Two-dimensional code jet printing quality control method and system

Abstract

The invention discloses a two-dimensional code jet printing quality control method and system, and relates to the technical field of industrial automation, and the method comprises the steps: obtaining ink batch information, paper model information and environment temperature and humidity data based on a production work order, comparing the information with pre-stored standard production conditions, and inputting the information to a predictive compensation model to output a predictive adjustment amount; setting nozzle voltage, nozzle temperature and trimming parameters of the code spraying equipment according to the predictive adjustment amount; performing two-dimensional code jet printing based on the set nozzle voltage, nozzle temperature and trimming parameters, performing appearance quality detection on the jet-printed two-dimensional code by using an online visual detection module to generate an initial detection result, and regularly extracting samples to perform offline two-dimensional code grade detection; and comparing the initial detection result with a preset qualified threshold, and combining the offline two-dimensional code grade data. According to the invention, qualitative change from passive error correction to active prevention is realized, and the quality stability, the production efficiency and the intelligent level of the two-dimensional code jet printing process are improved.

Description

Technical Field

[0001] This invention relates to the field of industrial automation technology, and in particular to a method and system for quality control of QR code printing. Background Technology

[0002] In the industrial printing field, the quality of QR code printing is of paramount importance. The current mainstream quality control technology is mainly based on the online appearance inspection-real-time feedback adjustment mode. The vision system detects appearance defects such as registration deviation, missing print, duplicate codes, black lines, ink splatter, and orientation of QR codes. After defects are found, the defective products are rejected and the operators adjust parameters such as printhead voltage and temperature based on their experience to ensure appearance consistency.

[0003] The readability of QR codes is affected by a combination of factors, including nozzle voltage, temperature, ambient temperature and humidity, nozzle height, waveform file, QR code trimming, nozzle plate overlap effect, and resolution. Existing online inspection systems can only identify appearance defects and cannot directly measure symbol levels online, resulting in blind spots in the monitoring of core quality indicators. Current adjustments are passive responses and cannot provide early warning and adaptive compensation for quality drift caused by equipment aging and material batch changes, thus restricting production stability and intelligent improvement. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a QR code printing quality control method that solves the problem that existing technologies cannot achieve forward-looking parameter optimization and self-healing of potential quality risks due to adjustment lag and reliance on human experience.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a QR code printing quality control method, which includes: acquiring ink batch information, paper type information and environmental temperature and humidity data based on a production work order, comparing them with pre-stored standard production conditions, and inputting them into a predictive compensation model to output predictive adjustment amounts.

[0008] The printhead voltage, printhead temperature, and trimming parameters of the inkjet printer are set according to the predictive adjustment amount.

[0009] QR code printing is performed based on the set printhead voltage, printhead temperature and trimming parameters. The appearance quality of the printed QR code is inspected using an online visual inspection module to generate initial inspection results. Samples are periodically sampled for offline QR code level inspection.

[0010] The initial test results are compared with the preset qualified threshold, and the quality judgment is obtained by combining the offline QR code level data. The time series data of environmental temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results and offline QR code level data are recorded.

[0011] The time series data is analyzed in real time, and the evolution patterns of defect types are matched with pre-stored defect evolution maps to generate self-healing instructions.

[0012] The self-healing command is executed to adjust the printhead voltage of the inkjet printer, and the predictive adjustment amount is returned to set the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer.

[0013] As a preferred embodiment of the QR code printing quality control method of the present invention, the method includes the following steps: based on the production work order, obtaining ink batch information, paper type information, and environmental temperature and humidity data, comparing them with pre-stored standard production conditions, and inputting them into a predictive compensation model to output predictive adjustment amounts.

[0014] Analyze the production work order to obtain the product brand and production quantity in the production work order. Based on the product brand, query and obtain the ink batch information and paper model information.

[0015] By reading the current ambient temperature and humidity data through sensors, and by using the ink batch information, paper type information, ambient temperature and humidity data, printhead voltage and printhead temperature parameters corresponding to the high-quality QR code printing results in historical production work orders, standard production conditions are established.

[0016] The ink batch information, paper type information, and ambient temperature and humidity data are compared with the pre-stored standard production conditions. The compared ink batch information, paper type information, and ambient temperature and humidity data are then input into the predictive compensation model, which outputs predictive adjustment amounts for key process parameters of the coding equipment.

[0017] As a preferred embodiment of the QR code printing quality control method of the present invention, the method includes the following steps: setting the printhead voltage, printhead temperature, and trimming parameters of the coding device according to the predictive adjustment amount:

[0018] Read the predictive adjustment amount, and calculate the result by combining the predictive adjustment amount with the current set values ​​of the printhead voltage, printhead temperature and trimming parameters of the inkjet printer.

[0019] The expression for setting the value is:

[0020]

[0021] in, This is the nozzle voltage setting value. For nozzle temperature setpoint Set values ​​for trimming parameters. This refers to the actual current setting value of the printhead voltage of the inkjet printer. This is the actual current set value of the printhead temperature of the inkjet printer. This is the actual current setting value of the trimming parameters. This is a predictive adjustment amount for the nozzle voltage. This is a predictive adjustment amount for nozzle temperature. This is a predictive adjustment amount for the trimming parameters;

[0022] Based on the calculation results, the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer are set to new values. The trimming parameters are represented as the percentage of black bars in the inkjet printing software.

