Production quality management method for aluminum trim locking assembly of passenger car

Abstract

The application discloses a production quality management method for an aluminum ornament locking assembly of a passenger car, and belongs to the technical field of production quality management, and specifically comprises the following steps: firstly, collecting locking force performance data and synchronous production line operation beat data of historical production batches, mapping and binding the time series data set containing the associated information of quality state and production intensity on a time axis; then, constructing a dynamic response surface of quality to production intensity based on the data set, identifying a quality mutation inflection point to determine a critical production rhythm threshold; then, generating a floating process control boundary according to the threshold, and dynamically translating a locking force early warning interval; comparing the locking force detection value with the floating boundary in real time, outputting a compensation instruction to fine tune the forming pressure when the warning interval is approached; and finally, collecting quality response data after compensation and returning, and iteratively correcting the dynamic response surface and the floating control boundary.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of production quality management, and in particular to a production quality management method for an aluminum trim part locking assembly for a passenger vehicle. BACKGROUND

[0002] The aluminum trim part locking assembly for a passenger vehicle is a key connecting component of the interior system of the whole vehicle, and its locking force performance directly determines the connection reliability and the safety of the whole vehicle. The related production quality management has become one of the core links of industry concern. At present, for the production quality management of the locking assembly, the existing technology generally adopts basic control means such as performance data collection, quality standard setting, real-time detection comparison and process parameter adjustment, realizes the preliminary monitoring of the locking force performance in the production process by constructing fixed quality control threshold and early warning interval, and provides certain technical support for guaranteeing the basic quality and production efficiency of the product.

[0003] However, the current quality management method in the production of the locking assembly fails to fully consider the overall trend of the quality performance changing dynamically with the production rhythm and organizational mode, resulting in lack of flexibility and adaptability in quality control. The existing method is often based on static or fixed quality control standards, and cannot capture the quality fluctuations caused by production adjustment, resource allocation or process change in the manufacturing process in real time, so it is difficult to form a stable and continuous quality management mechanism. This limitation makes the quality control vulnerable to changes in the production environment, and cannot guarantee the long-term quality stability, especially in multi-batch production, the quality maintenance may be interrupted, affecting the product reliability and the long-term quality guarantee basis of the whole vehicle use link. SUMMARY

[0004] The purpose of the present application is to provide a production quality management method for an aluminum trim part locking assembly for a passenger vehicle, which solves the problems in the background art:

[0005] The purpose of the present application can be achieved by the following technical solutions:

[0006] A production quality management method for an aluminum trim part locking assembly for a passenger vehicle, comprising the following steps:

[0007] S1: collecting locking force performance data and synchronous production line running rhythm data of historical production batches of the aluminum trim part locking assembly for a passenger vehicle; mapping and binding the locking force performance data and the running rhythm data on the time axis to generate a time series data set containing quality state and production intensity association information;

[0008] S2: constructing a dynamic response surface of the quality performance of the locking assembly relative to the production intensity by using the time series data set; identifying the inflection point where the quality index on the dynamic response surface occurs mutation, and determining the critical production rhythm threshold for maintaining quality stability under the current production organization mode;

[0009] S3: generating a floating process control boundary that changes in real time with the running pace of the production line according to the critical production rhythm threshold; during the acceleration or deceleration of production, the upper limit value and the lower limit value of the locking force early warning interval are automatically translated to build a dynamic quality monitoring range that matches the current production capacity;

[0010] S4: comparing the real-time detected locking force value of the locking assembly with the floating process control boundary; when the detection value approaches the warning interval of the floating boundary, outputting a process parameter compensation instruction to the production equipment control unit to offset the quality fluctuation tendency caused by the change of production rhythm by fine-tuning the molding pressure;

[0011] S5: collecting the quality response data of the locking assembly after the process parameter compensation and returning to the dynamic response surface model; using the response data to correct the dynamic response surface and the subsequent floating process control boundary, and completing the iterative update of the correlation between production rhythm and quality performance.

[0012] As a further scheme of the application: in step S1, the process of mapping and binding the locking force performance data and the running pace data on the time axis to generate a time series data set containing the quality state and production intensity correlation information is:

[0013] reading the production line running pace data and the locking force performance data from the production line main controller and the locking force detection unit respectively; using the high-precision network time protocol of industrial Ethernet to synchronize the clocks of the two independent data sources, and marking each collected data with a nanosecond-level time stamp;

[0014] Taking the time stamp sequence of the production line running pace data as the reference coordinate, mapping the locking force performance data into the corresponding pace period according to the time stamp proximity principle; by comparing the consistency of the time mark, the discrete locking force detection value is logically associated and bound with the continuously changing production line running pace data at that moment;

[0015] Merging and storing the running pace value and the locking force performance data that have completed logical association; constructing a multi-dimensional structured data table with time stamp as the only index key value, and outputting a time series data set containing the quality state and production intensity correlation information of the corresponding relationship between the production intensity input and the quality state output at each moment.

[0016] As a further scheme of the application: in step S2, the process of constructing the dynamic response surface of the locking assembly quality performance with respect to the production intensity using the time series data set; identifying the inflection point where the quality index on the dynamic response surface changes abruptly, and determining the critical production rhythm threshold that maintains the quality stability under the current production organization method is:

[0017] Extract production intensity data from the time series data set as the independent variable and the locking assembly quality performance data as the dependent variable; use the least squares method to nonlinearly fit the independent variable and the dependent variable, and establish a dynamic response surface mathematical model describing the continuous change of the quality index with the production intensity;

[0018] Second-order derivatives are calculated with respect to the production intensity variable of the established dynamic response surface model; the curvature values of each discrete point on the surface are calculated to generate a curvature distribution curve reflecting the sensitivity of the locking assembly quality performance to the change in production intensity;

[0019] The coordinate position where the curvature value in the curvature distribution curve presents a local maximum is searched; the coordinate position is marked as an inflection point where the quality index changes abruptly, which is a critical dividing point for distinguishing between stable production state and quality nonlinear fluctuation state;

[0020] The identified inflection point coordinate is orthogonally projected to the production intensity coordinate axis of the dynamic response surface; the specific production intensity value corresponding to the projection point is read and locked as the critical production rhythm threshold for maintaining quality stability under the current production organization mode.

