Module dynamic assembly and life cycle management system based on dependency injection strategy

Abstract

The application relates to the technical field of electric digital data processing, and discloses a module dynamic assembly and life cycle management system based on a dependency injection strategy, which comprises a data interaction module, a logic assembly unit and a dependency injection center, wherein the data interaction module acquires an interactive load data sequence of a running state of a controlled physical object and a logic trigger signal; the logic assembly unit determines a physical elastic characteristic factor according to the interactive load data sequence, injects a phase hedging item in a characteristic package weight switching process, and generates a dynamic weight adjustment instruction; the dependency injection center reconstructs a new and old characteristic package weight proportion according to the dynamic weight adjustment instruction, adopts state fingerprint inheritance logic after the physical elastic characteristic factor reaches a preset steady state judgment threshold, and destroys an old instance, the application solves the decoupling problem of a control logic and a physical deformation field through real-time coupling of a physical elastic characteristic factor and control weight, eliminates data breakpoints in a switching transient state, and maintains smooth system operation.

Description

Technical Field

[0001] This invention belongs to the field of electronic digital data processing technology, and in particular relates to a module dynamic assembly and lifecycle management system based on a dependency injection strategy. Background Technology

[0002] Current control kernels generally have pre-set static process procedure calculation logic to cope with continuous production of a single specification through preset parameter sets, ensuring steady-state system operation. With the increasing demand for flexible production, production lines need to perform rapid switching between materials of different specifications, which requires the control software to have the function of dynamically adjusting the control algorithm. Existing technology adopts a switching method based on material tracking position. When the system detects that the head of the rolled piece is close to the stand entrance, it triggers parameter refresh. However, there is a mismatch between the logic assembly cycle of the control software and the physical response time constant of the rolling mill pressing mechanism. During the replacement of control modules, the task scheduling mechanism of the real-time operating system generates instantaneous jitter of the calculation cycle. This uncertainty at the digital processing level causes the control command to have a phase lag at the biting moment, resulting in a timing disconnect between the control output and the physical action.

[0003] Hardware-based positioning triggering or preset parameter overlay cannot solve the dynamic mismatch of the system. Besides hardware improvements, the control algorithm lags behind in handling the difference between the digital logic assembly cycle and the response constant of the physical actuator. For example, Chinese invention patent CN110449469B discloses a vibration detection and control device and method for a corrugated roll mill. It uses an infrared rangefinder to monitor vibration displacement; when the displacement difference exceeds a preset threshold, an alarm is triggered and a damping signal is generated. This scheme is a passive response mode, with control logic activation lagging behind physical vibration. Software module loading is limited by the operating system's task scheduling mechanism. Under extremely high-speed engagement or multi-specification switching conditions, the non-deterministic delay in digital processing leads to control command interruptions. The switching process generates phase shifts, causing a disconnect between instruction generation and physical action timing. The solution lacks the ability to inherit the environmental characteristics of materials from the previous specification. The instantaneous control step during module replacement induces elastic oscillations in the rack, affecting the continuity of flexible production and product accuracy. To address these challenges, the industry has attempted to mitigate the switching impact by adding data buffers or introducing smoothing filtering algorithms. However, analysis shows that these methods achieve numerical stability by sacrificing the dynamic response performance of the control system, failing to solve the synchronization problem between digital logic assembly and physical displacement cycles at the physical perception level. Increasing hardware computing power cannot eliminate the nondeterministic delays during software module loading, making it difficult for the system to maintain high-precision real-time response under extreme switching conditions.

[0004] Therefore, how to construct a dynamic module assembly mechanism with physical state awareness to generate deterministic and smooth migration of control logic weights during material displacement is the technical problem to be solved by this invention. Summary of the Invention

[0005] This invention provides a module dynamic assembly and lifecycle management system based on a dependency injection strategy, including a data interaction module, a logic assembly unit, and a dependency injection center:

[0006] The data interaction module is used to acquire the interactive payload data sequence and logical trigger position signal that characterize the operating state of the controlled physical object;

[0007] The logic assembly unit is used to determine the physical elastic characteristic factor based on the interactive load data sequence, and to inject a phase offset term during the weight switching process of the process control feature package according to the logic trigger position signal, so as to generate a dynamic weight adjustment instruction.