[0023] As a preferred embodiment of the QR code printing quality control method of the present invention, the method includes: printing QR codes based on set printhead voltage, printhead temperature, and trimming parameters; using an online visual inspection module to perform appearance quality inspection on the printed QR codes, generating initial inspection results; and periodically sampling samples for offline QR code level inspection, comprising the following steps:

[0024] Using the set printhead voltage, printhead temperature, and trimming parameters, a QR code is printed on the paper, and the online visual inspection module captures the image of the printed QR code.

[0025] The online visual inspection module analyzes the collected images of the printed QR codes and generates initial inspection results, including appearance quality scores and defect codes, based on the appearance quality inspection results.

[0026] Samples are periodically taken from products that have passed the online visual inspection module, and offline QR code level detection is performed on the sampled products using an offline QR code detector to obtain offline QR code level data.

[0027] As a preferred embodiment of the QR code printing quality control method of the present invention, the following steps are included: comparing the initial detection result with a preset qualified threshold, combining it with offline QR code level data to obtain a quality judgment, and recording the time series data of environmental temperature and humidity data, printhead voltage, printhead temperature, trimming parameters, initial detection results, and offline QR code level data:

[0028] The appearance quality score in the initial test results is compared with the preset pass threshold, and the offline QR code level data is compared with the preset level threshold. Based on the comparison results of the initial test results and the comparison results of the offline QR code level data, the quality judgment result is obtained by combining the results.

[0029] Associate ambient temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results, and offline QR code level data with the current timestamp;

[0030] The environmental temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results, and offline QR code level data associated with the timestamp are stored as records in the time series data.

[0031] As a preferred embodiment of the QR code printing quality control method of the present invention, the method includes the following steps: real-time analysis of the time series data, matching the evolution pattern of defect types with a pre-stored defect evolution map, and generating a self-healing instruction:

[0032] Continuously monitor the numerical changes of defect codes and offline QR code level data recorded in the initial detection results in the time series data stream to identify the evolution pattern of defect types in the time series data;

[0033] By analyzing the correlation between nozzle voltage, nozzle temperature, ambient temperature and humidity data, trimming parameters, initial detection results, and offline QR code level data in historical time series data, a defect evolution map was established after summarizing the evolution pattern of defect types with parameter drift.

[0034] As a preferred embodiment of the QR code printing quality control method of the present invention, the method includes the following steps: executing the self-healing command to adjust the printhead voltage of the coding device, and returning the predictive adjustment amount to set the printhead voltage, printhead temperature, and trimming parameters of the coding device.

[0035] Parse the self-healing command to obtain the adjustment amount of the printhead voltage of the inkjet printer;

[0036] The current setting of the printhead voltage of the inkjet printer is modified according to the adjustment amount of the self-healing command. The process then jumps to set the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer according to the predictive adjustment amount.

[0037] Secondly, the present invention provides a QR code printing quality control system, including a predictive compensation module, which acquires ink batch information, paper type information and environmental temperature and humidity data based on production work orders, compares them with pre-stored standard production conditions, and inputs them into a predictive compensation model to output predictive adjustment amounts.

[0038] The parameter control module sets the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer based on predictive adjustment amounts.

[0039] The quality judgment module prints QR codes based on the set printhead voltage, printhead temperature and trimming parameters. It uses an online visual inspection module to inspect the appearance quality of the printed QR codes, generates initial inspection results, and periodically extracts samples for offline QR code level inspection.

[0040] The intelligent diagnostic module compares the initial test results with the preset qualified threshold, combines the offline QR code level data to obtain the quality judgment, and records the time series data of environmental temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results and offline QR code level data.

[0041] The self-healing decision module performs real-time analysis on the time series data, matches the evolution patterns of defect types with pre-stored defect evolution maps, and generates self-healing instructions.

[0042] The dynamic feedback module executes the self-healing command to adjust the printhead voltage of the inkjet printer and returns the predictive adjustment amount to set the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer.

[0043] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the QR code printing quality control method as described in the first aspect of the present invention.

[0044] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the QR code printing quality control method as described in the first aspect of the present invention.

[0045] The beneficial effects of this invention are as follows: By integrating production work order information, material data, and environmental parameters, and utilizing a predictive compensation model, the printing process parameters are pre-adjusted in a forward-looking and intelligent manner, thus avoiding quality risks at the beginning of production. Real-time quality data streams are generated through online visual inspection, and time series analysis technology is used to dynamically diagnose defect evolution patterns. When a potential quality degradation trend is identified, precise self-healing instructions are automatically triggered and executed to adjust equipment parameters. By feeding back the self-healing control results to the parameter setting stage, a closed-loop intelligent control loop that can adaptively cope with fluctuations in production conditions and degradation of equipment performance is constructed, achieving a qualitative change from passive error correction to proactive prevention, and improving the quality stability, production efficiency, and intelligence level of the QR code printing process. Attached Figure Description

[0046] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a flowchart of a QR code printing quality control method.

