Paper packaging box strength detection device based on pressure sensor

Abstract

The invention discloses a paper packaging box strength detection device based on a pressure sensor, relates to the technical field of paper packaging box strength detection, and constructs a complete detection device composed of a pressure applying mechanism, the pressure sensor, a displacement sensor, a control module, an accuracy analysis module, an adjustment module and a detection result output module. According to the system, through combined preprocessing of pressure and displacement data, a force value change rate and a deformation trend characteristic are extracted, and accuracy classification judgment is carried out based on a combination relation of the force value change rate and the deformation trend characteristic; and when the judgment result is not completely accurate, an adaptive adjustment mechanism is automatically started to dynamically prolong the compression process, continuously collect data and optimize the judgment result, so that the accuracy of damage identification and the intelligent level of the system are remarkably improved, and finally, a maximum pressure bearing value, a state classification result and a complete force-displacement curve are output. And the reliability and adaptability of strength evaluation are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of paper packaging box strength detection, and in particular to a paper packaging box strength detection device based on a pressure sensor. Background Art

[0002] Strength testing of paper packaging boxes involves using a series of physical tests to assess the box's ability to withstand external forces during transportation, stacking, and handling, ensuring it resists deformation or damage during actual use. Common test items include compressive strength, bursting strength, and edge crush strength. These indicators reflect the robustness of the box's structure and its ability to reliably protect the contents. Strength testing can be used to determine whether the box meets safety standards for transportation and storage.

[0003] The prior art has the following deficiencies:

[0004] Existing strength testing systems for paper packaging boxes, based on pressure and displacement data, can cause sudden drops in pressure or increases in displacement during actual testing due to local collapse or brief elastic rebound. This can lead the system to misinterpret damage and record an understated maximum pressure value, severely impacting the accuracy of strength assessments. This issue stems from the fact that the testing logic relies on a single threshold and lacks dynamic recognition of material behavior. If not improved, this could lead to misleading product designs, increased costs, and inaccurate quality control. Summary of the Invention

[0005] The purpose of the present invention is to provide a paper packaging box strength detection device based on a pressure sensor to solve the shortcomings of the background technology.

[0006] In order to achieve the above-mentioned object, the present invention provides the following technical solution: a paper packaging box strength detection device based on a pressure sensor, comprising a pressure applying mechanism, a pressure sensor, a displacement sensor, a control module, an accuracy analysis module, an adjustment module, and a detection result output module;

[0007] A pressure mechanism, used to apply vertical pressure to the paper packaging box being inspected;

[0008] Pressure sensor, used to collect real-time pressure data of paper packaging boxes during the pressure process;

[0009] Displacement sensor, used to collect real-time data of the pressure plate displacement during the pressure application process;

[0010] A control module, wherein the control module pre-processes the collected pressure data and displacement data, and extracts force value change rate characteristics and deformation trend characteristics respectively;

[0011] An accuracy analysis module, based on the combined relationship between the force value change rate feature and the deformation trend feature, classifies the current detection state and divides the judgment result of the local deformation of the carton into accuracy judgment, partial accuracy judgment and inaccuracy judgment;

[0012] The adjustment module, when it is judged that the accuracy is not complete, activates the adaptive adjustment mechanism, including dynamically extending the compression time window, allowing the platen to continue pressing down and re-collecting data;

[0013] The test result output module outputs the test results including the maximum pressure value, state classification results and pressure-displacement curve.

[0014] Preferably, the pressure mechanism includes an upper pressure plate and a lower pressure plate, and the upper pressure plate is driven by an electric screw drive device or a servo motor system and moves in a vertical direction under the guidance of a guide mechanism to apply controllable pressure to the carton.

[0015] Preferably, after analyzing the force value change rate characteristics, the force value change rate abnormal value is generated by the following method: the discrete sampling points of the pressure sensor during the test are recorded as: ;in: is the pressure value at the nth moment, calculate the pressure change rate at each moment: ;in: is the pressure change rate at the i-th time point, Δt is the time interval between two consecutive sampling points, and the result is a change rate sequence: ; is the pressure change rate at the nth time point, and the mean of the change rate is calculated and standard deviation ; Calculate the Z-score value, the expression is: ;like , is the abnormal rate point, where is the Z-score value of the i-th rate point, A is the abnormal threshold coefficient, and traverses all , record satisfaction Time point , which is the potential abnormal point, and the maximum pressure change rate of all potential abnormal points is taken as the abnormal value of the force change rate.

