Quantum sensing-based product production quality inspection management method and system

Abstract

The invention provides a product production quality inspection management method and system based on quantum sensing, and relates to the technical field of industrial automation and quantum detection, and the method comprises the steps: collecting the quantum phase difference, spin relaxation time and energy density data of a product through a quantum sensing array; performing quantum state decoupling processing on the multi-dimensional field data, and extracting a quantum hue consistency coefficient and a structure anomaly index; constructing a multi-scale convolutional neural network to generate a three-dimensional quality feature tensor; dynamically optimizing a defect judgment threshold value based on a quantum annealing algorithm; and generating a visual thermodynamic diagram and triggering a process control instruction. The system comprises a quantum sensing module, an edge computing unit, a dynamic decision engine, a man-machine interaction terminal and a feedback control module, and closed-loop management is realized through multi-physics field synchronous acquisition and real-time control. According to the method, traditional detection resolution limitation is broken through by utilizing quantum sensing, micro defects are accurately identified, and complex working conditions are adapted through dynamic threshold optimization.

Description

Technical Field

[0001] The present invention relates to the field of industrial automation and quantum detection technology, and in particular to a product production quality inspection management method and system based on quantum sensing. Background Art

[0002] In the production process of industrial products, quality inspection is the core link to ensure product consistency and reliability. Traditional quality inspection technologies mainly rely on optical imaging, ultrasonic detection or electromagnetic sensing, but these methods have significant limitations in detecting microscopic defects. Optical detection is limited by wavelength resolution and it is difficult to capture atomic-level lattice distortion or subsurface microcracks inside the material; ultrasonic technology has insufficient detection accuracy for complex structure products and is easily interfered by environmental noise; electromagnetic sensing cannot effectively identify the microscopic characteristics of non-metallic materials. In addition, existing detection systems mostly use fixed thresholds to determine defects, which cannot adapt to the dynamic fluctuations of parameters such as temperature and pressure during the production process, resulting in high false detection and missed detection rates. As the sophistication of products increases, traditional methods are insufficient in processing multi-physical field coupling data, making it difficult to establish an accurate mapping relationship between high-dimensional features and quality status.

[0003] At the data processing and decision-making level, existing technologies rely on traditional machine learning models or manual experience rules, and lack efficient analysis methods for the high-dimensional heterogeneous data generated by quantum sensing. Most systems use a single-scale feature extraction strategy, which cannot simultaneously capture the spatial correlation between microscopic defects and macroscopic anomalies, and there is a risk of information loss in the feature fusion process. At the same time, the visualization of quality inspection results is single, mostly based on two-dimensional grayscale images or numerical reports. It is difficult for operators to quickly locate the defect position and judge the severity, and manual re-inspection is inefficient. In the control feedback link, traditional systems lack a real-time closed-loop mechanism. There is a significant delay between the detection results and the adjustment of production line equipment. The calibration of process parameters relies on manual experience, making it difficult to achieve precise adaptive control.

[0004] Although existing technologies have demonstrated the potential for high-precision detection in laboratory environments, their industrial applications still face multiple challenges. Quantum signals are susceptible to environmental electromagnetic interference, resulting in a low signal-to-noise ratio. The timing synchronization accuracy during collaborative multi-sensor acquisition is insufficient, affecting the extraction of spatiotemporal features. Furthermore, the integration of quantum data with traditional control systems faces technical barriers, making it difficult to establish a complete closed loop from detection to execution. These issues severely restrict the practical application of quantum sensing technology in large-scale industrial production, and the industry urgently needs a systematic solution that integrates high-precision detection, intelligent decision-making, and real-time control. Summary of the Invention

[0005] In order to solve the technical problems in the prior art, the present invention provides a product production quality inspection management method and system based on quantum sensing.