[0048] Figure 2 This is a schematic diagram of a QR code printing quality control system.

[0049] Figure 3 Flowchart for setting up multi-parameter collaboration. Detailed Implementation

[0050] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0051] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0052] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0053] Reference Figures 1-3 As one embodiment of the present invention, this embodiment provides a QR code printing quality control method, including the following steps:

[0054] S1. Based on the production work order, obtain ink batch information, paper type information and environmental temperature and humidity data, compare them with the pre-stored standard production conditions, and input them into the predictive compensation model to output predictive adjustment amounts.

[0055] S1.1 Parse the production work order to obtain the product brand and production quantity in the production work order. Based on the product brand, query and obtain the ink batch information and paper model information.

[0056] Furthermore, the purpose of analyzing production work orders is to accurately locate the specific material base required for this production task. The product brand clearly identified in the production work order is a key index linking all production elements. By querying and obtaining ink batch information and paper type information based on the product brand, the accuracy of material traceability in the production preparation stage is ensured. Ink batch information is associated with the physicochemical properties of a specific formula, such as viscosity and conductivity, while paper type information defines the surface properties of the substrate, such as porosity and smoothness. Obtaining accurate ink batch information and paper type information provides indispensable material attribute input for subsequent adaptive process adjustments, thereby achieving precise matching between production conditions and material status, and laying an accurate material data foundation for quality control.

[0057] It should be noted that, in addition to obtaining ink batch information and paper type information, it is also necessary to expand the scope of production factor data collection to include printhead height setting value, currently used waveform file identifier, QR code trimming parameter configuration, printhead overlap area calibration parameters, and resolution setting parameters. These parameters, together with printhead voltage, printhead temperature, and ambient temperature and humidity data, constitute a set of key process parameters that affect the quality of QR code printing.

[0058] S1.2. Read the current ambient temperature and humidity data through the sensor, and establish standard production conditions by using the ink batch information, paper type information, ambient temperature and humidity data, and printhead voltage and temperature parameters corresponding to the high-quality QR code printing results in the historical production work orders.

[0059] Furthermore, establishing pre-stored standard production conditions constitutes the benchmark reference system for the entire predictive compensation. This is accomplished by collecting and analyzing production data from historical production work orders that ultimately correspond to high-quality QR code printing results. The ink batch information, paper type information, environmental temperature and humidity data, and the printhead voltage and temperature parameters used at the time corresponding to the high-quality results are screened out and integrated. Through statistical analysis or feature extraction of successful case datasets, the optimized parameter ranges and condition combinations that can stably produce high-quality QR codes for different product brands and under different material and environmental combinations are summarized, thereby forming a set of pre-stored standard production conditions. This transforms discrete expert experience or historical best practices into a quantifiable and reusable digital knowledge base.

[0060] S1.3. Compare the ink batch information, paper type information, and ambient temperature and humidity data with the pre-stored standard production conditions. Input the compared ink batch information, paper type information, and ambient temperature and humidity data into the predictive compensation model and output the predictive adjustment amount for the key process parameters of the coding equipment.

[0061] Furthermore, the real-time acquired ink batch information, paper type information, and environmental temperature and humidity data are compared with pre-stored standard production conditions to identify the differences between the current production situation and the known optimal production situation. Multiple influencing factors are comprehensively evaluated to quantify the degree of difference. The compared ink batch information, paper type information, and environmental temperature and humidity data are used as feature vectors representing the current production conditions and input into the predictive compensation model. This predictive compensation model, a mapping function trained on historical data, learns the complex nonlinear relationship between specific production condition characteristics and optimal process parameter adjustments. Based on the currently input feature vector, the predictive compensation model can deduce the predictive adjustment amounts of printhead voltage and printhead temperature required to compensate for the differences between the current and standard conditions.

[0062] Specifically, intelligent decision-making, from perceiving the current situation to generating predictive operational instructions, transforms parameter settings that rely on manual judgment into precise pre-adjustments automatically generated by data-driven models, thereby proactively offsetting potential quality fluctuations before production begins.

[0063] This was accomplished by collecting a large amount of historical production data, including information on different ink batches, paper types, ambient temperature and humidity, and corresponding printhead voltage and temperature parameters. The data was also correlated with the QR code quality detection results generated under these parameter combinations. The data was cleaned and feature-engineered to identify key features related to QR code quality. Machine learning algorithms, such as gradient boosting decision trees or neural networks, were used to train the processed data, enabling the predictive compensation model to learn the complex nonlinear mapping relationship between given ink batch information, paper type information, and ambient temperature and humidity data and the optimal printhead voltage and temperature adjustment. The model's prediction accuracy was evaluated through cross-validation and a test set. The predictive compensation model that met the performance criteria was deployed for online inference, thus obtaining the predictive compensation model.

[0064] It should be noted that the input feature vector of the predictive compensation model needs to be expanded accordingly. The compared data should include not only ink batch information, paper type information and environmental temperature and humidity data, but also process parameters. The pre-stored standard production conditions also need to be expanded to cover the historical best values ​​or effective ranges of these parameters. The trained predictive compensation model should be able to learn and output predictive adjustment amounts for this complete set of process parameters.