[0016] Preferably, after analyzing the deformation trend characteristics, the compression deformation rate anomaly value is generated by: obtaining the displacement sequence during the compression process from the displacement sensor: , is the displacement of the platen at the mth moment, Δs is the sampling time interval; the acceleration of each point is calculated using the central difference method : ; Define a sliding window of size w and intercept each window from the acceleration sequence: ; For each window, calculate its mean , calculate the current acceleration The deviation from the window mean is used as the compression deformation rate anomaly, and the expression is: sd ; Where sd is the abnormal value of compression deformation rate.

[0017] Preferably, the accuracy analysis module classifies the current detection state based on the combined relationship between the force value change rate feature and the deformation trend feature, and divides the judgment result of the local deformation of the carton into accuracy judgment, incomplete accuracy judgment and inaccuracy judgment, specifically including:

[0018] The force value change rate anomalies and compression deformation rate anomalies are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the accuracy score value label of the local deformation judgment result of the carton as the prediction target, and takes minimizing the sum of the prediction errors of the accuracy score value labels of the local deformation judgment result of all cartons as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence, and the model training is stopped. The accuracy score value of the local deformation judgment result of the carton is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0019] Preferably, the obtained accuracy score value of the local deformation judgment result of the carton is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the accuracy score value of the local deformation judgment result of the carton is compared with the first standard threshold and the second standard threshold respectively;

[0020] If the accuracy score of the carton local deformation judgment result is greater than the second standard threshold, it is classified as an accuracy judgment and can be directly adopted;

[0021] If the accuracy score of the carton local deformation judgment result is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it is classified as an incomplete accuracy judgment and enters the adaptive adjustment mechanism for secondary judgment or delayed confirmation;

[0022] If the accuracy score of the carton local deformation judgment result is less than the first standard threshold, it is classified as an inaccuracy judgment and abnormal processing or parameter reset is performed.

[0023] Preferably, the adjustment module, when determining that the accuracy is not complete, starts an adaptive adjustment mechanism, including dynamically extending the compression time window, allowing the pressing plate to continue pressing down and re-collecting data, specifically:

[0024] Set the initial extended compression time ΔT0 for additional sampling of compressed data;

[0025] At the same time, the delay mode of pressing the plate is enabled to keep the compression rate constant;

[0026] Start a new round of data collection;

[0027] At fixed time intervals, the latest accuracy score S(t) is recalculated to form a score curve sequence: ; Calculate the rate of change of the score in the last two time periods , the expression is: If ΔS>0, it means the score value is rising, structural information is gradually emerging, and the time window is extended; if ΔS≤0, the score value is stagnant or decreasing, and the window is ended or an alarm is issued;

[0028] If the score is judged to have a significant upward trend, the time window will be automatically extended: ; is the compression time after extension; if the score value has stabilized or decreased, the extension window is terminated and the maximum score value is fixed for subsequent judgment.

[0029] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0030] 1. This invention addresses the misjudgment problem inherent in traditional paper packaging box strength testing systems, often caused by reliance on a single threshold, by introducing a dual-dimensional judgment mechanism based on pressure change rate and compression deformation trend characteristics. By extracting outliers and introducing an accuracy scoring mechanism, combined with a machine learning model to classify and judge structural damage states, the system significantly improves the detection system's accuracy in distinguishing local deformation from actual damage, avoids distortion of maximum pressure values caused by prematurely triggering damage judgments, and enhances the accuracy and credibility of assessments.

