unit; The trusted traceability module is used to generate a sample traceability chain that includes detection, identification, sorting, storage, and anomaly handling events; The exception handling module is used to perform exception review processing when verification fails, sorting is abnormal, or storage is abnormal.

2. The adaptive sorting and traceability system for fragrance testing samples according to claim 1, characterized in that, The sample receiving execution module includes a collection frame, a conveyor track, a sorting execution mechanism, and multiple classification and storage units; The input end of the conveyor track is connected to the automatic sampling module of the fragrance testing instrument. The sorting execution mechanism is set at the position of the corresponding classification and storage unit on the conveyor track. The classification and storage unit includes at least a qualified storage unit, a slightly different storage unit, a significantly different storage unit, and an abnormal verification storage unit. The data acquisition module includes an RFID reader / writer, an image acquisition module, a positioning sensor, a conveying speed acquisition unit, an actuator feedback unit, a storage inlet sensor, and a storage box capacity sensor.

3. The adaptive sorting and traceability system for fragrance testing samples according to claim 1, characterized in that, The fusion verification module outputs the first... After the test result data package for each sample is generated, a digital twin record is created for that sample. This digital twin record includes the sample number, batch number, test result, set of test instrument parameters, sample morphology, sample delivery sequence number, sample transport status, target storage unit, and sample processing status, and is represented as follows: in, Indicates the first A digital twin record of a sample, Indicates the sample number. Indicates the batch number. This indicates the test result. This represents the set of parameters for the testing instrument. Indicates the sample morphology. Indicates the sample output sequence number. Indicates the sample at time [time]. The delivery status, Indicates the target storage unit. This indicates the sample processing status.

4. The adaptive sorting and traceability system for fragrance testing samples according to claim 3, characterized in that, The sample delivery status includes sample position, delivery speed, and position confidence level, and is expressed as: in, Indicates the first Each sample at time The delivery location, Indicates the first Each sample at time The conveying speed, This represents the confidence level of the location determined jointly by location sensor data and sampling order prediction; The fusion verification module continuously updates the correspondence between sample number, delivery location, detection judgment result and target storage unit based on the digital twin record.

5. The adaptive sorting and traceability system for fragrance testing samples according to claim 3, characterized in that, The fusion verification module converts the detection instrument output data, RFID read data, image acquisition data, and queue time-series data into detection features, RFID features, visual features, and queue features, respectively, forming multi-source fusion input features: in, Indicates the first Multi-source fusion input features of each sample This indicates the characteristics of the output data from the testing instrument. Indicates the characteristics of RFID read data. Indicates visual characteristics, Represents the temporal characteristics of the queue; The visual features include at least the sample container outline features, label area recognition features, sample appearance features, and sample anomaly detection features.

6. The adaptive sorting and traceability system for fragrance testing samples according to claim 5, characterized in that, The fusion verification module outputs the identity consistency confidence level through the consistency verification model: in, Indicates the first Confidence level of identity consistency among individual samples. This represents the Sigmoid activation function. Represents a non-linear activation function. This represents the first weight matrix. This represents the second weight matrix. This represents the first bias vector. Indicates the second bias vector; when Not less than the preset consistency threshold When the sample identities are consistent, Less than When this happens, the sample is marked as an abnormal sample and transferred to the abnormal review process.

7. The adaptive sorting and traceability system for fragrance testing samples according to claim 6, characterized in that, When the sample identities match, the predictive control module calculates the estimated time for the sample to reach the target sorting position based on the sample's current position and conveying speed. in, Indicates the first The estimated time for each sample to reach the target sorting location. Indicates the current sampling time. This indicates the sorting location corresponding to the target storage unit. Indicates the first The position of each sample at the current sampling time Indicates the first The transport speed of a sample at the current sampling time. Indicates the minimum speed threshold; The timing of issuing sorting instructions is determined based on communication latency, sorting mechanism response latency, and safety compensation time. in, Indicates the first The time when the sorting instruction for each sample is issued. This indicates the communication delay between the control unit and the sorting actuator. This indicates the response time required for the sorting actuator to reach the valid push position from receiving the instruction. This indicates the time for safety compensation.

8. The adaptive sorting and traceability system for fragrance testing samples according to claim 7, characterized in that, When the deviation between the actual arrival time and the expected arrival time of the sample exceeds a preset deviation threshold, the predictive control module corrects the conveying speed. in, This indicates the corrected conveyor speed. This indicates the conveying speed before the correction. This represents the speed correction factor. This indicates the actual location determined by location sensor data. Indicates the predicted location. Indicates the time interval between two consecutive samples; If the arrival time deviation of multiple consecutive samples exceeds the preset deviation threshold, the delivery speed will be reduced or the sample output of the fragrance detection instrument will be paused.

9. The adaptive sorting and traceability system for fragrance testing samples according to claim 1, characterized in that, After the sorting action is completed, the sorting verification module generates the post-sorting verification result based on the target storage unit inlet sensor signal, the target storage unit capacity change, the sorting execution mechanism action feedback signal, and the target storage area image recognition result. The trusted traceability module breaks down the processing of each sample into detection events, sample release events, RFID identification events, image acquisition events, consistency verification events, sorting instruction events, execution feedback events, storage confirmation events, and anomaly handling events. It then generates the hash value of the current event based on the hash value of the previous traceability event for the same sample and the hash value of the current traceability event. in, Indicates the first The first sample The hash value of a source event, This represents the hash value of a previous tracing event for the same sample. Indicates the current source tracing event, Indicates the sample number. This indicates a concatenation operation. This represents the hash operation function.

10. An adaptive sorting and traceability method for fragrance testing samples, characterized in that, The method, applied to the adaptive sorting and traceability system for fragrance testing samples according to any one of claims 1 to 9, comprises: Receive sample test result data packets output by the fragrance testing instrument; Establish digital twin records for the samples after testing; Collect data from the testing instrument output, RFID reading data, image acquisition data, and queue time sequence data; verify the identity consistency of the current sample; When the identity consistency verification is successful, the sorting instruction issuance time is determined based on the sample location, conveying speed, target storage unit location, communication delay, and sorting execution mechanism response delay. The drive sorting mechanism sorts the samples to the target storage unit; The actual sample collection results were sorted and verified. Generate a sample traceability chain based on detection events, identification events, sorting events, storage confirmation events, and anomaly handling events; If the identity consistency verification fails, the post-sorting verification fails, or the storage is abnormal, the sample will be transferred to the abnormal review process.

Images