[0065] S2. Set the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer based on the predictive adjustment amount.

[0066] S2.1 Read the predictive adjustment amount, and calculate the result by combining the predictive adjustment amount with the current set values ​​of the printhead voltage, printhead temperature and trimming parameters of the inkjet printer.

[0067] Furthermore, the predictive adjustment values ​​output by the predictive compensation model include not only adjustments for printhead voltage and temperature, but also adjustments for trimming parameters. These predictive adjustments are then used in conjunction with the actual current settings of the printhead voltage, printhead temperature, and trimming parameters (represented as the percentage of black bars in the coding software) to obtain optimized results. This achieves a smooth transition and coordinated optimization of multiple key process parameter settings. The core of this is accomplished through a specific, extended setpoint calculation expression, defined as the FCC_Multi function. Its input parameters, in addition to the original ones, include the actual current setting of the trimming parameters and their predictive adjustment values. The internal logic of the FCC_Multi function is to comprehensively consider the complex coupling relationship between printhead voltage, printhead temperature, and trimming parameters (for example, changes in trimming parameters may affect the ink droplet adhesion state, thus interacting with printhead temperature), as well as the nonlinear response characteristics of the current operating state of the equipment to each adjustment value. Through this comprehensive calculation, the final calculation results—the new nozzle voltage setting, nozzle temperature setting, and trimming parameter setting—not only fully integrate the predictive compensation model's compensation intentions for each parameter, but also maintain the synergy between parameter combinations and the overall stability of the setting process. This effectively avoids conflicts between parameters or drastic changes in setting values, ultimately resulting in a globally optimized and robust calculation result.

[0068] Specifically, the extended FCC_Multi function handles the coupling and nonlinearity between multiple parameters by establishing a fuzzy rule base for printhead voltage, printhead temperature, and trimming parameter adjustment amounts. It transforms the actual current settings of the printhead voltage, printhead temperature, and trimming parameters, as well as the predictive adjustment amounts of the printhead voltage, printhead temperature, and trimming parameters, into corresponding fuzzy sets through their respective membership functions. Inference is then performed using a predefined, more complex fuzzy rule base. This rule base includes empirical knowledge on multi-parameter collaborative control, such as appropriately reducing the positive adjustment amount of the trimming parameter to avoid excessive ink volume if the printhead voltage needs a significant adjustment and the current trimming parameter (black bar percentage) is already high. Through defuzzification, the fuzzy inference results are transformed into clear printhead voltage, printhead temperature, and trimming parameter settings (black bar percentage), thereby achieving the goal of multi-parameter collaborative optimization and smooth transition.

[0069] The expression for setting the value is:

[0070]

[0071] in, This is the nozzle voltage setting value. For nozzle temperature setpoint Set values ​​for trimming parameters. This refers to the actual current setting value of the printhead voltage of the inkjet printer. This is the actual current set value of the printhead temperature of the inkjet printer. This is the actual current setting value of the trimming parameters. This is a predictive adjustment amount for the nozzle voltage. This is a predictive adjustment amount for nozzle temperature. This is a predictive adjustment amount for the trimming parameters;

[0072] S2.2. Based on the calculation results, set the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer to new values. The trimming parameters are represented as the percentage of black bars in the inkjet printing software.

[0073] Furthermore, the control unit of the coding equipment receives new printhead voltage settings, new printhead temperature settings, and new trimming parameter settings (black bar percentage), and drives the corresponding voltage regulation circuit, temperature control unit, and trimming parameter configuration interface to precisely adjust the printhead's working parameters and trimming function to the target values. This ensures that the multi-parameter collaborative decisions output by the predictive compensation model can be accurately executed on the physical device, completing a full closed loop from digital instructions to physical world actions. This enables data-driven, forward-looking multi-parameter collaborative optimization to take effect in the actual production environment, laying a precise technological foundation for obtaining high-quality QR code printing results.

[0074] It should be noted that, based on the extended predictive adjustment amount output by the predictive compensation model, the setting operation should not only target the nozzle voltage and nozzle temperature, but also include trimming parameters (black bar percentage), and should fully consider other key process parameters such as nozzle height, waveform file (switch to a more suitable waveform), nozzle plate overlap parameters, and resolution. Through this comprehensive parameter setting, it is ensured that all known key influencing factors are adjusted to an optimized state before production begins, thereby achieving the goal of controlling quality fluctuations from the source.

[0075] S3. Based on the set printhead voltage, printhead temperature and trimming parameters, QR code printing is performed. The appearance quality of the printed QR code is inspected using an online visual inspection module to generate initial inspection results. Samples are periodically sampled for offline QR code level inspection.

[0076] S3.1 Using the set printhead voltage, printhead temperature and trimming parameters, print QR codes on paper, and the online visual inspection module collects the image of the printed QR codes.