[0031] 2. By incorporating an adaptive adjustment module, this invention dynamically extends the compression time window and recollects data when test results are inaccurate. This module automatically optimizes the judgment results based on changes in scoring trends, enabling intelligent adjustment of the test process and correction of misjudgments. Finally, the test result output module outputs structured data and force-displacement curves, facilitating user visualization and historical traceability. The overall system boasts high precision, strong robustness, and excellent industrial applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0033] Figure 1 This is a mind map of the device module of the present invention. DETAILED DESCRIPTION

[0034] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0035] For examples, see Figure 1 As shown, the paper packaging box strength detection device based on the pressure sensor of this embodiment includes a pressure mechanism, a pressure sensor, a displacement sensor, a control module, an accuracy analysis module, an adjustment module and a detection result output module;

[0036] A pressure mechanism, used to apply vertical pressure to the paper packaging box being inspected;

[0037] Pressure sensor, used to collect real-time pressure data of paper packaging boxes during the pressure process;

[0038] Displacement sensor, used to collect real-time data of the pressure plate displacement during the pressure application process;

[0039] A control module, wherein the control module pre-processes the collected pressure data and displacement data, and extracts force value change rate characteristics and deformation trend characteristics respectively;

[0040] An accuracy analysis module, based on the combined relationship between the force value change rate feature and the deformation trend feature, classifies the current detection state and divides the judgment result of the local deformation of the carton into accuracy judgment, partial accuracy judgment and inaccuracy judgment;

[0041] The adjustment module, when it is judged that the accuracy is not complete, activates the adaptive adjustment mechanism, including dynamically extending the compression time window, allowing the platen to continue pressing down and re-collecting data;

[0042] The test result output module outputs the test results including the maximum pressure value, state classification results and pressure-displacement curve.

[0043] The pressure mechanism is used to apply controllable vertical pressure to the paper packaging box being tested, simulating the pressure conditions of the packaging box during actual stacking or transportation. Its structure generally includes:

[0044] Upper and lower platens: The upper platen moves vertically via a drive mechanism, forming a pressure clamping area with the fixed lower platen. Both surfaces should be flat and rigid to ensure uniform force distribution.

[0045] Driving device: preferably an electric screw drive, an electric cylinder, a servo motor system or a hydraulic cylinder device, which is used to accurately control the moving speed and displacement range of the upper pressing plate.

[0046] Guide mechanism: such as guide columns, linear slide rails, etc., used to ensure that the upper pressure plate moves only in the vertical direction, avoiding deviation or tilt during the pressure application process, and ensuring the stability and accuracy of the pressure application.

[0047] Load support structure: The base and frame design have good rigidity and stability to withstand the reaction force under high load.

[0048] The pressure mechanism supports setting different loading speeds (such as 5mm / min~100mm / min) and stop conditions (such as force or displacement thresholds), meeting the compression testing requirements under multiple standards (such as ISO 12048 and GB / T 4857.4).

[0049] The pressure sensor is used to collect real-time vertical pressure data on the paper packaging box during the pressure application process and is a key component for the device to obtain compressive strength. Its typical parameters and configuration are as follows:

[0050] Sensor type: preferably a strain gauge pressure sensor, piezoelectric sensor or torque sensor with high sensitivity and high linearity.

[0051] Installation location: It can be installed below the upper platen, in the force path of the drive structure, or at the mechanical connection with the loading end to accurately reflect the actual load on the carton.

[0052] Range and accuracy: The sensor range is selected according to the carton size and pressure bearing capacity, such as 0–5000N, 0–10kN, etc. The output accuracy can reach ±0.5%FS or better.

[0053] Output method: The sensor outputs an analog voltage or current signal (such as 0–5V, 4–20mA), which is connected to an A / D acquisition card or embedded controller. It supports high-frequency sampling (≥100Hz) to track force changes in real time.

[0054] Sensor data is used to determine the maximum bearing pressure value and force mutation point, and serves as an important basis for judging structural damage.

[0055] The displacement sensor is used to collect the real-time movement distance of the upper platen during the compression process to reflect the compression deformation process of the carton, which is the basis for drawing the force-displacement curve. The specific configuration is as follows:

[0056] Sensor type: Linear displacement sensor (LVDT), potentiometer linear displacement sensor or photoelectric encoder are commonly used. Some high-end systems use laser displacement measurement devices.

[0057] Installation method: Install in conjunction with the moving parts of the upper pressure plate to measure its displacement relative to the initial position. The installation position should avoid mechanical interference and signal obstruction.