[0006] The technical solutions provided by the present invention are as follows:

[0007] First aspect:

[0008] The present invention provides a product production quality inspection management method based on quantum sensing, comprising:

[0009] S1. Real-time acquisition of multi-dimensional field data of the target product through a quantum sensing array, wherein the multi-dimensional field data includes a quantum phase difference spectrum, a spin relaxation time distribution, and an energy density waveform;

[0010] S2. Perform quantum state decoupling processing on the multidimensional field data to extract the spatiotemporal correlation feature matrix, which contains two core parameters: the quantum hue consistency coefficient QCHC and the structural anomaly index SAI;

[0011] S3. Build a dynamic quality assessment model, input the spatiotemporal correlation feature matrix into the pre-trained multi-scale convolutional neural network MSCNN, and output a three-dimensional quality feature tensor;

[0012] S4, optimizes the quality decision boundary based on the quantum annealing algorithm and dynamically adjusts the defect judgment threshold according to real-time production parameters;

[0013] S5. Generate a visual quality heat map and trigger a feedback control instruction, wherein the feedback control instruction includes a process parameter adjustment amount and an equipment calibration vector.

[0014] Second aspect:

[0015] The present invention provides a product production quality inspection and management system based on quantum sensing, comprising:

[0016] A quantum sensing module equipped with a multi-physics field synchronous acquisition device; an edge computing unit with integrated quantum noise suppression and feature extraction accelerator; and a dynamic decision-making engine based on a quantum-inspired optimization algorithm library.

[0017] The beneficial effects brought about by the technical solution provided by the present invention include at least:

[0018] (1) In this invention, multi-dimensional quantum field data is collected through a quantum sensing array, combined with the high sensitivity of diamond NV color center sensors, to break through the physical limits of traditional detection technology and accurately capture microscopic defects such as atomic-level lattice distortion and microcracks within the material. Quantum state decoupling processing technology effectively separates environmental noise from target signals, extracts quantum hue consistency coefficients and structural anomaly indices, significantly improves the spatial resolution and reliability of defect identification, and solves the problem of insufficient microscopic feature detection capabilities of traditional methods.

[0019] (2) In this invention, a dynamic quality decision boundary is constructed based on the quantum annealing algorithm, the production process parameters are encoded as a potential barrier function in the quantum optimization model, and the energy state of the quality feature vector is analyzed in real time. When the production environment fluctuates and the feature shifts, the system automatically adjusts the defect judgment threshold to achieve intelligent matching of the detection standard and the real-time working conditions, overcoming the adaptability defects of the fixed threshold system in dynamic production scenarios and significantly reducing the risk of false detection and missed detection;

[0020] (3) In this invention, by mapping the three-dimensional quality feature tensor to the HSV color space, a multi-dimensional visualization heat map of hue, brightness, and saturation is generated, which intuitively presents the distribution and severity of defects. Combining an augmented reality interface with a real-time feedback control module, a response from test results to equipment calibration instructions is achieved in seconds, forming a closed loop of "perception-analysis-decision-execution" throughout the entire process. This significantly improves quality inspection efficiency and process control accuracy, and solves the technical bottlenecks of lag in human-computer interaction and isolated control in traditional systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 A flowchart of a product production quality inspection management method based on quantum sensing provided by an embodiment of the present invention;

[0023] Figure 2 A schematic diagram of a quantum state decoupling process flow for a product production quality inspection management method based on quantum sensing provided by an embodiment of the present invention;

[0024] Figure 3 A schematic diagram of a multi-scale convolutional neural network implementation process for a product production quality inspection management method based on quantum sensing provided by an embodiment of the present invention;

[0025] Figure 4 A schematic diagram of the barrier function construction and dynamic threshold control process of a product production quality inspection management method based on quantum sensing provided by an embodiment of the present invention;

[0026] Figure 5 A schematic diagram of the process flow of a visualized quality heat map for a product production quality inspection management method based on quantum sensing provided by an embodiment of the present invention;

[0027] Figure 6A schematic diagram of the flow of visual quality heat map generation and feedback control instruction triggering for a product production quality inspection management method based on quantum sensing provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0029] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0030] In the embodiments of the present invention, the terms "image" and "picture" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same. The terms "of," "corresponding," and "corresponding" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same.

[0031] In the embodiments of the present invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0032] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0033] Reference Manual Figure 1 , which shows a flow chart of a product production quality inspection management method based on quantum sensing provided by an embodiment of the present invention.

[0034] An embodiment of the present invention provides a method for managing product production quality inspection based on quantum sensing. The method can be implemented by a device for managing product production quality inspection based on quantum sensing. The device can be a terminal or a server. The process flow of the method for managing product production quality inspection based on quantum sensing can include the following steps:

[0035] S1. Collect multi-dimensional field data of the target product in real time through a quantum sensing array. The multi-dimensional field data includes quantum phase difference spectrum, spin relaxation time distribution and energy density waveform.