[0077] Furthermore, printing QR codes on paper using pre-set printhead voltage, temperature, and trimming parameters is the core material realization step in the entire quality control process, transforming optimized digital process instructions into physical product quality. The coding equipment drives the printhead to perform precise ink droplet ejection based on the precisely set printhead voltage, temperature, and trimming parameters. The printhead voltage determines the ejection power and flight speed of the ink droplets, the printhead temperature affects the viscosity and flowability of the ink, and the trimming parameters directly regulate the percentage of black bars used to improve the clarity of the QR code edges, i.e., the distribution of ink output in the edge area. The synergistic effect of these three factors forms a QR code graphic with a specific physical shape at a predetermined position on the printing paper. The online visual inspection module uses a pre-configured high-resolution industrial camera to trigger and capture images of the printed QR codes in real time under specific lighting conditions, ensuring the immediacy, consistency, and high fidelity of image acquisition. This provides a reliable and standardized visual information source for subsequent quantitative analysis, completing the accurate conversion from digital process parameters to quantifiable and detectable physical objects.

[0078] S3.2 The online visual inspection module analyzes the collected images of the printed QR codes and generates initial inspection results of appearance quality score and defect code based on the appearance quality inspection results.

[0079] Furthermore, the preprocessing operations on the collected printed QR code images include grayscale conversion and noise filtering. Based on predefined image processing algorithms, key features of the QR code graphic are extracted, such as edge sharpness, module shape integrity, and the presence of unexpected ink dots. Appearance quality inspection is based on the extracted features for quantitative evaluation. Edge sharpness is scored by edge gradient intensity, and module shape integrity is determined by morphological analysis. Any defects that do not conform to the standard graphic, such as striations or blurring, will be classified into specific defect codes.

[0080] Specifically, the appearance quality score integrates the evaluation results of various features and outputs them in a percentage format, while the defect code accurately identifies the type and location of the flaw, thereby generating a structured and quantifiable initial inspection result, transforming subjective visual judgment into objective and traceable data indicators.

[0081] S3.3 Periodically extract samples from products that have been inspected by the online visual inspection module, and use an offline QR code detector to perform offline QR code level detection on the extracted samples to obtain offline QR code level data.

[0082] Furthermore, to supplement and verify the key online inspection capabilities, the sampling operation randomly selects representative samples from the products that have been inspected and recorded by the online visual inspection module, or at time intervals, according to the preset sampling frequency and sampling rules. This ensures that the production batch and process parameters of the samples can be accurately traced. The selected samples are then inspected using an offline QR code detector. The offline QR code detector uses international standard algorithms to precisely measure the micro-parameters of the QR code, such as symbol contrast, modulation ratio, and decoding rate, and outputs offline QR code level data that conforms to ISO standards.

[0083] S4. Compare the initial test results with the preset qualified threshold, combine them with the offline QR code level data to obtain the quality judgment, and record the time series data of environmental temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results and offline QR code level data.

[0084] S4.1 Compare the appearance quality score in the initial test results with the preset pass threshold, and compare the offline QR code level data with the preset level threshold. Based on the comparison results of the initial test results and the comparison results of the offline QR code level data, obtain the overall quality judgment result.

[0085] Furthermore, the appearance quality score in the initial inspection results is logically compared with a preset pass threshold to determine whether the appearance meets the standard. The offline QR code level data is compared with a more stringent preset level threshold to assess whether the symbol level meets the higher-level quality requirements. The comprehensive judgment logic makes a decision based on the two comparison conclusions. For example, when both the appearance quality score and the offline QR code level data are better than their respective thresholds, it is judged as a high-quality product. When the appearance quality score meets the standard but the offline QR code level data does not meet the standard, it is judged as a qualified product with potential risks and triggers in-depth analysis. When the appearance quality score does not meet the standard, it is directly judged as a non-qualified product. By integrating the dual standards of online rapid appearance inspection and offline precise level inspection, a multi-dimensional and refined grading judgment of product quality is achieved. It not only focuses on immediately visible defects but also on the intrinsic quality that affects the final user experience, thus providing a more accurate guide for the optimization of the production process.

[0086] S4.2. Associate the ambient temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results, and offline QR code level data with the current timestamp.

[0087] Furthermore, by acquiring the current timestamp of a high-precision clock source and binding it with environmental temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results, and offline QR code level data generated at the same production moment to form a data unit, this unique link of timestamps precisely correlates multi-source heterogeneous data generated at different times to the same production event. This ensures that each quality inspection result uniquely corresponds to the precise process conditions, environmental state, and authoritative level standards at the time the result was generated, laying an indispensable and solid foundation for subsequent analysis of the dynamic causal relationship between quality fluctuations and changes in multiple parameters.

[0088] S4.3. Record the environmental temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results, and offline QR code level data after the associated timestamp into the time series data.

[0089] Furthermore, a big data foundation for process traceability and intelligent analysis is constructed. The storage operation persistently stores each complete data record with a timestamp in the order of its timestamps into a dedicated time-series database or data file, forming a continuous time-series data stream. This systematically organizes massive amounts of instantaneous production and quality information into a holistic historical archive that evolves over time. Its data structure naturally supports efficient range queries and trend analysis, making it possible to conduct retrospective statistics on the production process, monitor real-time trends, and make predictive diagnoses based on historical patterns. This provides high-quality data support for higher-level intelligent decision-making.