[0008] The dependency injection center, connected to the logic assembly unit, is used to reorganize the weight ratio of the old and new process control feature packages according to the dynamic weight adjustment instruction. After the weight ratio reorganization is completed and the physical elasticity feature factor reaches the preset steady-state judgment threshold, the life cycle instance of the old process control feature package is destroyed. The phase offset term is used to adjust the injected weight in advance to offset the response lag of the actuator at the moment of intervention when the controlled physical object undergoes physical elastic deformation. Before destroying the old process control feature package, the dependency injection center extracts the control residual amount of the current active feature package at the end of the operation and maps it to the initial value register of the target module.

[0009] Furthermore, the operation of the logic assembly unit to inject the phase offset term includes: determining the physical elastic displacement of the controlled physical object at the moment of intervention based on the absolute value of the load change rate in the interactive load data sequence and the preset structural elastic coefficient, and converting the physical elastic displacement into a phase offset term; the logic assembly unit superimposes the phase offset term according to the basic weight ratio determined by the logic trigger position signal to construct a dynamic weight adjustment instruction.

[0010] Before destroying the old process control feature package, the dependency injection center intercepts the output residual and integral cumulative value of the old process control feature package in the current sampling period, and uses the output residual and integral cumulative value as the initial feature vector to inject into the input stack of the target module.

[0011] The logic assembly unit is equipped with a dynamic weight adjustment operator, and the adjustment amount output follows the following logic: Where U is the output control quantity of the dynamic weight adjustment command. The basic weight distribution is determined based on the logical trigger position signal. The current logical coordinate point is represented by the logical trigger position signal. The logical output value of the new process control feature package. This is the logical output value of the old process control feature package. This is a phase offset term generated based on physical elastic displacement.

[0012] Preferably, the logic assembly unit also performs active excitation detection: during the intervention gap of the controlled physical object, the real-time total physical delay of the control loop is intercepted by the feedback time stamp, and the trigger threshold of the logic trigger position signal is corrected based on the real-time total physical delay, so as to realize the phase advance compensation of the dynamic weight adjustment command relative to the instant of physical intervention.

[0013] Preferably, the system also includes a shadow preloading layer, which is used to perform the preloading action of the process control feature package before the phase offset term is injected: when the logic trigger position signal reaches the preset time window start point, the shadow preloading layer retrieves the binary image of the target module from the strategy storage area and completes the dependency resolution of the target module, so that the target module is in a ready suspended state.

[0014] Preferably, the system further includes a deterministic kernel locking module for locking the operation clock cycle during weight ratio reorganization; during phase offset injection, the deterministic kernel locking module disables background tasks other than high-priority interrupts, limiting the weight switching action to a time deterministic window of 10μs to 50μs.

[0015] Preferably, when the dependency injection center destroys the old process control feature package, it cleans up the memory address space occupied by the old process control feature package and releases the hardware channel occupancy rights associated with the old process control feature package.

[0016] Preferably, the logic assembly unit determines the intervention state of the controlled physical object by calculating the time partial derivative of the load change rate in the interactive load data sequence; when the absolute value of the time partial derivative drops to within the preset zero-point threshold range, the logic assembly unit determines that the physical elastic characteristic factor has reached the preset steady-state determination threshold and outputs a phase offset term cancellation signal.

[0017] Preferably, the data interaction module also acquires physical fatigue characteristic data of the controlled physical object; the dependency injection center corrects the gain coefficient of the phase offset term based on the physical fatigue characteristic data to compensate for the control phase deviation caused by the drift of physical environment attributes.

[0018] Compared to existing technologies, the module dynamic assembly and lifecycle management system based on the dependency injection strategy of this invention has the following advantages:

[0019] 1. In the dynamic assembly and lifecycle management of modules, the physical coordinates of the rolled piece relative to the mill stand are determined in real time through the material tracking module. And the logic assembly unit according to the physical coordinates By triggering atomic pointer jumps in the process control feature package in shadow memory, a deterministic mapping mechanism is constructed that directly drives digital logic assembly by physical space displacement. This decouples the originally static and resident control algorithm into discrete feature packages that are injected on demand and destroyed at regular intervals. By using physical displacement as the time axis of software logic assembly, the life cycle of control logic and the physical displacement cycle of materials are synchronized. Compared with the traditional full code refresh or passive parameter switching method, this invention eliminates the logic response lag of the control program when switching between multiple specifications continuously, and ensures the absolute determinism of the output command of the digital control system at the moment of material bite.