[0021] As a further scheme of the present application, the process of calculating the second-order derivative of the established dynamic response surface model with respect to the production intensity variable and calculating the curvature values of each discrete point on the surface to generate a curvature distribution curve reflecting the sensitivity of the locking assembly quality performance to the change in production intensity is as follows:

[0022] The function polynomial expression of the dynamic response surface mathematical model is analyzed by calling the numerical calculation unit; the second-order differential operation is performed on the production intensity variable in the expression, and the second-order derivative function analytical expression describing the gradient change rate of the quality index is derived;

[0023] The production intensity coordinate values of each discrete point on the dynamic response surface model are substituted into the second-order derivative function analytical expression; combined with the first-order derivative value of the corresponding point and the differential geometry curvature calculation formula, the geometric curvature value at each discrete point is calculated;

[0024] An analysis coordinate system with production intensity as the horizontal axis and geometric curvature value as the vertical axis is established; the geometric curvature value is mapped to the coordinate system and connected by using a smooth interpolation fitting algorithm to generate a curvature distribution curve reflecting the continuous change in sensitivity.

[0025] As a further scheme of the present application, in step S3, the process of generating a floating process control boundary that changes in real time with the production line running tempo according to the critical production rhythm threshold is as follows:

[0026] Collecting real-time production line running tempo and calculating load difference ratio of the tempo and critical production tempo threshold value; substituting the load difference ratio into preset boundary compensation algorithm to obtain boundary translation correction value adapted to current production load state through calculation;

[0027] Retrieving basic static control interval data of the locking assembly; performing same-direction weighted offset calculation on upper limit value and lower limit value of the basic static control interval by using the boundary translation correction value to obtain instant coordinate value of the dynamic interval;

[0028] Continuously outputting the calculated instant coordinate value of the dynamic interval to the process monitoring module in time sequence; forming the floating process control boundary which is real-time linked with the production line running tempo in time domain by continuously updating the instant coordinate value.

[0029] As a further scheme of the application, in the step S3, the process of automatically translating the upper limit value and the lower limit value of the locking force early warning interval and constructing the dynamic quality monitoring range matched with the current production load capacity in the production acceleration or deceleration process is:

[0030] Calculating real-time differential value of the production line running tempo with respect to time to obtain tempo change rate; determining the current production acceleration state or production deceleration state according to the positive or negative sign of the tempo change rate and quantifying the specific change intensity synchronously;

[0031] Substituting the quantified change intensity into the preset tempo-deviation compensation function model to perform calculation; and analytically obtaining the interval translation compensation value which is in linear mapping relationship with the current acceleration or deceleration trend;

[0032] Stacking and correcting the interval translation compensation value to the current upper limit value and the current lower limit value of the floating process control boundary; locking the value interval after the stacking calculation as the final dynamic quality monitoring range adapted to the current production load capacity.

[0033] As a further scheme of the application, in the step S4, the process of outputting the process parameter compensation instruction to the production equipment control unit to offset the quality fluctuation tendency caused by the production tempo change by fine-tuning the molding pressure is:

[0034] Calculating deviation value of the real-time locking force value with respect to the center line of the floating process control boundary; and inversely calculating the molding pressure compensation value required for correcting the current deviation in combination with the deformation modulus of the locking assembly material;

[0035] Encoding the molding pressure compensation value into parameter correction instruction special for the production equipment control unit; and preferentially sending the instruction to the bottom programmable logic controller responsible for performing the pressing action through the high-speed industrial bus network;

[0036] The programmable logic controller dynamically adjusts the output power of the press-fitting actuator according to the instruction; and directly outputs the changed forming pressure in the subsequent press-fitting stroke to physically correct the quality data deviation of the locking assembly.

[0037] As a further scheme of the present application, in the step S5, the process of completing the iterative update of the correlation between the production rhythm and the quality performance is as follows:

[0038] The quality response data of the locking assembly after the process parameter compensation and the corresponding production line operation rhythm data are imported into the original time sequence data set; the latest data is given a higher calculation weight than the historical data, and a weighted update data set focusing on reflecting the recent equipment state is generated;

[0039] The nonlinear relationship between the production intensity and the quality performance is fitted again by using the weighted update data set to generate a corrected dynamic response surface; the curvature extreme point is recalculated to locate the inflection point coordinates reflecting the sudden change of the quality index of the current equipment capacity;

[0040] The critical production rhythm threshold is refreshed according to the new inflection point coordinates, and the subsequent floating process control boundary is reconstructed; the updated boundary parameters are sent to the process monitoring module to cover the old standard, thereby completing the iterative update of the correlation between the production rhythm and the quality performance.

[0041] The present application has the following beneficial effects:

[0042] Based on the above technical scheme, the present application has the following beneficial effects:

[0043] By constructing the dynamic response surface of the quality performance relative to the production intensity, the present application breaks through the limitation that the traditional static quality control standard cannot adapt to flexible production rhythm. The inflection point of the sudden change of the quality index is accurately identified by using the time sequence data set, the critical production rhythm threshold under the current equipment capacity and process state can be scientifically defined, and the quality monitoring range is upgraded from the rigid fixed numerical interval to the floating process control boundary linked in real time with the production line operation rhythm. This mechanism can automatically adapt to the inertia load change in the process of production acceleration, deceleration or steady high-speed operation, effectively solving the problem that the locking force of the aluminum decoration part fluctuates due to the adjustment of production organization mode, the switching of rhythm speed and is difficult to be captured by the traditional standard, ensuring the accuracy and physical applicability of the quality judgment standard under different production loads, and realizing the dynamic balance between the improvement of production efficiency and the maintenance of quality stability.