[0090] S5. Perform real-time analysis on time series data, match the evolution patterns of defect types with pre-stored defect evolution maps, and generate self-healing instructions.

[0091] S5.1 Continuously monitor the numerical changes of the defect codes and offline QR code level data recorded in the initial detection results in the time series data stream, and identify the evolution pattern of defect types in the time series data.

[0092] Furthermore, it achieves advanced sensing capabilities for predictive diagnostics. The monitoring process tracks the latest records of continuously flowing time-series data using sliding time window technology, focusing on the temporal trends of specific defect code frequency in the initial detection results and the continuous drift direction and rate of offline QR code level data values. Identifying evolutionary patterns aims to capture slowly accumulating signs of quality degradation that characterize the development of potential faults from the continuous production data stream. For example, the frequency of a certain defect code changes from occasional to continuously increasing, or the offline QR code level data shows a monotonous decline. This elevates the perspective of quality analysis from the static acceptance of individual products to a dynamic assessment of the health status of the entire production line. It can keenly extract abnormal and directional evolutionary clues from seemingly normal production fluctuations, providing high-quality input signals for subsequent accurate diagnosis.

[0093] S5.2 By analyzing the correlation between nozzle voltage, nozzle temperature, ambient temperature and humidity data, trimming parameters, initial detection results and offline QR code level data in historical time series data, the evolution pattern of defect types with parameter drift is summarized and a defect evolution map is established.

[0094] Furthermore, the offline learning and knowledge mining process for constructing a diagnostic knowledge base is explored. The analysis process delves into long-term records stored in historical time-series data, employing data analysis techniques such as temporal association rule mining to study the causal relationships and temporal sequence between the long-term slow decay of nozzle voltage, the periodic fluctuations of nozzle temperature, the gradual changes in environmental temperature and humidity data, the adjustment history of trimming parameters and the appearance of specific defect codes in initial detection results, and the deterioration of offline QR code level data. The summarized patterns are abstracted and formalized into various typical defect evolution scenarios.

[0095] Specifically, for example, there is a strong correlation between the continuous slight decay of nozzle voltage, the monotonic decrease of offline QR code level data, and the gradual increase in the frequency of specific wire drawing defect codes, exhibiting a temporal sequence. These verified causal chains, after being structured and stored, form a defect evolution map. This explicit and structured encoding of the deep-seated patterns of equipment performance degradation and process parameter imbalances hidden within vast historical data transforms ineffable expert experience into precise, computable knowledge models, providing an intelligently matched, causally-rich reference map library for online real-time diagnostic systems.

[0096] S5.3 Match the evolution patterns of the identified defect types with the pre-stored defect evolution maps.

[0097] Furthermore, based on knowledge-based real-time diagnostic decision-making, the matching operation performs similarity pattern recognition by comparing the characteristics of the evolutionary patterns of defect types with the characteristics of various typical patterns recorded in the pre-stored defect evolution map. The matching process needs to consider the morphology of the evolutionary trend, the direction of change of key parameters, and the consistency of the time scale, which enables rapid benchmarking of the current production process status with historical experience and lessons learned. This allows for finding the most likely explanation for the observed quality evolution trend, i.e., locating the root cause, and providing a decision-making basis for generating precise corrective measures.

[0098] S5.4. Based on the pre-stored defect evolution map that has been successfully matched, generate a self-healing instruction to adjust the parameters.

[0099] Furthermore, the logic for generating self-healing instructions is directly derived from the causal laws contained in the pre-stored defect evolution map that has been successfully matched. It clearly indicates which process parameters should be adjusted in what direction to correct the identified defect evolution pattern. For example, if the map shows that the current trend is caused by the slow decay of the nozzle voltage, then the self-healing instruction includes the specific operational requirements for increasing the nozzle voltage.

[0100] Specifically, it automatically transforms complex quality status assessments into concise, clear, and immediately executable equipment control commands, realizing an automated closed loop from problem perception to problem handling, thereby proactively suppressing quality degradation and preventing the generation of ultimately defective products.

[0101] S6. Execute the self-healing command to adjust the printhead voltage of the inkjet printer, and return to the predictive adjustment setting to set the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer.

[0102] S6.1. Parse the self-healing command to obtain the adjustment amount of the printhead voltage of the inkjet printer.

[0103] Furthermore, the core of parsing self-healing instructions lies in accurately extracting executable parameter adjustment information. Self-healing instructions are usually carried in a structured data format, clearly containing target parameter identifiers and adjustment values. The parsing process obtains the adjustment amount of the printhead voltage of the inkjet printer by identifying specific fields in the instruction. For example, extracting parameters from the instruction: printhead voltage, operation: increase, magnitude: adjustment amount key information; transforming high-level decision instructions into precise operational quantities that can be directly understood by the low-level device control interface, ensuring that the self-healing intent can be transmitted and executed without loss, and providing clear input for subsequent physical parameter modifications.