[0020] 2. By using the dynamic weight adjustment operator configured in the logic assembly unit, the nonlinear drift of the control weight is executed within the process procedure switching interval, so that the output command U of the physical stand follows the transition logic jointly decided by the old and new feature packages. This nonlinear transition mode based on the bite depth of the rolled piece transforms the discrete software module assembly process into a continuous energy evolution process that matches the physical deformation process. When the mechanical impact generated by the high deformation resistance material entering the stand overlaps with the switching point of the control logic, the control step caused by the model switching is eliminated from the command generation level by dynamically adjusting the contribution ratio of the old and new feature packages, avoiding mechanical oscillations caused by logical mutations and maintaining the dynamic stability of the control system during extreme working condition transitions.

[0021] 3. By utilizing the dependency injection center to execute the state fingerprint inheritance mechanism before the module is destroyed, the control residual quantity of the current active state feature package at the end of the operation is extracted and mapped to the initial value register of the target module. This breaks the technical limitation of independent and stateless modules in the conventional dependency injection mode. Based on the new injection module's ability to perceive the environmental characteristics of the previous life cycle, in continuous production scenarios, by converting the physical deviations caused by rack thermal fluctuations or mechanical fatigue into the initial compensation variables of digital logic, seamless continuation of cross-life cycle logic in environmental attributes is achieved, eliminating data breakpoints caused by instantaneous module hot switching, and ensuring the system's adaptive compensation accuracy for process environment drift. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating the interaction flow of system operation data and the dynamic assembly logic of modules in this invention.

[0023] Figure 2 This is a diagram showing the overall system topology and radial connection relationships of the functional units of the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0025] It should be noted that all directional and positional terms used in this invention, such as: up, down, left, right, front, back, vertical, horizontal, inner, outer, top, bottom, transverse, longitudinal, center, etc., are only used to explain the relative positional relationship and connection between components in a specific state (as shown in the accompanying drawings). They are only for the convenience of describing this invention and do not require that this invention be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention. In addition, the descriptions of "first," "second," etc., in this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated.

[0026] In the description of this invention, unless otherwise explicitly specified and limited, the terms installation, connection, and linking should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections; they can refer to direct connections or indirect connections through an intermediate medium; they can refer to the internal connection of two components. For those skilled in the art, the specific meaning of the above terms in this invention can be understood according to the specific circumstances.

[0027] In the description of this specification, references to the terms "an embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example, and the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0028] This invention provides a module dynamic assembly and lifecycle management system based on a dependency injection strategy. It mainly includes a data interaction module, a logic assembly unit, and a dependency injection center. The data interaction module and the logic assembly unit communicate via a data bus to capture the kinematic characteristics and logical trigger signals of the controlled physical object in real time. The logic assembly unit performs dynamic calculations and phase compensation of control weights based on the captured data and sends the generated dynamic weight adjustment instructions to the dependency injection center. The dependency injection center is responsible for loading process control feature packages, weight reconstruction, state inheritance, and cleaning up and destroying old instances. The system realizes control logic by constructing a dynamic assembly mechanism that is coupled in real time with the physical deformation field at the digital processing level. Migration and deterministic response within the material switching window; controlled physical objects generate load fluctuations during operation. To enable the digital system to accurately perceive changes in physical boundaries, the data interaction module adopts a high-frequency sampling method, acquiring interactive load data sequences representing the operating state of the controlled physical object through a real-time data bus, while simultaneously monitoring the logical trigger position signal representing the position of the rolled piece head. The interactive load data sequence includes real-time pressure values ​​and velocity vectors with a sampling frequency of not less than 1kHz. The data interaction module converts the collected raw data into feature vectors that conform to a preset data structure, that is, defines a structured data packet containing load amplitude, rate of change, and time stamp information, and pushes it to the input buffer of the logic assembly unit in real time.