[0044] The application establishes a closed-loop active compensation mechanism based on real-time detection and model iteration, which significantly improves the long-term stability and system adaptive ability of quality management. By actively outputting process parameter compensation instructions to the equipment when the detection value approaches the warning interval of the floating boundary, the quality deviation tendency caused by rhythm changes can be offset by fine-tuning the molding pressure, and passive post-screening is transformed into active pre-intervention. More importantly, the dynamic response surface and control boundary are continuously corrected using the response data after process compensation, so that the management method has self-learning and self-evolution ability, and can automatically digest the cumulative errors caused by equipment aging, mold wear or environmental drift. This iterative updating mechanism not only guarantees the high consistency of the locking assembly in long-period and multi-batch production, but also provides a solid bottom process guarantee for the connection reliability of aluminum trim parts in the vehicle use link. BRIEF DESCRIPTION OF DRAWINGS

[0045] The application will be further described below in conjunction with the accompanying drawings.

[0046] Figure 1 is a flowchart of a production quality management method for an aluminum trim part locking assembly of a passenger vehicle. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the application will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0048] Please refer to Figure 1 The application is a production quality management method for an aluminum trim part locking assembly of a passenger vehicle, which comprises the following steps:

[0049] S1: Collect the locking force performance data of the historical production batches of the aluminum trim part locking assembly of the passenger vehicle and the synchronous production line running rhythm data; map and bind the locking force performance data and the running rhythm data on the time axis to generate a time series data set containing the associated information of quality state and production intensity;

[0050] S2: Construct a dynamic response surface of the quality performance of the locking assembly relative to the production intensity using the time series data set; identify the inflection points where the quality indicators on the dynamic response surface mutate, and determine the critical production rhythm threshold that maintains quality stability under the current production organization method;

[0051] S3: generating a floating process control boundary that varies in real time with the running pace of the production line according to the critical production rhythm threshold; during the acceleration or deceleration of production, automatically translating the upper and lower limit values of the locking force early warning interval to build a dynamic quality monitoring range that matches the current production capacity;

[0052] S4: comparing the real-time detected locking force value of the locking assembly with the floating process control boundary; when the detection value approaches the warning interval of the floating boundary, outputting a process parameter compensation instruction to the production equipment control unit to offset the quality fluctuation tendency caused by the change of production rhythm by fine-tuning the molding pressure;

[0053] S5: collecting the quality response data of the locking assembly after the process parameter compensation and returning it to the dynamic response surface model; using the response data to correct the dynamic response surface and the subsequent floating process control boundary, and completing the iterative update of the correlation between production rhythm and quality performance.

[0054] In one embodiment of the present application, in step S1, the process of mapping and binding the locking force performance data and the running pace data on the time axis to generate a time series data set containing the correlation information between quality state and production intensity is:

[0055] First, carry out the basic data physical collection work of the aluminum decoration locking assembly for the vehicle. In the face of the large batch continuous manufacturing scene of the aluminum decoration locking assembly for the vehicle, the key performance indicator data covering the complete historical production batch is obtained as the analysis benchmark. The specific operation process selects the equipment operation records in the past 3 to 6 months as the sample source, and focuses on the locking force performance data of the locking assembly after the completion of the press fitting and the real-time running pace data of the production line at that time. The locking force performance data is directly read by a high-precision mechanical sensor installed at the end of the production line, and the actual holding force value of each single product is recorded in Newton, which reflects the final delivery quality of the product. The production line running pace data is obtained by encoder or counter, which reflects the current output efficiency and equipment running speed, usually measured in seconds or per minute. During the collection process, the integrity of the data should be ensured, and the samples should cover various periods from the low-speed trial production stage to the high-speed full-load running stage, so as to provide sufficient sample space for establishing the correlation model of quality and speed. These historical data not only contain the parameters of qualified products, but also contain the data of part of the edge products or unqualified products, so as to fully reflect the actual influence of production intensity change on product quality, and lay a solid data foundation for establishing a precise control model.

[0056] The main controller and the end locking force detection unit of the production line are connected by an industrial Ethernet bus to capture the data streams generated by the two independent sources in real time. To solve the problem of possible deviation of the internal clock oscillator of different devices, the IEEE 1588 precision time protocol (PTP) is used to synchronize the time of all nodes in the network. The master clock node sends synchronization messages to the slave clock node, calculates and automatically compensates for network transmission delay and clock drift, thereby strictly controlling the clock deviation between the main controller and the detection unit to the sub-microsecond level. On the basis of this high-precision synchronization, each running beat data read from the main controller and the locking force performance data read from the detection instrument are marked with nanosecond-level absolute time stamps. This high-precision time marking can accurately record the physical time of each signal generation, completely eliminate the risk of time sequence disorder caused by signal transmission delay or processing queue accumulation, and ensure the strict alignment of multi-source heterogeneous data in the time dimension, providing a unique physical reference benchmark for accurate data correlation.

[0057] Since the production line running beat data is usually a continuous change of analog quantity or high-frequency sampling sequence with the acceleration and deceleration of the equipment, and the locking force performance data is discrete event data based on single product output, it is necessary to use high-frequency running beat data timestamp sequence as the reference coordinate axis. The processing unit traverses the nanosecond-level timestamp of each locking force data, and searches for the time point closest to its time marker on the time axis of the running beat through binary search or sliding window algorithm. By comparing the consistency of the two time markers, the discrete locking force detection value is logically matched with the instantaneous running speed of the production line at that specific time point. If the difference between the two time stamps is within the preset minimum tolerance range, it is determined that they belong to the same production action cycle, thereby establishing a one-to-one logical correlation. This binding method can accurately restore the actual production intensity experienced by the product at the moment of forming, effectively eliminate the errors caused by relying only on average speed calculation, realize the precise hooking of quality data and process rhythm at the microscopic level, and ensure that each quality data point can find its speed background at the time of generation.