[0036] It is understandable that the core reason for real-time collection of multi-dimensional field data (quantum phase difference maps, spin relaxation time distribution, energy density waveforms) through quantum sensing arrays is that traditional sensors cannot capture microscopic quantum state characteristics, while quantum sensing technology can break through the limits of classical physics and achieve non-destructive detection of deep-level characteristics such as the internal lattice structure and energy distribution of materials. The role of this step is to provide a high-dimensional raw data base for quality detection. The quantum phase difference map reflects the consistency of atomic arrangement, the spin relaxation time reveals the lifetime characteristics of the defect state, and the energy density waveform characterizes the microscopic energy dissipation pattern. Logically, this step constitutes the input end of the technology chain and provides a physical basis for subsequent feature extraction and decision analysis.

[0037] S2. Perform quantum state decoupling processing on the multidimensional field data to extract the spatiotemporal correlation feature matrix, which contains two core parameters: the quantum chromatic consistency coefficient QCHC and the structural anomaly index SAI.

[0038] It is understandable that the necessity of implementing quantum state decoupling processing to extract the spatiotemporal correlation feature matrix stems from the strong coupling characteristics of quantum sensing data - environmental noise and multi-physical field interference cause the original signal to be mixed. After separating the eigenstate signal through the quantum entanglement decoupling equation, the two parameters of QCHC (quantified surface uniformity) and SAI (characterizing the degree of structural distortion) are extracted using spatiotemporal correlation analysis. Its function is to convert the original quantum data into computable engineering features. QCHC detects microscopic chromatic aberrations through quantum phase consistency, and SAI combines spatiotemporal gradients to identify hidden defects. Logically, this step is the hub for the conversion of data to information, providing feature input for the quality assessment model.

[0039] Furthermore, if Figure 2 As shown, the quantum state decoupling process is as follows:

[0040] S201, signal preprocessing:

[0041] S2011, using Daubechies9 wavelet basis to decompose the original signal into 5 layers, the decomposition formula is:

[0042]

[0043] Among them, cA k represents the k-th layer approximation coefficient, cD k represents the detail coefficient of the kth layer;

[0044] S2012. Calculate the time offset between multiple sensors using a cross-correlation algorithm:

[0045] Δt=arg max τ δs1(t)s2(t+τ)dt

[0046] Among them, s1(t) and s2(t) are the original signals of the two sensors, and is the time delay variable.

[0047] S202, quantum channel separation:

[0048] S2021. Construct the total Hamiltonian model:

[0049] H total =H system +H noise

[0050] Among them, H system represents the intrinsic Hamiltonian of the target system, H noise represents the environmental noise interference term, and the eigenstates are separated by the variational quantum eigensolver VQE. The number of iterations is set to 200 and the convergence tolerance is 1e-6;

[0051] S2022. Perform window Fourier transform on the decoupled signal:

[0052] STFT(t,f)=∫s(τ)w(τ-t)e -j2πfτ dτ

[0053] Where w(τ-t) is the window function, the window length is set to 10ms, and the window type is Hanning window.

[0054] S203, spatiotemporal feature fusion:

[0055] S2031. Calculate the quantum coherence of adjacent sensor nodes:

[0056]

[0057] Among them, |ψ i > represents the quantum state of the i-th sensor, <ψ i |ψ j > is the inner product of two quantum states;

[0058] S2032. Construct a Markov state transition matrix:

[0059]

[0060] Among them, N m→n Indicates the number of transitions from state m to state n.

[0061] S3. Build a dynamic quality assessment model, input the spatiotemporal correlation feature matrix into the pre-trained multi-scale convolutional neural network MSCNN, and output a three-dimensional quality feature tensor.

[0062] It is understandable that a dynamic quality assessment model is constructed and a multi-scale convolutional neural network (MSCNN) is used to process the spatiotemporal correlation feature matrix. Traditional single-scale models find it difficult to capture both microscopic defects (such as lattice dislocations) and macroscopic anomalies (such as surface cracks) at the same time. The MSCNN extracts atomic-level features through a 1×1 quantum convolution kernel and captures millimeter-level defect distribution through a 3×3 spatiotemporal convolution kernel, and then fuses multi-scale information through an attention mechanism to ultimately output a three-dimensional quality feature tensor. Its function is to establish a mapping relationship between quantum features and quality levels. The three-dimensional tensor (spatial coordinates × defect type × confidence level) can comprehensively characterize the product status. Logically, this step is the core decision-making unit of the technical solution, realizing intelligent mapping from features to quality criteria.