[0029] To achieve adaptive hedging of physical inertia at control logic switching points, the system executes an active excitation detection procedure to determine the real-time total physical delay of the control loop. During the material bite gap, the logic assembly unit sends a detection pulse to the actuator of the controlled physical object through the data interaction module and intercepts the feedback time stamp at the actuator. The system records the time difference from the moment the detection pulse is emitted to the moment the feedback signal is transmitted as the real-time total physical delay. The logic assembly unit is based on the real-time total physical delay. The advance compensation value of the logic trigger position signal is calculated by combining the current running speed v of the controlled physical object. The advance compensation value satisfies the following relationship: ,in, This is the advance compensation value, in units of v represents the running speed, in m / s; The total physical latency is measured in milliseconds (ms), and the logic assembly unit compensates for this latency with lead time. The position trigger threshold is dynamically adjusted to counteract the phase shift caused by the coupling between information transmission resistance and mechanical inertia.

[0030] Due to a magnitude mismatch between the logic assembly cycle of the control software and the physical response time constant of the actuator, the logic assembly unit executes a weight adjustment procedure with physical sensing characteristics. Based on the absolute value of the load change rate in the interactive load data sequence and a preset structural elasticity coefficient, the logic assembly unit determines the physical elastic displacement of the controlled physical object at the instant of intervention, and uses this displacement to calculate the phase offset term. The logic assembly unit generates the output control quantity U by calling a built-in dynamic weight adjustment operator. The operational logic of the dynamic weight adjustment operator follows the following relationship: Where U is the output control quantity of the dynamic weight adjustment command; The basic weight distribution is determined based on the logical trigger position signal; The current logical coordinate point is represented by the logical trigger position signal; The logical output value of the new process control feature package; This is the logical output value of the old process control feature package; The phase offset term is generated based on the physical elastic displacement, and the phase offset term is injected during the weight switching process. The system performs instantaneous adjustments upon the mechanical impact generated by the workpiece biting in, to counteract the execution lag caused by the elastic deformation of the stand arch; the logic assembly unit determines the basic weight distribution. When the function follows an S-shaped logistic regression curve, the expression is: α is the control switching slope factor, which limits the degree of weight migration; β is the displacement of the switching center point calibrated by the material tracking system. Logical coordinate points acquired by the displacement sensor; phase offset term The sensitivity coefficient λ is determined as follows: A pressure step test is performed under no-load conditions. The elongation data sequence of the bearing during the process of increasing the rolling force from zero to 20% of the rated load is recorded. The least squares method is used to perform linear fitting on the load change sequence and the displacement feedback sequence to determine the inherent longitudinal stiffness coefficient K. The reciprocal of K is then converted into a dimensionless weighted correction operator to reduce micron-level physical deformation. Convert to a phase offset gain value between 0 and 1.

[0031] In multi-specification continuous production processes, the loading process of software modules can lead to competition for computing resources and task scheduling jitter. To ensure a constant computing cycle for the real-time kernel, this application employs a shadow preloading method, including a shadow preloading layer, which is triggered when a logical position signal is received. When the preset time window begins, the shadow preloading layer retrieves the binary image of the target module from the policy storage area and completes dependency resolution and environment initialization in the inactive area of ​​memory, putting the target module into a ready-suspended state. Simultaneously, the system also includes a deterministic kernel locking module, used to lock the computation clock cycle during weight ratio reorganization. During phase offset injection, the deterministic kernel locking module blocks background tasks except for high-priority interrupts, limiting weight switching actions to a deterministic time window of 10μs to 50μs, achieving zero-impact operation at microsecond precision. During weight ratio reorganization, the deterministic kernel locking module blocks background tasks except for hard real-time interrupts through the operating system kernel scheduling interface, limiting the task locking period to a fixed window of 10μs to 50μs determined by the processor clock frequency and task stack depth. When the controlled physical object reaches steady state, the logic assembly unit intercepts the original sample of the interactive load data sequence during the no-load operation phase, calculates the standard deviation σ of the load change rate within a 500ms time window, and the steady-state determination threshold. The absolute value of the time partial derivative of the load change rate is three times the standard deviation σ, excluding high-frequency electromagnetic interference signals; the absolute value of the time partial derivative of the load change rate remains below the steady-state determination threshold for 20ms. At that time, the destruction of old instances and the cleanup of memory address space are triggered.