[0058] The processing unit opens a dedicated storage space in the database, merges the running beat values that have been successfully paired with the locking force performance data, and processes them. A multi-dimensional structured data table is constructed with a nanosecond timestamp as the only index key value. The table includes a serial number, absolute time, instantaneous production beat, measured locking force, and environmental temperature, etc. Each row of data strictly corresponds to a specific production time, clearly showing the production intensity input conditions and the corresponding quality state output results at that time. After cleaning and formatting, the data table finally forms a standardized time series data set, which can be directly called by the mathematical modeling unit for analyzing the specific influence of production rhythm on quality stability. In this way, the originally disorganized and independent device parameters are transformed into high-value information assets with strong correlation characteristics, which can intuitively reflect the dynamic evolution trend of product performance under different load conditions of the production line, providing reliable data support for intelligent process parameter compensation, and also providing a quantitative basis for long-term evaluation of device capacity.

[0059] In an embodiment of the present application, in step S2, a dynamic response surface of the locking assembly quality performance with respect to the production intensity is constructed using the time series data set; and the process of identifying the inflection point where the quality index changes abruptly on the dynamic response surface to determine the critical production rhythm threshold for maintaining quality stability under the current production organization mode is as follows:

[0060] First, the extraction and variable definition of production intensity data and locking assembly quality performance data are performed. The data processing unit directly accesses the previously constructed time series data set, accurately filters out the production line running beat values as the independent variable data sequence, and extracts the locking force detection values corresponding to the same timestamp as the dependent variable data sequence. In this process, the processing unit pre-processes the original data to eliminate invalid null values caused by sensor failure or network packet loss, ensuring that each set of data input into the model has complete physical meaning. The independent variable represents the workload and operating speed of the production equipment, which is the active input parameter in process control, while the dependent variable objectively reflects the physical performance of the product delivered under this load condition, which is the passive output result of the process. By aligning the two sets of data in the mathematical space, the processing unit prepares the data for establishing the analytical relationship between the two, aiming to extract the inherent law of the influence of device operating speed on product quality from a large number of discrete production records, and thus to transform the complex physical production process into a numerical analysis object that can be operated mathematically.

[0061] The least square method is used to extract the independent variables and dependent variables for nonlinear fitting, which is the core step of constructing a mathematical model. The calculation module selects a polynomial function of appropriate order as the basic model architecture, and determines the optimal solution of each coefficient in the polynomial through iterative operation. In the operation process, the calculation module calculates the residual sum of squares between each actual observation point and the fitting surface, and adjusts the function coefficients to minimize the residual sum of squares, so as to obtain a continuous curve or surface that can best approximate the historical data distribution trend. This fitting process effectively smooths the random mass fluctuation noise caused by small differences in materials or environmental vibration, and extracts the main evolution trend between production intensity and locking quality. The dynamic response surface mathematical model established is no longer a set of discrete numbers, but a continuous and differentiable function expression, which can predict the corresponding theoretical locking force value at any given production cycle value point.

[0062] The second-order derivative of the established dynamic response surface model with respect to the production intensity variable is calculated; the curvature values of each discrete point on the surface are calculated to generate a curvature distribution curve reflecting the sensitivity of the locking assembly quality performance to changes in production intensity; the specific process is as follows:

[0063] The numerical calculation unit first starts the deep analysis program for the dynamic response surface mathematical model. The unit reads the polynomial function expression that has been established from the storage module, accurately identifies the symbol representing the production intensity, the core independent variable in the expression, and the corresponding constant coefficients. According to the basic principles of calculus, the calculation unit performs strict second-order differentiation operation on the production intensity variable in the function expression. This operation process first performs the first derivative operation on the original function, thereby obtaining the first-order derivative expression describing the rate of change of the quality index with the production intensity, and then performs the second derivative on the first-order derivative, and finally derives the second-order derivative function expression describing the gradient change rate of the quality index. This analytical expression represents the concave-convex feature of the function curve in the mathematical level, and reveals the acceleration characteristics of the locking assembly quality performance fluctuation with the production rhythm change in the physical level. Through this analytical operation, the original static surface model is transformed into a mathematical tool that can dynamically describe the change intensity, providing a bottom-level mathematical logic support for accurately capturing the quality mutation characteristics, and ensuring that the process sensitivity analysis is based on rigorous mathematical derivation.

[0064] The computing module then performs large-scale numerical substitution and geometric operation tasks to obtain the curvature characteristics of the discrete points. It traverses each discrete sampling point within the domain of the dynamic response surface model, extracts the production intensity coordinate values corresponding to these points as input parameters, substitutes these values into the previously derived second-order derivative function and first-order derivative function analytical expressions, and thus accurately obtains the second-order derivative value and first-order derivative value at the specific point. The computing module uses the obtained first-order derivative and second-order derivative to construct an operation relationship according to the standard plane curve curvature calculation formula in differential geometry, with the denominator of the relationship being the square root of one plus the square of the first-order derivative, and the numerator being the absolute value of the second-order derivative. Through this complex algebraic operation, the computing module obtains the geometric curvature value at each discrete point. This value not only reflects the size of the second-order derivative, but also takes into account the influence of the tangent slope, thus obtaining a normalized geometric quantity that can accurately quantify the severity of the bending deformation of the locking assembly quality indicator under the corresponding production intensity, providing quantitative data indicators for identifying potential process risk points.