[0104] S6.2 Modify the current setting value of the printhead voltage of the inkjet printer according to the adjustment amount of the self-healing instruction, and the process jumps to set the printhead voltage, printhead temperature and trimming parameters of the inkjet printer according to the predictive adjustment amount.

[0105] Furthermore, the adjustment amount of the received self-healing command is directly applied to the current setting value of the printhead voltage through the control interface provided by the inkjet printer, thereby completing the online update and correction of the core parameters of the equipment during operation. Then the process jumps to the step of setting the printhead voltage, printhead temperature and trimming parameters of the inkjet printer according to the predictive adjustment amount. This jump command means the start of a brand new quality control cycle, but the initial state of this cycle has already incorporated the correction effect achieved by the previous round of self-healing command.

[0106] Specifically, the mandatory process jump enables dynamic reset and iterative optimization of the entire quality control process, ensuring that the effective parameters verified after self-healing adjustment can be immediately used as the benchmark input for the next round of predictive compensation. This builds a perpetual optimization closed loop that can continuously adapt to fluctuations in production conditions and has autonomous optimization capabilities, fundamentally improving the adaptability and stability of production.

[0107] This embodiment also provides a QR code printing quality control system, including: a predictive compensation module, which acquires ink batch information, paper type information and environmental temperature and humidity data based on the production work order, compares them with the pre-stored standard production conditions, and inputs them into the predictive compensation model to output predictive adjustment amounts;

[0108] The parameter control module sets the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer based on predictive adjustment amounts.

[0109] The quality judgment module prints QR codes based on the set printhead voltage, printhead temperature and trimming parameters. It uses an online visual inspection module to inspect the appearance quality of the printed QR codes, generates initial inspection results, and periodically extracts samples for offline QR code level inspection.

[0110] The intelligent diagnostic module compares the initial test results with the preset qualified threshold, combines the offline QR code level data to obtain the quality judgment, and records the time series data of environmental temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results and offline QR code level data.

[0111] The self-healing decision module performs real-time analysis on the time series data, matches the evolution patterns of defect types with pre-stored defect evolution maps, and generates self-healing instructions.

[0112] The dynamic feedback module executes the self-healing command to adjust the printhead voltage of the inkjet printer and returns the predictive adjustment amount to set the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer.

[0113] This embodiment also provides a computer device applicable to the QR code printing quality control method, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the QR code printing quality control method proposed in the above embodiment.

[0114] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0115] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the QR code printing quality control method proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0116] In summary, this invention integrates production work order information, material data, and environmental parameters, and utilizes a predictive compensation model to achieve forward-looking intelligent pre-adjustment of printing process parameters. This mitigates quality risks at the initial production stage, generates real-time quality data streams through online visual inspection, and dynamically diagnoses defect evolution patterns based on time series analysis technology. When a potential quality degradation trend is identified, it automatically triggers and executes precise self-healing commands to adjust equipment parameters. By feeding back the self-healing control results to the parameter setting stage, a closed-loop intelligent control loop capable of adaptively responding to fluctuations in production conditions and equipment performance degradation is constructed. This achieves a qualitative leap from passive error correction to proactive prevention, improving the quality stability, production efficiency, and intelligence level of the QR code printing process.

[0117] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for quality control of QR code printing, characterized in that: This includes acquiring ink batch information, paper type information, and environmental temperature and humidity data based on production work orders, comparing them with pre-stored standard production conditions, and inputting them into a predictive compensation model to output predictive adjustment amounts. The printhead voltage, printhead temperature, and trimming parameters of the inkjet printer are set according to the predictive adjustment amount. QR code printing is performed based on the set printhead voltage, printhead temperature and trimming parameters. The appearance quality of the printed QR code is inspected using an online visual inspection module to generate initial inspection results. Samples are periodically sampled for offline QR code level inspection. The initial test results are compared with the preset qualified threshold, and the quality judgment is obtained by combining the offline QR code level data. The time series data of environmental temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results and offline QR code level data are recorded. The time series data is analyzed in real time, and the evolution patterns of defect types are matched with pre-stored defect evolution maps to generate self-healing instructions. The self-healing command is executed to adjust the printhead voltage of the inkjet printer, and the predictive adjustment amount is returned to set the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer.

2. The QR code printing quality control method as described in claim 1, characterized in that: Based on the production work order, ink batch information, paper type information, and environmental temperature and humidity data are obtained and compared with pre-stored standard production conditions. The data is then input into the predictive compensation model to output predictive adjustment amounts, including the following steps: Analyze the production work order to obtain the product brand and production quantity in the production work order. Based on the product brand, query and obtain the ink batch information and paper model information. By reading the current ambient temperature and humidity data through sensors, and by using the ink batch information, paper type information, ambient temperature and humidity data, printhead voltage and printhead temperature parameters corresponding to the high-quality QR code printing results in historical production work orders, standard production conditions are established. The ink batch information, paper type information, and ambient temperature and humidity data are compared with the pre-stored standard production conditions. The compared ink batch information, paper type information, and ambient temperature and humidity data are then input into the predictive compensation model, which outputs predictive adjustment amounts for key process parameters of the coding equipment.