[0063] S4. Optimize the quality decision boundary based on the quantum annealing algorithm and dynamically adjust the defect judgment threshold according to real-time production parameters.

[0064] It is understandable that step S4 optimizes the quality decision boundary based on the quantum annealing algorithm. The fundamental reason is that the traditional static threshold cannot adapt to the dynamic fluctuations of production parameters (such as temperature and pressure). By encoding the quality characteristics as quantum bit spin states, embedding process constraints in the Hamiltonian, and using the quantum tunneling effect to cross the local optimal solution, the defect judgment threshold is dynamically adjusted. The purpose is to achieve adaptive optimization of defect detection. When the production environment changes, the decision boundary automatically drifts with the energy ground state. Logically, this step constitutes the dynamic adjustment center of quality control to ensure that the detection standards are strictly matched with the real-time working conditions.

[0065] S5. Generate a visual quality heat map and trigger a feedback control instruction, wherein the feedback control instruction includes a process parameter adjustment amount and an equipment calibration vector.

[0066] It is understandable that the design motivation for generating a visual quality heat map and triggering feedback control instructions is to solve the problems of poor interpretability and high control delay of traditional quality inspection results. By mapping the three-dimensional quality tensor to the HSV color space (H maps uniformity deviation, S maps defect probability, and V maps energy density), the heat map can intuitively locate the position and severity of the defect. At the same time, based on the response relationship between the quality feature tensor and the process parameter space, the equipment calibration vector (such as the robot arm posture correction) is calculated. The role is to form a "detection-display-control" closed loop, the heat map guides manual re-inspection, and the control instructions drive the production line to self-adjust. Logically, this step is the output terminal of the technical solution, completing the last link of quality control.

[0067] In a possible implementation, S2 specifically includes:

[0068] The quantum hue consistency coefficient QCHC is calculated by the following formula:

[0069]

[0070] Among them, μ c′ is the quantum phase shift mean of color channel c, σ max is the maximum standard deviation of the three channels.

[0071] In a possible implementation, S2 specifically includes:

[0072] The calculation method of the structural abnormality index SAI is:

[0073]

[0074] in, represents the second-order derivative norm of the quantum gradient field, ΔT is the energy fluctuation in the time dimension, σ G , σ T are the standard deviations of spatial gradient and temporal fluctuation, respectively.

[0075] In a possible implementation, S3 specifically includes:

[0076] The multi-scale convolutional neural network comprises:

[0077] 1×1 quantum convolution kernel for extracting microscopic defect features;

[0078] 3×3 spatiotemporal convolution kernel for capturing macroscopic anomaly features;

[0079] Dynamically fuse multi-scale features through attention mechanism.

[0080] Furthermore, if Figure 3 As shown in Figure 2, the implementation details of the multi-scale convolutional neural network are as follows:

[0081] S301. Network architecture implementation: A three-level feature processing architecture is used to achieve full-scale detection from micro-defect capture to macro-anomaly identification:

[0082] S3011 uses a 1×1 convolution kernel to intelligently compress and reorganize feature channels. The input layer receives a 64-channel feature map from the quantum decoupling module and compresses the channels to 16 dimensions using a trainable weight matrix, preserving key quantum features while removing redundant information. A channel shuffling mechanism is introduced to permute the output channels after each convolution operation, enhancing information exchange between different feature channels.

[0083] S3012. Dynamically evaluate the signal quality of each channel and randomly deactivate low-quality channels with a signal-to-noise ratio below 20dB. Specifically, during the training phase, the output of this channel is blocked with a 50% probability, forcing the network to learn robust feature representations.

[0084] S302, macro feature extraction layer:

[0085] S3021 uses a 3×3 convolution kernel with periodic boundary conditions to effectively capture spatial correlation features across sensor nodes. In the temporal dimension, a space-time cube is constructed by concatenating features across frames to identify anomalous propagation patterns of energy fluctuations. Dilated convolution technology is deployed with a dilation rate of 2, expanding the receptive field to three times that of traditional convolution without increasing computational complexity, ensuring sensitive detection of large-area defects.