[0065] A two-dimensional rectangular coordinate system dedicated to sensitivity analysis is constructed to visualize the data results. The coordinate system sets the horizontal axis to represent the production intensity variable and the vertical axis to represent the geometric curvature value. All discrete geometric curvature values calculated in the previous step are mapped and projected into this analysis coordinate system according to their corresponding production intensity coordinates, forming a series of data points scattered on the coordinate plane. To obtain a continuous and intuitive trend view, a cubic spline interpolation algorithm or a Gaussian smoothing fitting algorithm is used to connect these discrete points. This algorithm can complete the missing information between points while ensuring data fidelity, ultimately generating a smooth and continuous curvature distribution curve. This curve intuitively reflects the distribution of the sensitivity of the locking assembly quality performance to the production intensity change within the entire production speed range, with the peak region on the curve clearly indicating the extremely unstable speed range of the process, and the flat region corresponding to the relatively robust safe range of the process, thus converting the complex differential calculation results into graphical decision-making basis that engineers can easily understand.

[0066] Retrieving the coordinate position of the local maximum of the curvature value in the curvature distribution curve is a key step for identifying the process state transition. The generated curvature distribution curve is scanned globally, and an extreme value search algorithm is used to find the coordinate point at the wave peak of the curve. These local maximum points of curvature correspond to the inflection points in the physical sense, which means that the variation law of the quality index with the production intensity has fundamentally changed at this point. On the left side of the inflection point, the production is in a relatively stable linear control area, and the quality fluctuation is small, while on the right side of the inflection point, the production may enter a nonlinear oscillation area, and the quality will deteriorate sharply with the increase of speed. The point with the maximum curvature value is marked as the most critical mutation inflection point, and it is defined as the critical dividing point between the stable production state and the nonlinear quality fluctuation state. This determination process is completely based on the objective calculation of mathematical characteristics, avoiding the subjective error caused by relying on artificial experience, and accurately capturing the critical moment of the qualitative change of device performance.

[0067] In the final stage, the critical production rhythm threshold under the current production organization mode is determined through a geometric projection operation. The processing unit orthogonally projects the identified quality index mutation inflection point coordinate in the mathematical space to the horizontal coordinate axis representing the production intensity. The intersection point of the projection line and the production intensity coordinate axis is the specific production beat value corresponding to the inflection point. The processing unit reads this value and locks it as the critical production rhythm threshold, which represents the upper limit of the maximum safe production speed under the combined action of the current device state, mold wear degree and material characteristics. Once the actual production speed exceeds this value, the product quality will face a high risk of losing control. This threshold is stored in the process parameter database as a rigid constraint condition for guiding future production scheduling and speed setting. In this way, the originally abstract mathematical model analysis result is transformed into specific process parameters that can be directly identified and executed by workshop managers and device controllers, realizing the closed-loop landing from data analysis to engineering application.

[0068] In an embodiment of the present application, in the step S3, the process of generating a floating process control boundary that changes in real time with the production line running beat according to the critical production rhythm threshold is:

[0069] Firstly, the collection of real-time production data and the quantitative calculation of load state are carried out. The data processing unit reads the actual running tact value of the production line at the current time through the high-speed communication interface connected with the line controller. The value accurately reflects the instantaneous processing speed and mechanical load state of the current equipment. The processing unit calls the critical production rhythm threshold value stored in the memory which has been determined in the previous stage to carry out numerical comparison operation with the real-time running tact value collected. Through specific division and difference operation logic, the processing unit calculates the approximation degree of the current tact relative to the critical threshold value, thereby obtaining a dimensionless load difference ratio. The ratio intuitively quantifies how much safety buffer space the current production state is away from the dangerous edge of quality mutation. Then, the processing unit substitutes the load difference ratio as a key input parameter into the preset boundary compensation algorithm model. The algorithm integrates a conversion function based on the material mechanics characteristics and equipment dynamics characteristics, which can map the abstract load ratio value to a specific physical mechanics compensation amount. After a series of floating point operations, the algorithm finally outputs a boundary translation correction value adapted to the current production load state. The correction value is in Newton units, representing the theoretical offset of the control interval required to maintain process stability at the current speed.

[0070] Next, the dynamic correction calculation operation of the static control interval of the locking assembly is performed. The calculation module retrieves the basic static control interval data of the locking assembly of this type measured in the standard laboratory environment from the process database. The data includes the upper limit threshold and the lower limit threshold of the static force value that meets the quality acceptance standard. The calculation module uses the boundary translation correction value obtained in the previous step to perform the same weighted offset calculation on the two basic static thresholds. This means that if the correction value is positive, the upper limit and the lower limit will move in the same direction to increase the value, and vice versa, to move in the direction of decreasing the value, thereby keeping the bandwidth of the control interval relatively constant or making minor adjustments according to the specific weight. During the calculation, different weighting coefficients are assigned to the upper limit and the lower limit according to the process characteristics to reflect the different effects of speed changes on the extreme boundaries. Through this real-time algebraic superposition operation, the calculation module converts the originally fixed static standard into a set of dynamic interval coordinate values that can adapt to the current working condition. This set of values is no longer a rigid inspection red line, but has physical meaning that matches the current equipment operating state, accurately defining the reasonable mechanical range for determining product eligibility at the current instant.

[0071] The last stage builds a dynamic process monitoring envelope that changes over time through continuous data stream output. The processing unit sends the calculated dynamic interval real-time coordinate values in strict time series format to the process monitoring module or visual human-computer interaction interface continuously through the high-speed bus. With every millisecond change in the production line's running tempo, this set of upper and lower limit coordinate values is constantly refreshed and reconstructed. In the time domain coordinate system, this series of continuously updated real-time coordinate values are connected to form a floating process control boundary similar to a strip-shaped channel. This boundary is like the skin that clings to the production rhythm. When the production line speeds up, the boundary will automatically shift upwards or downwards to accommodate reasonable fluctuations caused by inertia. When the production line slows down, the boundary will quickly fall back to tighten the quality control standards.