3. The QR code printing quality control method as described in claim 2, characterized in that: The printhead voltage, printhead temperature, and trimming parameters of the inkjet printer are set according to the predictive adjustment amount, including the following steps: Read the predictive adjustment amount, and calculate the result by combining the predictive adjustment amount with the current set values ​​of the printhead voltage, printhead temperature and trimming parameters of the inkjet printer. The expression for setting the value is: in, This is the nozzle voltage setting value. For nozzle temperature setpoint Set values ​​for trimming parameters. This refers to the actual current setting value of the printhead voltage of the inkjet printer. This is the actual current set value of the printhead temperature of the inkjet printer. This is the actual current setting value of the trimming parameters. This is a predictive adjustment amount for the nozzle voltage. This is a predictive adjustment amount for nozzle temperature. This is a predictive adjustment amount for the trimming parameters; Based on the calculation results, the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer are set to new values. The trimming parameters are represented as the percentage of black bars in the inkjet printing software.

4. The QR code printing quality control method as described in claim 3, characterized in that: QR code printing is performed based on the set printhead voltage, printhead temperature, and trimming parameters. The printed QR codes are then inspected for appearance quality using an online visual inspection module to generate initial inspection results. Samples are periodically sampled for offline QR code quality testing, including the following steps: Using the set printhead voltage, printhead temperature, and trimming parameters, a QR code is printed on the paper, and the online visual inspection module captures the image of the printed QR code. The online visual inspection module analyzes the collected images of the printed QR codes and generates initial inspection results, including appearance quality scores and defect codes, based on the appearance quality inspection results. Samples are periodically taken from products that have passed the online visual inspection module, and offline QR code level detection is performed on the sampled products using an offline QR code detector to obtain offline QR code level data.

5. The QR code printing quality control method as described in claim 4, characterized in that: The initial test results are compared with the preset pass threshold, and combined with offline QR code level data to obtain a quality judgment. Time series data of environmental temperature and humidity, nozzle voltage, nozzle temperature, trimming parameters, initial test results, and offline QR code level data are recorded, including the following steps: The appearance quality score in the initial test results is compared with the preset pass threshold, and the offline QR code level data is compared with the preset level threshold. Based on the comparison results of the initial test results and the comparison results of the offline QR code level data, the quality judgment result is obtained by combining the results. Associate ambient temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results, and offline QR code level data with the current timestamp; The environmental temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results, and offline QR code level data associated with the timestamp are stored as records in the time series data.

6. The QR code printing quality control method as described in claim 5, characterized in that: The time-series data is analyzed in real time, and the evolution patterns of defect types are matched with pre-stored defect evolution maps to generate self-healing instructions, including the following steps: Continuously monitor the numerical changes of defect codes and offline QR code level data recorded in the initial detection results in the time series data stream to identify the evolution pattern of defect types in the time series data; By analyzing the correlation between nozzle voltage, nozzle temperature, ambient temperature and humidity data, trimming parameters, initial detection results and offline QR code level data in historical time series data, the evolution pattern of defect types with parameter drift was summarized and a defect evolution map was established. The evolution patterns of the identified defect types are matched with pre-stored defect evolution maps; Based on the successfully matched pre-stored defect evolution map, a self-healing instruction with adjusted parameters is generated.

7. The QR code printing quality control method as described in claim 6, characterized in that: Executing the self-healing command to adjust the printhead voltage of the inkjet printer, and returning the predictive adjustment amount to set the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer, includes the following steps: Parse the self-healing command to obtain the adjustment amount of the printhead voltage of the inkjet printer; The current setting of the printhead voltage of the inkjet printer is modified according to the adjustment amount of the self-healing command. The process then jumps to set the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer according to the predictive adjustment amount.

8. A QR code printing quality control system, based on the QR code printing quality control method according to any one of claims 1 to 7, characterized in that: This includes a predictive compensation module that, based on production work orders, acquires ink batch information, paper type information, and environmental temperature and humidity data, compares them with pre-stored standard production conditions, and inputs them into the predictive compensation model to output predictive adjustment amounts. The parameter control module sets the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer based on predictive adjustment amounts. The quality judgment module prints QR codes based on the set printhead voltage, printhead temperature and trimming parameters. It uses an online visual inspection module to inspect the appearance quality of the printed QR codes, generates initial inspection results, and periodically extracts samples for offline QR code level inspection. The intelligent diagnostic module compares the initial test results with the preset qualified threshold, combines the offline QR code level data to obtain the quality judgment, and records the time series data of environmental temperature and humidity data, nozzle voltage, nozzle temperature, trimming parameters, initial test results and offline QR code level data. The self-healing decision module performs real-time analysis on the time series data, matches the evolution patterns of defect types with pre-stored defect evolution maps, and generates self-healing instructions. The dynamic feedback module executes the self-healing command to adjust the printhead voltage of the inkjet printer and returns the predictive adjustment amount to set the printhead voltage, printhead temperature, and trimming parameters of the inkjet printer.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the QR code printing quality control method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the QR code printing quality control method according to any one of claims 1 to 7.

Images