[0032] Conventional dependency injection lacks state continuity, leading to control step jumps during module replacement. To address this, the dependency injection center executes a state fingerprint inheritance procedure before destroying the old process control feature package. The center intercepts the output residual and integral cumulative value of the old process control feature package in the current sampling period and maps them as environmental feature vectors to the initial value register of the target module. The newly injected module acquires the environmental features of the previous lifecycle, compensating for physical deviations caused by rack thermal fluctuations or mechanical fatigue at the logic level. Once the physical elasticity characteristic factor reaches a preset steady-state threshold (i.e., the absolute value of the time partial derivative of the load change rate in the interactive load data sequence drops to a preset zero-point threshold), the dependency injection center executes an atomic pointer jump to destroy the old instance, clearing its occupied memory address space and releasing associated hardware channel occupancy rights, achieving a smooth, non-linear transition of control logic from the previous process procedure to the current process procedure. The dependency injection center executes state fingerprint inheritance to extract the integral cumulative value of the old process control feature package in the current sampling period. Mapped to the target module's initial value register; excluding output step caused by differences in control algorithm structure, the initial integral compensation amount of the new instance is calculated by dividing the old instance's integral control quantity by the ratio of the new and old feature packet proportional gains through energy equivalent conversion logic. At the instant of the atomization pointer jump, the initial output value of the newly injected module and the residual control value at the end point of the old instance's execution maintain continuity with a deviation within 5%; the environmental feature vector contains a physical deformation trend field, which is calculated by the dynamic jump displacement. The slope of change is determined over 10 consecutive sampling periods, and the new instance loading instant provides compensation for the predicted physical background.

[0033] Example 1: The rate of change of rolling force fed back by the pressure sensor during the biting operation of the metal workpiece received on the rolling mill stand. A load surge occurs within a timescale of 5ms to 10ms. Constrained by physical inertia, the rolling mill stand exhibits elastic longitudinal displacement, with a mechanical response time constant less than the task scheduling cycle of the electronic digital data processing system. The data interaction module acquires the interactive load data sequence at a sampling frequency of 1kHz, and the logic assembly unit determines the load based on the rolling force change rate. The dynamic runout displacement, as a physical elastic characteristic factor, is determined by combining the inherent longitudinal stiffness coefficient K of the rolling mill stand. When the logic triggers the position signal When the preset time window begins, the shadow preloading layer retrieves the binary image of the target module from the policy store and completes environment initialization in the inactive area of ​​memory. The logic assembly unit then dynamically adjusts its displacement. Calculate the phase offset term The logic assembly unit calls the dynamic weight adjustment operator to generate the output control quantity U, which follows the mathematical relationship as follows: Where U is the output control quantity of the dynamic weight adjustment command; The basic weight distribution is determined based on the logical trigger position signal; The current logical coordinate point is represented by the logical trigger position signal; The logical output value of the new process control feature package; This is the logical output value of the old process control feature package; Based on dynamic jump displacement The generated phase offset term is an injection weight of the new process control feature package that is added instantaneously when the system undergoes deformation at the physical level, in order to offset the execution lag caused by mechanical deformation.

[0034] The dependency injection center reconstructs the weight ratio of the old and new process control feature packages based on the output control quantity U. After the physical elasticity feature factor reaches the preset steady-state judgment threshold, it intercepts the output residual and integral cumulative value of the old process control feature package in the current sampling period, maps the output residual and integral cumulative value as environmental feature vectors to the initial value register of the new process control feature package, executes the state fingerprint inheritance logic, and executes atomic pointer jump to destroy the life cycle instance of the old process control feature package and release the associated hardware channel occupancy rights. The system achieves alignment between the logic switching action and the physical instantaneous movement of the rolling mill stand at the time scale level by converting physical deformation features into dynamic correction variables of injection weights.

[0035] Example 2: In the experiment used to verify the adaptability of the coupling between industrial control logic and physical entity state, the thickness control scenario of a metal rolling production line was selected as the background. Targeting the pressure signal transmission resistance and physical execution inertia generated under ultra-high speed biting conditions, the test platform was set up in an electrical digital data processing environment that includes a processing unit with a computing frequency of no less than 2.4 GHz and a distributed clock system with sub-microsecond precision. The data interaction module collected the original load signal at a sampling frequency of 1 kHz and superimposed Gaussian white noise with a signal-to-noise ratio of 20 dB into the signal source. The sampling frequency was set based on the analysis of the upper limit of the load fluctuation frequency of the controlled physical object. Under the premise of ensuring that the acquisition accuracy is no less than 0.05%, the system data processing load was kept within the preset processing unit performance boundary.