[0086] S3022. Implement a grouped convolution strategy, dividing the input channels into four independent processing groups. Each group performs convolution operations in parallel before performing feature fusion, reducing the computational load to 25% of traditional methods. Dynamic precision quantization technology is used to represent low-frequency components in feature maps using 8-bit fixed-point numbers, while retaining 16-bit floating-point precision for high-frequency components. This reduces memory usage by 40% while maintaining accuracy.

[0087] S303, multi-scale feature fusion:

[0088] S3031. Design a dual-path attention mechanism. Path one generates channel attention weights through global average pooling, while path two utilizes spatial pyramid pooling to generate spatial attention masks. The two are superimposed to form a dynamic feature selection matrix, guiding the weighted fusion of micro and macro features. Residual connections are introduced during the fusion process, directly injecting 30% of the original input features into the final output to avoid degradation of deep-level features.

[0089] In one possible implementation, S4 optimizes the decision boundary through the quantum tunneling effect, specifically including:

[0090] The production parameter constraints are encoded as a barrier function, and the threshold adjustment is triggered when the quality feature vector crosses the barrier region.

[0091] Furthermore, if Figure 4 As shown, the process of barrier function construction and dynamic threshold control is as follows:

[0092] S401. Automatically extract key process parameters from the production line control system, including variables that directly impact product quality, such as temperature, pressure, and flow rate. The screening criteria are parameter fluctuations exceeding the historical normal range and strong correlation with past quality defects. For example, if the pressure value of an injection molding machine consistently exceeds the statistical range of qualified samples, this parameter will be marked as requiring monitoring.

[0093] S402: Convert the selected process parameter operating ranges into constraint boundaries in the quality feature space. In the high-dimensional space composed of quantum features, a spherical safety zone is defined, centered around the characteristic vector distribution of historically qualified products. The radius of this zone is dynamically adjusted based on the degree of dispersion of qualified samples to ensure coverage of normal production fluctuations.

[0094] S403: An energy barrier mechanism is established outside the safe zone. When the real-time detected quality feature vector approaches or exceeds the constraint boundary, the system automatically increases the energy level of the area. This energy barrier forms a "peak" structure in the feature space similar to a topographic map, with the energy value increasing more dramatically the closer to the danger zone.

[0095] S404: Calculate the current position of the quality feature vector in the energy topography in real time and evaluate the overall energy level using the structural anomaly index. If the energy values for multiple consecutive detection cycles exceed the dynamically calculated warning threshold, a valid barrier crossing event is identified. The warning threshold is automatically updated based on the recent energy mean and standard deviation to mitigate environmental noise.

[0096] S405: Generate threshold adjustment instructions based on the direction and distance the feature vector deviates from the safe zone. If the vector continues to deviate toward the high-temperature sensitive side, the defect determination threshold in the high-temperature direction is correspondingly relaxed. If the vector suddenly jumps into the high-voltage danger zone, the detection sensitivity in that area is temporarily increased. The adjustment amplitude is logarithmically related to the deviation distance, ensuring fine-tuning for small fluctuations and rapid response to large deviations.

[0097] S406: Threshold adjustment is performed using smoothing filtering technology to avoid misjudgments caused by sudden changes in the detection standard. By setting a transition time constant, the new threshold gradually approaches the target value within a preset time window, while retaining a priority channel for manual intervention. The adjusted threshold is synchronized to all detection nodes in real time, and the implementation effect is verified through closed-loop feedback.

[0098] In a possible implementation, S5 includes:

[0099] In the visualization quality heat map:

[0100] Hue component mapping surface uniformity deviation value;

[0101] The brightness component is positively correlated with the energy density distribution;

[0102] The saturation component reflects the defect probability intensity.

[0103] Furthermore, if Figure 5 As shown in the figure, the relevant process of visualizing the quality heat map is as follows:

[0104] S501. Establish a mapping rule between surface uniformity deviation value and hue component. In the HSV color model, the base hue of qualified products is set to green (120 degrees). As the deviation value increases, the hue gradually changes in a clockwise direction. Severely exceeded areas are displayed in red (0 degrees), and moderate deviations transition to yellow (60 degrees), forming an intuitive hue alarm system.