[0058] Measuring range and resolution: Designed according to the carton compression range, such as 0–300mm range, with a resolution of up to 0.01mm to ensure accurate identification of tiny deformations.

[0059] Data output: Sensor data and pressure data are collected synchronously to calculate compression rate, compression mutation point and structural instability behavior.

[0060] Displacement data is not only used to determine structural damage, but also assists in structural behavior analysis and adaptive adjustment of algorithm judgment. It is one of the necessary inputs for realizing intelligent judgment systems.

[0061] The control module is the core functional module of this device, which is used to coordinate the pressure application process, collect and analyze pressure and displacement data in real time, and realize intelligent judgment of the damage state of the carton structure based on multi-dimensional characteristics. Its main components and functions are as follows:

[0062] The control module continuously acquires the raw force and displacement data during the pressure application process through real-time communication with the pressure sensor and displacement sensor, and performs the following preprocessing operations on it:

[0063] Filtering: Use sliding average, median filtering or low-pass filtering algorithms to effectively remove high-frequency noise and mechanical jitter interference;

[0064] Data registration: align pressure and displacement data with a unified timestamp to ensure data synchronization;

[0065] Abnormal rejection: Identify and eliminate abnormal values caused by sensor instability or initial oscillation.

[0066] Based on preprocessed data, this module extracts the force change rate characteristics and deformation trend characteristics to comprehensively determine whether the carton has entered a state of structural damage:

[0067] The force change rate characteristic represents the pressure change rate per unit time obtained by performing a first-order derivative calculation on continuous pressure data. It is used to identify whether there is a sudden drop in force during the compression process, which is usually a direct sign of carton instability or crushing. If the derivative value exceeds a preset negative threshold (such as -100N / s), it is judged to be a rapid drop.

[0068] After analyzing the force value change rate characteristics, the force value change rate abnormal value is generated. The generation method is:

[0069] The discrete sampling points of the pressure sensor during the test are recorded as: ;in: is the pressure value at the nth moment, in N (Newton), the sampling interval is Δt, in seconds (s), usually 0.1s, 0.01s, etc.; calculate the pressure change rate at each moment (pressure change per unit time): ;in: is the pressure change rate at the i-th time point (N / s), Δt is the time interval between two consecutive sampling points, and the result is a change rate sequence: ; is the pressure change rate at the nth time point, and the mean of the change rate is calculated and standard deviation ; Calculate the Z-score value, the expression is: ;like , is the abnormal rate point, where is the Z-score value of the i-th rate point, A is the abnormal threshold coefficient, and the common value is 2 or 3; traverse all , record satisfaction Time point , which is a potential abnormal point (which may be a structural damage point or a local instability point). The maximum pressure change rate of all abnormal points is taken as the abnormal value of the force change rate.

[0070] A large, abnormally large force rate of change, especially a significantly negative value, indicates a dramatic drop in pressure within a very short period of time, often corresponding to a sudden collapse or structural failure of the carton. This high-amplitude, sudden rate change indicates a high degree of certainty regarding the failure. Therefore, existing detection systems are generally accurate in determining structural failure in such situations, with a low probability of misjudgment.

[0071] Conversely, when the force change rate is abnormally small, it indicates a slow drop or weak fluctuation in the force, likely due to localized denting, elastic rebound, or short-term structural instability rather than actual damage. In this case, existing systems that rely solely on a set threshold to trigger damage determination are prone to inaccurate or misjudgment, misidentifying local deformation as overall structural failure.

[0072] The deformation trend feature indicates that the second-order derivative of the displacement data is calculated to capture the inflection point of the change in the compression deformation rate. The sudden change in the second-order derivative value represents the behavior of the structural deformation transitioning from linear deformation to structural instability. If the second-order derivative suddenly changes in a short period of time (exceeds the threshold), it is determined to be a sudden change in the deformation trend.