[0036] After the test is started, the raw interactive load data sequence containing noise is recorded, and the rate of change of rolling force of the controlled physical object at the instant of workpiece bite is measured. The value is 1528.6 kN / s, and the logic assembly unit is based on the rolling force variation rate. The phase offset term is generated by combining the inherent longitudinal stiffness coefficient K. The value was 0.0834. During the feature package weight switching process, the peak value of the control output residual of the experimental group was 4.18%, while the peak value of the residual of the control group with the phase offset term removed and the standard weight distribution executed was 13.25% under the same working conditions. The experimental data showed that the sample group of the present invention achieved a 68.45% suppression effect on the phase deviation caused by mechanical deformation. Further boundary condition verification was performed. When the rolling force change rate was adjusted to 1757.9 kN / s, which is 15% higher than the upper limit of the preset protection range, the phase offset term was affected because the physical entity entered the elastic deformation limit zone. The correction efficiency reached saturation, and the residual suppression rate dropped to 51.12%, confirming that the parameter range defined by the present invention is the working range for achieving control accuracy optimization. The experimental results confirm that the physical elastic characteristic factor and the dynamic weight adjustment operator have a synergistic effect. By converting the physical displacement into a digital correction variable, the response deviation between high-frequency real-time calculation and large inertia physical actuator is solved, and the control logic is migrated under environmental disturbances.

[0037] Example 3: This example combines Figures 1 to 2 This section describes a module dynamic assembly and lifecycle management system based on dependency injection strategy, such as... Figure 1 As shown, the controlled physical object is in operation and transmits its state data to the data interaction module. The data interaction module contains functional units for acquiring interactive load data sequences, acquiring logic trigger signals, and monitoring the physical object's operating state. The data interaction module outputs the interactive load data sequences and logic trigger signals to the logic assembly unit. The logic assembly unit performs operations such as determining physical elasticity characteristic factors, injecting phase offset terms, and generating dynamic weight adjustment instructions during feature package weight switching. The logic assembly unit outputs the dynamic weight adjustment instructions to the dependency injection center. The dependency injection center performs operations such as reconstructing the weight ratio of the old and new feature packages, executing state fingerprint inheritance logic, and destroying the old process control feature package instance after satisfying the steady-state judgment threshold.

[0038] like Figure 2 As shown, the module dynamic assembly and lifecycle management system based on the dependency injection strategy serves as the central node, and is radially connected to the data interaction module, the logic assembly unit, the dependency injection center, and the physical support environment. The data interaction module is connected to the interaction payload data sequence and the logic trigger position signal, respectively. The logic assembly unit is connected to the physical elasticity feature factor, the phase offset term, and the dynamic weight adjustment instruction, respectively. The dependency injection center is connected to the weight ratio reorganization, the state fingerprint inheritance logic, and the old instance destruction, respectively. The physical support environment is connected to the controlled physical object, the initial value register, and the physical sensor group, respectively.

[0039] Example 4: When switching between different specifications of sheet metal, considering the nonlinear behavior of the frame stiffness under different reduction rates, the logic assembly unit executes a parameter calibration method to determine the phase offset term. To determine the corrected strength, the system performs an unloaded rolling test under no-load conditions. The data interaction module records the elongation of the roll cage as the rolling force increases from zero to 20% of the rated value. The least squares method is used to perform linear fitting between the rolling force change sequence and the displacement feedback sequence to calibrate the inherent longitudinal stiffness coefficient K. Based on this, the logic assembly unit determines the phase offset term. With dynamic jump displacement The sensitivity coefficient λ, with a value range of 0.015 to 0.025, is used to map micrometer-level physical deformations to a dimensionless weighted correction operator between 0 and 1. The phase offset term satisfies the following relation. ,in, The phase offset term is represented by λ, which is the sensitivity coefficient. The dynamic jump displacement is expressed in μm; through this quantitative mapping path, the system realizes parameter transmission from the physical deformation field to the digital control domain.