[0105] S502. Normalize the energy density data collected by the quantum sensor array, and use gamma correction to nonlinearly map the energy value to the brightness channel. High energy density areas are highlighted, and low energy areas remain dark. At the same time, dynamic range compression is implemented to prevent local overexposure.

[0106] S503. Control the intensity of the saturation component based on the defect probability confidence level output by the multi-scale convolutional neural network. When the defect probability is less than 5%, a low saturation soft display is used. Within the probability range of 5%-95%, the saturation increases linearly. When the probability exceeds 95%, the display switches to a high saturation flashing warning mode.

[0107] S504: Integrate hue, brightness, and saturation three-dimensional data to generate a pseudo-color heat map with overlaid spatial coordinate information. Sub-pixel anti-aliasing processing is implemented in the rendering engine to ensure clear color boundaries at different resolutions, and a temporal integration algorithm is used to eliminate flicker interference caused by transient noise.

[0108] S505. Deploy a dynamic visual optimization mechanism. When the operator focuses on a specific area, the system automatically enhances the color contrast and overlays contour lines. The system continuously monitors the position of the human eye's gaze point and performs real-time resolution enhancement on key observation areas.

[0109] S506: The resulting heat map is overlaid with the 3D model of the production equipment, creating a fusion of reality and virtuality within the augmented reality interface. Defective areas are automatically associated with process parameter adjustment suggestions. Clicking a color block on the heat map retrieves raw sensor data, such as the quantum phase difference map, at that location.

[0110] In one possible implementation, a time reversal algorithm is used to perform traceability analysis on historical quality data to establish a defect evolution prediction model.

[0111] Furthermore, if Figure 6 As shown in the figure, the process of generating a visual quality heat map and triggering feedback control instructions is as follows:

[0112] S601: Receive a 3D quality feature tensor input and map its spatial dimensions to the HSV color space. Normalize the surface uniformity deviation value at each detection point, and create a linear gradient of hue values from green (120°) to red (0°) corresponding to the deviation from low to high, forming a hue component reference layer.

[0113] S602: Analyze the energy density distribution data and convert the energy values at each point into luminance components using a logarithmic relationship. Dynamic luminance suppression is enabled for areas where the energy density is 200% above the average level to prevent local overexposure. For areas where the energy density is below 50%, dark details are enhanced to ensure visibility in low-energy areas.

[0114] S603. Calculate the saturation component based on the defect probability intensity. When the probability is less than 5%, use a 20% basic saturation. For every 1% increase in probability, the saturation increases by 2%. When the probability exceeds 95%, enable a high-saturation pulse flashing mode. The flashing frequency is positively correlated with the severity of the defect.

[0115] S604: Fusion of hue, brightness, and saturation three-dimensional data generates an initial heat map. A spatial interpolation algorithm is used to eliminate gaps in sensor placement. Edge sharpening technology is used to enhance the display of defect boundaries. The final output is a pseudo-color image with a resolution of no less than 4K.

[0116] S605. Calculate process parameter adjustments based on the defect distribution characteristics in the thermal map. For continuous, large-area defects, generate instructions for synchronously adjusting the equipment's pressure and temperature downwards. For discrete, point-like defects, trigger the calculation of the robot arm's position calibration vector, with an accuracy requirement of ±0.1mm.

[0117] S606. Establish a feedback control instruction priority mechanism, set the critical defect instructions that directly affect product safety as the highest priority (response delay <50ms), set the appearance defect instructions as the medium priority (response delay <200ms), and automatically add diagnostic log recording requirements for historical repeated defects.

[0118] In one possible implementation, the quantum sensing array uses a diamond NV color center sensor whose sensitivity satisfies:

[0119]

[0120] Where δB is the magnetic field detection accuracy, h is the Planck constant, γ is the gyromagnetic ratio, T2 is the coherence time, and N is the number of effective sensing units.

[0121] In one possible implementation, generating the feedback control instruction includes:

[0122] Map the quality feature tensor to the process parameter space through dimensionality reduction and calculate the parameter adjustment vector:

[0123] Δθ=W·(qq ref )

[0124] Where W is the pre-trained process response matrix, q is the real-time quality vector, and q ref is the standard mass vector.

[0125] Reference Manual Figure 2 , which shows a structural diagram of a product production quality inspection and management system based on quantum sensing provided by an embodiment of the present invention.