[0072] In one embodiment of the present application, the step S3, during the acceleration or deceleration of production, automatically shifts the upper limit value and the lower limit value of the locking force warning interval to build a dynamic quality monitoring range that matches the current production load capacity, is:

[0073] First, differential calculation and trend identification are performed for changes in the production line running state. The calculation module continuously reads the real-time value stream of the production line running tempo at a millisecond level time interval, and uses numerical differentiation algorithms to perform real-time operation on this set of data sequences that change continuously over time. This operation process calculates the difference between the current time tempo value and the last sampling time tempo value, and divides it by a very short time interval increment, to accurately obtain the real-time differential value of the production line running tempo with respect to time. This value is the change rate of the tempo in physical terms. The calculation module determines the current dynamics state of the device according to the positive and negative signs of this differential value. If the differential value is positive, it may represent a tempo extension, i.e. a deceleration state, and if the differential value is negative, it represents a tempo shortening, i.e. an acceleration state, thereby accurately identifying the current production acceleration state or production deceleration state. At the same time, the calculation module takes the absolute value of this differential value, thereby stripping the directional information and simply quantifying the specific change intensity value.

[0074] Next, the operation of converting the speed change intensity into a mechanical compensation value using a preset function model is performed. The processing unit calls a rhythm and deviation compensation function model previously constructed in the memory, which is a linear or nonlinear relational expression fitted based on a large amount of historical mechanical experimental data, and is specially used to describe the specific influence of the inertial force generated by the acceleration and deceleration of the device on the forming quality of the locking assembly. The processing unit inputs the quantified variation intensity value obtained in the previous step into the function model as the independent variable, and through the execution of algebraic operations such as multiplication and addition, combined with the inherent stiffness coefficient and damping coefficient in the model, the interval translation compensation value strictly linearly mapped with the current acceleration or deceleration trend is analytically obtained. The compensation value is in newtons or megapascals as a physical unit, and its numerical size is directly proportional to the speed change of the device. For example, in the case of rapid acceleration, the model will calculate a larger positive compensation value to offset the pressure loss caused by the machine vibration, and in the case of slow deceleration, a smaller negative compensation value is calculated.

[0075] In the final stage, the dynamic correction and final locking operation of the floating process control boundary are completed. The control module obtains the calculated interval translation compensation value and applies it to the floating process control boundary at the current time. The specific operation is to add the compensation value to the current upper limit value and the current lower limit value of the floating process control boundary respectively, and perform an algebraic weighted calculation of the same frequency and amplitude. Through this superposition correction, the originally only speed absolute value changing floating boundary is again shifted based on the acceleration, making the monitoring interval show a very elastic dynamic adaptation characteristic on the time axis. The control module then locks the value interval after superposition calculation, and establishes it as the final dynamic quality monitoring range adapted to the current production load capacity. This range is a special standard for the current millisecond instant, which can perfectly contain the reasonable quality data fluctuations caused by the inertia of the device acceleration and deceleration, and at the same time can sensitively intercept the real abnormal defects.

[0076] In an embodiment of the present application, in the step S4, the process of outputting a process parameter compensation instruction to the production device control unit to offset the quality fluctuation tendency caused by the change of production rhythm by fine-tuning the molding pressure is as follows:

[0077] Firstly, the mathematical calculation and physical quantity inverse calculation operation for real-time deviation are performed. The calculation module first acquires the real-time value of the locking force detected at the current time, and reads the upper limit coordinate and the lower limit coordinate of the floating process control boundary at the current time. The calculation module constructs a virtual process center line by calculating the arithmetic mean of the upper limit and the lower limit, and the center line represents the ideal quality target value under the current production rhythm. Then, the real-time locking force value is subtracted from the process center line to obtain a deviation value with positive and negative signs, which accurately quantifies the magnitude of the current product quality deviation from the ideal state. In order to convert this quality deviation into a control quantity that can be executed by the equipment, the calculation module introduces the modulus of deformation of the locking component material as a key conversion parameter. According to the stress-strain relationship in material mechanics and the generalized form of Hooke's law, the calculation module establishes an inverse calculation model. The model takes the deviation value as input, combines the elastic modulus and plastic deformation characteristic parameters of the material, and inversely deduces the additional forming pressure required to eliminate the deviation. After a series of floating point operations, the module finally outputs the forming pressure compensation value required to correct the current deviation, which is accurate to two decimal places, providing a clear mechanical adjustment target for the actuator.

[0078] Next, the encoding and transmission task of converting the physical compensation value into industrial control instructions is performed. After receiving the calculated forming pressure compensation value, the communication processing module first digitizes it according to the standard format of the underlying device communication protocol. The module converts the analog pressure value into a hexadecimal control word, adds a frame header, a frame trailer, and a cyclic redundancy check code, and encapsulates it as a parameter correction instruction data packet dedicated to the production equipment control unit. In order to ensure that the instruction can be responded in time, the communication module establishes a dedicated transmission channel using a high-speed industrial bus network. In the network protocol stack, the communication module marks the parameter correction instruction as a real-time data frame with the highest priority, so that it can preempt the bandwidth and surpass the transmission queue of other non-critical state data. Through this instant communication mechanism, the instruction is sent to the underlying programmable logic controller responsible for the pressing action in priority.

[0079] The last stage is executed by programmable logic controller to perform physical level power adjustment and quality correction action. The underlying programmable logic controller immediately triggers the internal interrupt service program to parse and verify the instruction after receiving the high priority parameter correction instruction. The controller adjusts the output power signal of the proportional valve or servo driver connected to the press assembly actuator through the digital analog conversion interface according to the pressure compensation value contained in the instruction. This adjustment is not just a simple on-off control, but a fine linear trimming of the output voltage or current. At the next press stroke start moment, the press assembly actuator will output the corrected forming pressure according to the changed power parameter. This new pressure value directly acts on the new aluminum trim part blank, changing the material compression deformation through physical means. This real-time physical intervention effectively offsets the quality fluctuation tendency caused by changes in production rhythm, forcing the newly produced lock assembly to return to the center of the floating process control boundary, thereby achieving closed-loop quality correction for each piece in the continuous production process.