[0073] After analyzing the deformation trend characteristics, the compression deformation rate anomaly value is generated. The generation method is as follows:

[0074] Get the displacement sequence during compression from the displacement sensor: , is the displacement of the platen at the mth moment (unit: mm), Δs is the sampling time interval (unit: s); the acceleration of each point is calculated using the central difference method (excluding the first and last two dots): ; Define a sliding window of size w (e.g. w = 5 data points) and intercept each window from the acceleration sequence: ; For each window, calculate its mean , calculate the current acceleration The deviation from the window mean is used as the compression deformation rate anomaly, and the expression is: sd ; Where sd is the abnormal value of compression deformation rate.

[0075] A larger compression deformation rate anomaly indicates a dramatic, sudden change in the carton's compression deformation within a very short period of time, often reflecting significant structural instability or overall collapse. This dramatic acceleration change corresponds to actual damage behavior, demonstrating that existing detection systems are generally accurate in determining damage in these situations, with high consistency and repeatability.

[0076] Conversely, if the compression deformation rate is abnormally small, it indicates that the carton deformed slowly during compression or only experienced local, non-abrupt depressions, indicating that the structure may still be in an elastic or transitional state. If existing systems trigger "structural failure" judgments based solely on set thresholds, they may mistakenly identify such local deformations as overall failures. This often results in inaccurate or even erroneous judgments, especially in high-strength or multi-layer cartons.

[0077] The accuracy analysis module classifies the current detection status based on the combined relationship between the force value change rate characteristics and the deformation trend characteristics, and divides the judgment results of the local deformation of the carton into accuracy judgment, incomplete accuracy judgment and inaccuracy judgment, specifically including:

[0078] The force value change rate anomalies and compression deformation rate anomalies are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the accuracy score value label of the local deformation judgment result of the carton as the prediction target, and takes minimizing the sum of the prediction errors of the accuracy score value labels of the local deformation judgment result of all cartons as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence, and the model training is stopped. The accuracy score value of the local deformation judgment result of the carton is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0079] Compare the obtained accuracy score of the local deformation judgment result of the carton with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and compare the accuracy score of the local deformation judgment result of the carton with the first standard threshold and the second standard threshold respectively;

[0080] If the accuracy score of the carton local deformation judgment result is greater than the second standard threshold, it is classified as an accuracy judgment, indicating that the force value and the deformation trend are highly consistent, the judgment credibility is high, and the system accurately identifies the structural damage of the carton local deformation and can be directly adopted;

[0081] If the accuracy score of the carton local deformation judgment result is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it is classified as incomplete accuracy, indicating that there is a certain deviation in the judgment or the information is incomplete. The system judges it as incomplete and recommends entering the adaptive adjustment mechanism for secondary judgment or delayed confirmation;

[0082] If the accuracy score of the local deformation judgment result of the carton is less than the first standard threshold, it is classified as an inaccurate judgment, indicating that the force value change and deformation trend do not show the characteristics of structural damage, but the system still makes a damage judgment, which is a misjudgment or erroneous trigger. The system accuracy is unreliable and requires exception handling or parameter reset.

[0083] The adjustment module, when it is judged that the accuracy is not complete, starts the adaptive adjustment mechanism, including dynamically extending the compression time window, allowing the platen to continue pressing down and re-collecting data, specifically:

[0084] If the accuracy score S of the current carton local deformation judgment result satisfies: greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it is determined to be an incomplete accuracy judgment, and the adjustment module is triggered to start;

[0085] Set the initial extended compression time ΔT0 (e.g., 2 to 5 seconds) for additional sampling and compression of data;

[0086] At the same time, enable the delayed pressing mode of the pressing plate to keep the compression rate constant (such as 10mm / min);

[0087] Start a new round of data collection: force value Fi, displacement xi.

[0088] At fixed time intervals (e.g., every 0.5s), the latest accuracy score S(t) is recalculated to form a score curve sequence: ; Calculate the rate of change of the score in the last two time periods , the expression is: If ΔS>0, it means that the score value increases, structural information gradually emerges, and the time window continues to be extended; if ΔS≤0, the score value stagnates or decreases, and the window should be ended or an alarm should be issued.

[0089] If the score is judged to have a significant upward trend, the time window will be automatically extended: , such as increasing by 1 second each time; is the compression time after extension; if the score value has stabilized or decreased, the extension window is terminated and the maximum score value is fixed for subsequent judgment.