[0126] The present invention also provides a product production quality inspection management system based on quantum sensing, which is applied to the product production quality inspection management method based on quantum sensing, comprising:

[0127] A quantum sensing module equipped with a multi-physics field synchronous acquisition device; an edge computing unit with integrated quantum noise suppression and feature extraction accelerator; and a dynamic decision-making engine based on a quantum-inspired optimization algorithm library.

[0128] The system consists of a quantum sensing module, an edge computing unit, a dynamic decision-making engine, a human-machine interaction terminal, and a feedback control module. The quantum sensing module, equipped with a diamond NV color center sensor and a superconducting quantum interference device, is deployed at a production line inspection node. Using a multi-physics field synchronous acquisition device, it acquires the target product's quantum phase difference spectrum, spin relaxation time distribution, and energy density waveform in real time. The raw data is transmitted to the edge computing unit via a fiber optic network. The edge computing unit integrates a quantum noise suppression unit and a feature extraction accelerator to perform noise reduction and spatiotemporal feature extraction on the sensor data, generating a spatiotemporal correlation feature matrix containing the quantum chromatic consistency coefficient and structural anomaly index. The processed data is then transmitted to the dynamic decision-making engine via an industrial communication protocol.

[0129] The dynamic decision engine deploys a multi-core heterogeneous computing architecture with a built-in quantum-inspired optimization algorithm library. After receiving the spatiotemporal correlation feature matrix, it uses a pre-trained process response matrix to calculate the deviation between the quality feature vector and the standard vector. Combined with a quantum annealing algorithm, it optimizes the decision boundary and generates a dynamic defect determination threshold. Simultaneously, it outputs process parameter adjustments and equipment calibration vectors to the feedback control module via a real-time operating system. The feedback control module, which includes an industrial communication protocol stack and directly connects to production line control equipment, converts calibration commands into executable signals for the equipment, driving the robot arm's posture correction or process parameter adjustment, forming a closed-loop detection-control link.

[0130] The human-computer interaction terminal supports augmented reality display and multi-terminal collaborative operation. After receiving the three-dimensional quality feature tensor output by the dynamic decision engine, it is mapped to the HSV color space to generate a visual quality heat map. The hue component reflects the surface uniformity deviation, the brightness component is associated with the energy density distribution, and the saturation component indicates the probability intensity of the defect. Operators can view defect location information in real time through the interactive interface and manually intervene in high-priority control instructions. Data synchronization between system modules is achieved through a distributed storage array. The quantum sensing module and the edge computing unit use a hard real-time communication link. The dynamic decision engine and the feedback control module are connected via a low-latency industrial network, ensuring end-to-end collaboration from data acquisition to control execution.

[0131] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0132] (1) In this invention, multi-dimensional quantum field data is collected through a quantum sensing array, combined with the high sensitivity of diamond NV color center sensors, to break through the physical limits of traditional detection technology and accurately capture microscopic defects such as atomic-level lattice distortion and microcracks within the material. Quantum state decoupling processing technology effectively separates environmental noise from target signals, extracts quantum hue consistency coefficients and structural anomaly indices, significantly improves the spatial resolution and reliability of defect identification, and solves the problem of insufficient microscopic feature detection capabilities of traditional methods.

[0133] (2) In this invention, a dynamic quality decision boundary is constructed based on the quantum annealing algorithm, the production process parameters are encoded as a potential barrier function in the quantum optimization model, and the energy state of the quality feature vector is analyzed in real time. When the production environment fluctuates and the feature shifts, the system automatically adjusts the defect judgment threshold to achieve intelligent matching of the detection standard and the real-time working conditions, overcoming the adaptability defects of the fixed threshold system in dynamic production scenarios and significantly reducing the risk of false detection and missed detection;

[0134] (3) In this invention, by mapping the three-dimensional quality feature tensor to the HSV color space, a multi-dimensional visualization heat map of hue, brightness, and saturation is generated, which intuitively presents the distribution and severity of defects. Combining an augmented reality interface with a real-time feedback control module, a response from test results to equipment calibration instructions is achieved in seconds, forming a closed loop of "perception-analysis-decision-execution" throughout the entire process. This significantly improves quality inspection efficiency and process control accuracy, and solves the technical bottlenecks of lag in human-computer interaction and isolated control in traditional systems.