[0080] In one embodiment of the present application, the step S5 uses response data to correct the dynamic response surface and the subsequent floating process control boundary, and the process of iterative updating the correlation between production rhythm and quality performance is as follows:

[0081] First, the weighted import and data set update operation for the latest production feedback data are performed. After the process parameter compensation instruction is executed and the new lock assembly is completed, the detection instrument immediately reads the final quality response data and pairs it with the corresponding production line running rhythm data at this moment. The processing unit imports this set of data stream containing the latest equipment state characteristics into the original time sequence data set in the storage module in real time. In order to accurately reflect the current wear state of the equipment and the influence of environmental thermal drift, the calculation algorithm introduces a time decay based weight distribution mechanism when merging data. This mechanism gives the newly generated latest data a very high calculation weight, usually set to a reference value of 1.0, while the historical data of several days or several weeks ago is gradually reduced in weight coefficient according to the time interval. Through this differentiated data processing strategy, the processing unit generates a weighted update data set that focuses on reflecting the recent state of the equipment. This data set gradually forgets the outdated equipment behavior patterns in statistical characteristics, and focuses on the current physical characteristics, thereby providing the most time-sensitive sample basis for the next mathematical modeling.

[0082] The computing module performs the re-fitting of the non-linear relationship and the model correction task using the constructed weighted update dataset. The module calls the weighted least squares algorithm to recalculate the polynomial coefficients in the dynamic response surface function, taking the weighted production intensity data as the independent variable and the quality performance data as the dependent variable. This iterative calculation process fine-tunes the geometric shape of the dynamic response surface, thereby more accurately approximating the real performance curve of the current device in actual operation. Next, the module performs second-order differentiation on the corrected dynamic response surface and recalculates the curvature values of each discrete point on the surface in combination with the first derivative. Through global scanning of the updated curvature distribution, the module can accurately capture the latest coordinate position where the curvature value presents a local maximum. This new coordinate point objectively reveals the specific critical point where the quality index changes abruptly with the production intensity under the current device capacity conditions, thereby completing the dynamic positioning of the inflection point coordinates and ensuring that the mathematical model can sensitively perceive small shifts in device performance.

[0083] In the final stage, the core control parameters are refreshed according to the repositioned inflection point coordinates and the closed-loop update is completed. The processing unit reads the projection value of the new inflection point coordinates on the production intensity axis and establishes it as the latest critical production rhythm threshold. Based on this updated threshold, the processing unit calls the boundary generation algorithm to reconstruct the next floating process control boundary and recalculates the upper and lower limit values of the safe pressure interval at different operating rhythms. These corrected boundary parameters are immediately packaged as an update instruction package and sent to the on-site process monitoring module through the industrial communication network. After receiving the instruction, the process monitoring module immediately performs parameter hot replacement and uses the new control boundary to overwrite the old monitoring standard. This operation ensures that the quality management logic in the next production decision is completely based on the latest device capacity and process correlation, achieving a fully automatic closed-loop iteration from data collection to model correction to standard update and eliminating the risk of misjudgment due to the lag of control standards behind device state changes.

[0084] The above describes one embodiment of the present application in detail, but the content described is only the preferred embodiment of the present application and cannot be considered as limiting the scope of the implementation of the present application. Any equivalent changes and improvements made in accordance with the scope of the present application should still be within the scope of the patent coverage of the present application.

Claims

1. A method for quality management in the production of locking assemblies for aluminum trim parts in passenger vehicles, characterized in that, Includes the following steps: S1: Collect locking force performance data of historical production batches of aluminum trim locking components for passenger vehicles and synchronous production line cycle data; map and bind the locking force performance data and cycle data on the time axis to generate a time-series dataset containing information related to quality status and production intensity. S2: Construct a dynamic response surface of the quality performance of locking components relative to production intensity using time-series datasets; identify inflection points where quality indicators on the dynamic response surface change abruptly, and determine the critical production rhythm threshold for maintaining quality stability under the current production organization method; S3: Generates a floating process control boundary that changes in real time with the production line's operating rhythm based on the critical production rhythm threshold; during production acceleration or deceleration, automatically shifts the upper and lower limits of the locking force warning range to construct a dynamic quality monitoring range that matches the current production load capacity. The process of generating a floating process control boundary that changes in real time with the production line's operating rhythm based on the critical production rhythm threshold is as follows: Collect the real-time production line operating cycle time and calculate the load difference ratio between it and the critical production rhythm threshold; substitute this load difference ratio into the preset boundary compensation algorithm to calculate the boundary shift correction value adapted to the current production load state. Retrieve the basic static control range data of the locking component; use the boundary translation correction value to perform a weighted offset calculation on the upper and lower limits of the basic static control range to obtain the instantaneous coordinate values ​​of the dynamic range; The calculated dynamic range real-time coordinate values ​​are continuously output to the process monitoring module in a time series; by continuously updating the real-time coordinate values, a floating process control boundary that is linked in real time with the production line's operating rhythm in the time domain is formed. The process of establishing the upper and lower limits of the automatic translation locking force early warning range to construct a dynamic quality monitoring range that matches the current production load capacity is as follows: Calculate the real-time differential value of the production line's operating cycle time to obtain the cycle time change rate; determine the current production acceleration or deceleration state based on the positive or negative sign of the cycle time change rate, and simultaneously quantify the specific intensity of the change. The quantified intensity of change is substituted into a preset rhythm-deviation compensation function model for calculation; the interval translation compensation value that maintains a linear mapping relationship with the current acceleration or deceleration trend is obtained analytically. The interval translation compensation value is superimposed and corrected to the current upper and lower limits of the floating process control boundary; the numerical interval after superposition calculation is locked as the final dynamic quality monitoring range adapted to the current production load capacity; S4: Compare the real-time detected locking force value of the locking component with the floating process control boundary; when the detected value approaches the warning range of the floating boundary, output the process parameter compensation command to the production equipment control unit, and offset the tendency of quality fluctuation caused by changes in production rhythm by fine-tuning the molding pressure. S5: Collect the quality response data of the locking component after process parameter compensation and send it back to the dynamic response surface model; use the response data to correct the dynamic response surface and subsequent floating process control boundaries, and complete the iterative update of the relationship between production rhythm and quality performance.