[0090] After the delay period, the maximum score value Smax is taken and compared with the standard threshold again;

[0091] If Smax is greater than the second standard threshold, the system update is judged to be accurate;

[0092] If it is still greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it is retained as incompletely accurate;

[0093] If the score drops to the first standard threshold, the system updates it to a misjudgment and records the anomaly.

[0094] The test result output module outputs the test results including the maximum pressure value, state classification results and pressure-displacement curve, specifically including:

[0095] After the carton compression test is complete, the test result output module outputs the key parameters collected and analyzed during the test in a structured format for user review, recording, and evaluation, as well as for subsequent quality control and production optimization. This module is directly connected to the control system's data processing unit and works in conjunction with the front-end sensor data acquisition module.

[0096] The output content and its composition include: maximum pressure value output, unit: Newton (N) or kilogram-force (kgf). Acquisition method: Take the maximum value in the force value sequence collected during the entire test process. It reflects the overall compressive strength of the carton and is an important indicator of compressive performance. Usage: Compare with product design requirements or standard limits to determine eligibility.

[0097] Status classification result output, parameter name: status label, such as "accurate", "not completely accurate" and "inaccurate"; judgment basis: combined with the abnormal value of the force value change rate and the abnormal value of the compression deformation rate, judgment is made according to the threshold gradient and combination logic; purpose: used to determine the reliability of the monitoring system, support traceability analysis or alarm control.

[0098] Pressure-displacement curve output, data structure: horizontal axis: platen displacement value xi (unit: mm); vertical axis: corresponding moment pressure value Fi (unit: N); presentation method: graphical curve (for local display or host computer GUI interface); or export in data table form (CSV, JSON format); function: visualize the mechanical response process during compression; assist in determining the structural failure mode (linear yield, sudden collapse, multi-stage deformation, etc.); support test data comparison and archiving.

[0099] Data output interface: local display (touch screen or LCD module); USB export or SD card storage; host computer communication interface (RS485, CAN, Ethernet, etc.); file format support: text data: CSV, TXT, JSON format; curve chart: PDF, JPG or display in the graphic module.

[0100] In this embodiment, a controllable vertical pressure is applied to the carton by a pressure mechanism, and the force and displacement data are collected in real time in combination with a pressure sensor and a displacement sensor. The collected data is filtered and pre-processed by a control system to extract the force value change rate characteristics and deformation trend characteristics to determine whether the carton has structural damage. During the detection process, the system classifies the accuracy of the judgment results according to the combination of the two characteristics, and introduces a scoring mechanism and a gradient standard threshold to divide the judgment results into three categories: accurate, inaccurate, and inaccurate. When it is judged to be inaccurate, the system can dynamically extend the compression time window through an adaptive adjustment mechanism, further collect data and update the judgment, and finally output the maximum pressure value, judgment status and pressure-displacement curve through the detection result output module to achieve high-precision and high-stability compression resistance detection.

[0101] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0102] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A paper packaging box strength detection device based on a pressure sensor, characterized by: It includes a pressure mechanism, a pressure sensor, a displacement sensor, a control module, an accuracy analysis module, an adjustment module and a detection result output module; A pressure mechanism, used to apply vertical pressure to the paper packaging box being inspected; Pressure sensor, used to collect real-time pressure data of paper packaging boxes during the pressure process; Displacement sensor, used to collect real-time data of the pressure plate displacement during the pressure application process; A control module, wherein the control module pre-processes the collected pressure data and displacement data, and extracts force value change rate characteristics and deformation trend characteristics respectively; An accuracy analysis module, based on the combined relationship between the force value change rate feature and the deformation trend feature, classifies the current detection state and divides the judgment result of the local deformation of the carton into accuracy judgment, partial accuracy judgment and inaccuracy judgment; The adjustment module, when it is judged that the accuracy is not complete, activates the adaptive adjustment mechanism, including dynamically extending the compression time window, allowing the platen to continue pressing down and re-collecting data; The test result output module outputs the test results including the maximum pressure value, state classification results and pressure-displacement curve.