[0040] To address the issue of steady-state determination during instance destruction, the system employs an automatic threshold calibration method based on signal statistical characteristics. The logic assembly unit captures raw samples of the interactive load data sequence within the no-load operation cycle, calculates the standard deviation σ of the rolling force change rate within a 500ms time window, and sets the steady-state determination threshold accordingly. Set to 3 times the standard deviation σ, and following the 3σ principle to eliminate interference from high-frequency electromagnetic noise on steady-state determination, the threshold is set when the absolute value of the time partial derivative of the rolling force change rate remains below the dynamically calibrated steady-state determination threshold for 20 ms. When the system determines that the stimulated oscillation caused by the physical bite has disappeared, it triggers the dependency injection center to execute an atomic pointer jump. When the dependency injection center inherits the execution state fingerprint, it constructs an environmental feature vector containing an output residual field, an integral cumulative term field, and a physical deformation trend field. This vector is written into the initial value register of the target module in the form of a structured data packet through memory mapping technology. The output residual field corresponds to the PID control deviation value of the previous sampling period, the integral cumulative term field corresponds to the accumulation of controlled variables under historical operating conditions, and the physical deformation trend field corresponds to the slope of the dynamic jump displacement. By injecting the above fingerprint information containing the physical background into the new instance, the system compensates for the initial position deviation caused by the rack thermal drift at the moment of logic switching, thereby improving the output accuracy of the control command to the sub-millimeter level and maintaining stable operation.

[0041] Example 5: In process adaptation scenarios for different board thicknesses, the logic assembly unit determines the weight distribution function by executing a calibration procedure. The system, under a controlled testing environment, inputs a logical coordinate sequence containing 512 steps and simultaneously collects position feedback data at the instant of workpiece bite-in. A nonlinear fitting algorithm is used to process the collected data point sequence to construct a transition model that conforms to the physical response characteristics. The basic weight distribution is determined based on the logical trigger position signal. The system determines the position offset β and slope factor α of the model in the 100mm to 300mm bite range through calibration experiments. β is the position offset and α is the slope factor. The system aligns the evolution rate of the injected weights with the establishment speed of the frame arch deformation field in the temporal dimension, thus solving the logical deviation of weight allocation caused by the change in rolling speed.

[0042] When the system faces the replacement of physical sensor groups or the drift of the actuator environmental reference, it relies on the calibration procedure of the environmental fingerprint reference executed by the injection center. When the controlled physical object is in an unloaded state, the system obtains rolling force data within 100 sampling cycles through the data interaction module, calculates the average static pressure deviation, and uses this value as a zero offset term to compensate the output residual field of the environmental feature vector. The system determines the mechanical backlash of the transmission system by driving the roll to perform one reciprocating motion, and converts the displacement loss data into a displacement compensation variable and maps it to the initial value register of the target module. This pre-calibration process enables the process control feature package lifecycle instance to obtain the physical boundary of the current hardware state at the moment of loading, eliminates cross-lifecycle data breakpoints caused by equipment fatigue, and maintains stable system operation.

[0043] Example 6: In a scenario involving the deployment of a multi-stand hot strip mill system, the logic assembly unit executes a stiffness mapping program to determine the inherent longitudinal stiffness coefficient K. The data interaction module records the vertical displacement data of the mill stand under a controlled preload ranging from 0 kN to 5000 kN in 500 kN increments. The system uses the least squares method to perform linear fitting on the corresponding sequence of load and displacement to determine the inherent longitudinal stiffness coefficient K. The logic assembly unit injects a test phase offset term under an unloaded simulation environment. The closed-loop delay from the generation of control commands to the response of the actuator is measured using a distributed clock system. The corresponding value that makes the control output residual less than 5% of the preset boundary is selected as the operating point of the sensitivity coefficient λ, where K is the inherent longitudinal stiffness coefficient, with the unit being kN / μm. λ is the phase offset term; λ is the sensitivity coefficient.

[0044] When the system is deployed on a hardware platform with high concurrency processing capabilities and faces real-time kernel scheduling jitter, the shadow preloading layer executes a resource scheduling calibration method to determine the safe time window for module loading. The system performs the target module loading action 1000 times in a test environment where the processor is at 80% of its rated load, and measures the binary image loading time. Environment initialization time And determine the safety margin based on the real-time kernel scheduling jitter variance. The logic assembly unit is based on the rolling speed. Calculate the advance displacement of the logic trigger position signal The advance displacement satisfies the relationship. ,in, The advance displacement of the logical trigger position signal, in units of v represents the rolling speed, measured in m / s. Loading time, in milliseconds; Initialization time, in milliseconds; As a safety margin, the unit is ms. This process determines the task lock width during the weight ratio reorganization period, which is used to shield non-real-time calculation task interference and maintain a constant calculation cycle during material switching.