2. The method for quality management of the production of locking components for aluminum trim parts in passenger vehicles according to claim 1, characterized in that, In step S1, the process of mapping and binding the locking force performance data and the operating cycle data on the time axis to generate a time-series dataset containing information related to quality status and production intensity is as follows: The production line main controller and locking force detection unit read the production line cycle time data and locking force performance data respectively; the high-precision network timing protocol of industrial Ethernet is used to synchronize the clocks of the two independent data sources, and each piece of collected data is marked with a nanosecond-level timestamp. Using the timestamp sequence of production line cycle data as a reference coordinate, the locking force performance data is mapped to the corresponding cycle based on the proximity principle of timestamps; by comparing the consistency of timestamps, discrete locking force detection values ​​are logically associated and bound to the continuously changing production line cycle data point by point. The logically related cycle time values ​​and locking force performance data are merged and stored; a multidimensional structured data table with timestamp as the unique index key is constructed, and a time-series dataset containing the correspondence between production intensity input and quality status output at each moment is output.

3. The method for quality management of the production of locking components for aluminum trim parts in passenger vehicles according to claim 1, characterized in that, In step S2, the process of constructing a dynamic response surface of the locking component's quality performance relative to production intensity using a time-series dataset, identifying inflection points where quality indicators on the dynamic response surface suddenly change, and determining the critical production rhythm threshold for maintaining quality stability under the current production organization method is as follows: Production intensity data was extracted from the time series dataset as the independent variable and locking component quality performance data as the dependent variable. The least squares method was used to perform nonlinear fitting on the independent and dependent variables to establish a dynamic response surface mathematical model describing the continuous change of quality indicators with production intensity. The second derivative of the established dynamic response surface model with respect to the production intensity variable is obtained; the curvature values ​​of each discrete point on the surface are calculated, and a curvature distribution curve reflecting the sensitivity of the locking component's quality performance to changes in production intensity is generated. Retrieve the coordinates of the curvature distribution curve where the curvature value exhibits a local maximum; mark these coordinates as the inflection point where the quality index undergoes a sudden change, serving as the critical dividing point between a stable production state and a nonlinear quality fluctuation state; The identified inflection point coordinates are orthogonally projected onto the production intensity coordinate axis of the dynamic response surface; the specific production intensity value corresponding to the projection point is read and locked as the critical production rhythm threshold for maintaining quality stability under the current production organization mode.

4. The method for quality management of the production of locking components for aluminum trim parts in passenger vehicles according to claim 3, characterized in that, The process of obtaining the second derivative of the established dynamic response surface model with respect to the production intensity variable, calculating the curvature values ​​at each discrete point on the surface, and generating a curvature distribution curve reflecting the sensitivity of the locking component's quality performance to changes in production intensity is as follows: The numerical computation unit is invoked to parse the function polynomial expression of the dynamic response surface mathematical model; second-order differential operations are performed on the production intensity variable in the expression to derive the analytical expression of the second-order derivative function describing the gradient change rate of the quality index; Extract the production intensity coordinates of each discrete point on the dynamic response surface model and substitute them into the analytical expression of the second derivative function; By combining the first derivative values ​​of the corresponding points and the formula for calculating differential geometric curvature, the geometric curvature values ​​at each discrete point are calculated. An analytical coordinate system is established with production intensity as the horizontal axis and geometric curvature values ​​as the vertical axis. The geometric curvature values ​​are mapped to this coordinate system and a smooth interpolation fitting algorithm is used to connect the points to generate a curvature distribution curve that reflects the continuous change of sensitivity.

5. The method for quality management of the production of locking components for aluminum trim parts in passenger vehicles according to claim 1, characterized in that, In step S4, the process of outputting process parameter compensation instructions to the production equipment control unit and offsetting the tendency of quality fluctuations caused by changes in production rhythm by fine-tuning the molding pressure is as follows: Calculate the deviation of the real-time locking force value relative to the center line of the floating process control boundary; combine the deformation modulus of the locking component material to reverse-calculate the molding pressure compensation value required to correct the current deviation; The molding pressure compensation value is encoded into a parameter correction instruction specific to the production equipment control unit; this instruction is sent first to the underlying programmable logic controller responsible for executing the pressing action via a high-speed industrial bus network. The programmable logic controller dynamically adjusts the output power of the press-fitting actuator according to instructions; in subsequent press-fitting strokes, it directly outputs the changed forming pressure to physically correct the mass data deviation of the locking component.

6. The method for quality management of the production of locking components for aluminum trim parts in passenger vehicles according to claim 1, characterized in that, In step S5, the process of using response data to correct the dynamic response surface and subsequent floating process control boundaries to complete the iterative update of the correlation between production rhythm and quality performance is as follows: Import the quality response data of the locking components after process parameter compensation and the corresponding production line cycle time data into the original time-series dataset; assign the latest data in this set a higher calculation weight than the historical data to generate a weighted update dataset that focuses on reflecting the recent equipment status; We refit the nonlinear relationship between production intensity and quality performance using a weighted updated dataset to generate a corrected dynamic response surface; we recalculate the curvature extrema to locate the inflection point coordinates where quality indicators reflecting current equipment capabilities undergo abrupt changes. The critical production rhythm threshold is refreshed based on the new inflection point coordinates, and the subsequent floating process control boundary is reconstructed. The updated boundary parameters are then sent to the process monitoring module to cover the old standards in order to complete the iterative update of the relationship between production rhythm and quality performance.

Images