[0135] The above content is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

[0136] There are a few points to note:

[0137] (1) The drawings of the embodiments of the present invention only relate to the structures related to the embodiments of the present invention. Other structures may refer to conventional designs.

[0138] (2) For the sake of clarity, the thickness of layers or regions in the drawings used to describe the embodiments of the present invention are exaggerated or reduced, that is, these drawings are not drawn to scale. It is understood that when an element such as a layer, film, region, or substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or intervening elements may be present.

[0139] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to form new embodiments.

[0140] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A product production quality inspection management method based on quantum sensing, characterized in that: include: S1. Real-time acquisition of multi-dimensional field data of the target product through a quantum sensing array, wherein the multi-dimensional field data includes a quantum phase difference spectrum, a spin relaxation time distribution, and an energy density waveform; S2. Perform quantum state decoupling processing on the multidimensional field data to extract the spatiotemporal correlation feature matrix, which contains two core parameters: the quantum hue consistency coefficient QCHC and the structural anomaly index SAI; S3. Build a dynamic quality assessment model, input the spatiotemporal correlation feature matrix into the pre-trained multi-scale convolutional neural network MSCNN, and output a three-dimensional quality feature tensor; S4, optimizes the quality decision boundary based on the quantum annealing algorithm and dynamically adjusts the defect judgment threshold according to real-time production parameters; S5. Generate a visual quality heat map and trigger a feedback control instruction, wherein the feedback control instruction includes a process parameter adjustment amount and an equipment calibration vector.

2. A product production quality inspection management method based on quantum sensing according to claim 1, characterized in that: The S2 specifically includes: The quantum hue consistency coefficient QCHC is calculated by the following formula: Among them, μ c′ is the quantum phase shift mean of color channel c, σ max is the maximum standard deviation of the three channels.

3. The method for product production quality inspection and management based on quantum sensing according to claim 1, characterized in that: The S2 specifically includes: The calculation method of the structural abnormality index SAI is: in, represents the second-order derivative norm of the quantum gradient field, ΔT is the energy fluctuation in the time dimension, σ G , σ T are the standard deviations of spatial gradient and temporal fluctuation, respectively.

4. The method for product production quality inspection and management based on quantum sensing according to claim 1, characterized in that: The S3 specifically includes: The multi-scale convolutional neural network comprises: 1×1 quantum convolution kernel for extracting microscopic defect features; 3×3 spatiotemporal convolution kernel for capturing macroscopic anomaly features; Dynamically fuse multi-scale features through attention mechanism.

5. The method for product production quality inspection and management based on quantum sensing according to claim 1, characterized in that: The S4 optimizes the decision boundary through the quantum tunneling effect, specifically including: The production parameter constraints are encoded as a barrier function, and the threshold adjustment is triggered when the quality feature vector crosses the barrier region.

6. The method for product production quality inspection and management based on quantum sensing according to claim 1, characterized in that: The S5 includes: In the visualization quality heat map: Hue component mapping surface uniformity deviation value; The brightness component is positively correlated with the energy density distribution; The saturation component reflects the defect probability intensity.

7. The method for product production quality inspection and management based on quantum sensing according to claim 1, characterized in that: include: The historical quality data is traced and analyzed through the time reversal algorithm to establish a defect evolution prediction model.

8. The method for product production quality inspection and management based on quantum sensing according to claim 1, characterized in that: include: The quantum sensing array uses diamond NV color center sensors, whose sensitivity meets the following requirements: Where δB is the magnetic field detection accuracy, h is the Planck constant, γ is the gyromagnetic ratio, T2 is the coherence time, and N is the number of effective sensing units.

9. The method for product production quality inspection and management based on quantum sensing according to claim 1, characterized in that: The feedback control instruction generation includes: Map the quality feature tensor to the process parameter space through dimensionality reduction and calculate the parameter adjustment vector: Δθ=W·(q-q ref ) Where W is the pre-trained process response matrix, q is the real-time quality vector, and q ref is the standard mass vector.

10. A product production quality inspection and management system based on quantum sensing, characterized in that: include: A quantum sensing module equipped with a multi-physics field synchronous acquisition device; Edge computing unit with integrated quantum noise suppression and feature extraction accelerator; Deploy a dynamic decision engine based on a library of quantum-inspired optimization algorithms.

Images