[0045] The embodiments of this application have been described above with reference to the accompanying drawings. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit of this application and the scope of protection of this invention, and all of these forms are within the protection scope of this application.

Claims

1. A module dynamic assembly and lifecycle management system based on dependency injection strategy, characterized in that, This includes a data interaction module, a logic assembly unit, and a dependency injection center. The data interaction module is used to acquire the interactive payload data sequence and logical trigger position signal that characterize the operating state of the controlled physical object; The logic assembly unit is used to determine the physical elastic characteristic factor based on the interactive load data sequence, and to inject a phase offset term during the weight switching process of the process control feature package according to the logic trigger position signal, so as to generate a dynamic weight adjustment instruction. The dependency injection center, connected to the logic assembly unit, is used to reorganize the weight ratio of the old and new process control feature packages according to the dynamic weight adjustment instruction. After the weight ratio reorganization is completed and the physical elasticity feature factor reaches the preset steady-state judgment threshold, the life cycle instance of the old process control feature package is destroyed. The phase offset term is used to adjust the injected weight in advance to offset the response lag of the actuator at the moment of intervention when the controlled physical object undergoes physical elastic deformation. Before destroying the old process control feature package, the dependency injection center extracts the control residual amount of the current active feature package at the end of the operation and maps it to the initial value register of the target module. Furthermore, the operation of the logic assembly unit to inject the phase offset term includes: determining the physical elastic displacement of the controlled physical object at the moment of intervention based on the absolute value of the load change rate in the interactive load data sequence and the preset structural elastic coefficient, and converting the physical elastic displacement into a phase offset term; the logic assembly unit superimposes the phase offset term according to the basic weight ratio determined by the logic trigger position signal to construct a dynamic weight adjustment instruction. Before destroying the old process control feature package, the dependency injection center intercepts the output residual and integral cumulative value of the old process control feature package in the current sampling period, and uses the output residual and integral cumulative value as the initial feature vector to inject into the input stack of the target module. The logic assembly unit is equipped with a dynamic weight adjustment operator, and the adjustment amount output follows the following logic: Where U is the output control quantity of the dynamic weight adjustment command. The basic weight distribution is determined based on the logical trigger position signal. The current logical coordinate point is represented by the logical trigger position signal. The logical output value of the new process control feature package. This is the logical output value of the old process control feature package. This is a phase offset term generated based on physical elastic displacement.

2. The module dynamic assembly and lifecycle management system based on dependency injection strategy according to claim 1, characterized in that, The logic assembly unit also performs active excitation detection: during the intermission interval of the controlled physical object, it intercepts the real-time total physical delay of the control loop through the feedback time stamp, and corrects the trigger threshold of the logic trigger position signal based on the real-time total physical delay, so as to realize the phase advance compensation of the dynamic weight adjustment command relative to the instant of physical intervention.

3. A module dynamic assembly and lifecycle management system based on dependency injection strategy according to claim 2, characterized in that, The system also includes a shadow preloading layer, which is used to perform the preloading action of the process control feature package before the phase offset term is injected: when the logic trigger position signal reaches the preset time window start point, the shadow preloading layer retrieves the binary image of the target module from the strategy storage area and completes the dependency resolution of the target module, so that the target module is in a ready suspended state.

4. A module dynamic assembly and lifecycle management system based on dependency injection strategy according to claim 3, characterized in that, The system also includes a deterministic kernel locking module, which locks the operation clock cycle during weight ratio reorganization; during phase offset injection, the deterministic kernel locking module masks background tasks except for high-priority interrupts, limiting the weight switching action to a time deterministic window of 10μs to 50μs.

5. A module dynamic assembly and lifecycle management system based on dependency injection strategy according to claim 1, characterized in that, When the dependency injection center destroys the old process control feature package, it cleans up the memory address space occupied by the old process control feature package and releases the hardware channel occupancy rights associated with the old process control feature package.

6. A module dynamic assembly and lifecycle management system based on dependency injection strategy according to claim 1, characterized in that, The logic assembly unit determines the intervention status of the controlled physical object by calculating the time partial derivative of the load change rate in the interactive load data sequence.

Images