2. The paper packaging box strength detection device based on a pressure sensor according to claim 1, characterized in that: The pressure mechanism includes an upper pressure plate and a lower pressure plate. The upper pressure plate is driven by an electric screw drive device or a servo motor system and moves in a vertical direction under the guidance of a guide mechanism to apply controllable pressure to the carton.

3. The paper packaging box strength detection device based on a pressure sensor according to claim 1, characterized in that: After analyzing the force value change rate characteristics, the force value change rate abnormal value is generated. The generation method is as follows: the discrete sampling points of the pressure sensor during the test are recorded as: ;in: is the pressure value at the nth moment, calculate the pressure change rate at each moment: ;in: is the pressure change rate at the i-th time point, Δt is the time interval between two consecutive sampling points, and the result is a change rate sequence: ; is the pressure change rate at the nth time point, and the mean of the change rate is calculated and standard deviation ; Calculate the Z-score value, the expression is: ;like , is the abnormal rate point, where is the Z-score value of the i-th rate point, A is the abnormal threshold coefficient, and traverses all , record satisfaction Time point , which is the potential abnormal point, and the maximum pressure change rate of all potential abnormal points is taken as the abnormal value of the force change rate.

4. The paper packaging box strength detection device based on a pressure sensor according to claim 3, characterized in that: After analyzing the deformation trend characteristics, the compression deformation rate anomaly value is generated by obtaining the displacement sequence during the compression process from the displacement sensor: , is the displacement of the platen at the mth moment, Δs is the sampling time interval; the acceleration of each point is calculated using the central difference method : ; Define a sliding window of size w and intercept each window from the acceleration sequence: ; For each window, calculate its mean , calculate the current acceleration The deviation from the window mean is used as the compression deformation rate anomaly, and the expression is: sd ; Where sd is the abnormal value of compression deformation rate.

5. The paper packaging box strength detection device based on a pressure sensor according to claim 4, characterized in that: The accuracy analysis module classifies the current detection status based on the combined relationship between the force value change rate characteristics and the deformation trend characteristics, and divides the judgment results of the local deformation of the carton into accuracy judgment, incomplete accuracy judgment and inaccuracy judgment, specifically including: The force value change rate anomalies and compression deformation rate anomalies are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the accuracy score value label of the local deformation judgment result of the carton as the prediction target, and takes minimizing the sum of the prediction errors of the accuracy score value labels of the local deformation judgment result of all cartons as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence, and the model training is stopped. The accuracy score value of the local deformation judgment result of the carton is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

6. The paper packaging box strength detection device based on a pressure sensor according to claim 5, characterized in that: Compare the obtained accuracy score of the local deformation judgment result of the carton with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and compare the accuracy score of the local deformation judgment result of the carton with the first standard threshold and the second standard threshold respectively; If the accuracy score of the carton local deformation judgment result is greater than the second standard threshold, it is classified as an accuracy judgment and can be directly adopted; If the accuracy score of the carton local deformation judgment result is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it is classified as an incomplete accuracy judgment and enters the adaptive adjustment mechanism for secondary judgment or delayed confirmation; If the accuracy score of the carton local deformation judgment result is less than the first standard threshold, it is classified as an inaccuracy judgment and abnormal processing or parameter reset is performed.

7. The paper packaging box strength detection device based on a pressure sensor according to claim 6, characterized in that: The adjustment module, when it is judged that the accuracy is not complete, starts the adaptive adjustment mechanism, including dynamically extending the compression time window, allowing the platen to continue pressing down and re-collecting data, specifically: Set the initial extended compression time ΔT0 for additional sampling of compressed data; At the same time, the delay mode of pressing the plate is enabled to keep the compression rate constant; Start a new round of data collection; At fixed time intervals, the latest accuracy score S(t) is recalculated to form a score curve sequence: ; Calculate the rate of change of the score in the last two time periods , the expression is: If ΔS>0, it means the score value is rising, structural information is gradually emerging, and the time window is extended; if ΔS≤0, the score value is stagnant or decreasing, and the window is ended or an alarm is issued; If the score is judged to have a significant upward trend, the time window will be automatically extended: ; is the compression time after extension; If the score value has stabilized or decreased, the extended window is terminated and the maximum score value is fixed for subsequent judgment.

Images