Office furniture manufacturing data sensing and executing system based on Internet of Things

Abstract

The invention discloses an office furniture manufacturing data perception and execution system based on the Internet of Things, and relates to the technical field of furniture manufacturing management, and the system comprises an index generation module which is used for collecting control instruction parameters, state perception data and historical execution data, constructing a residual error factor group comprising a control residual error, a time synchronization deviation and an abnormal recording frequency, and generating a control instruction parameter; a control credible index is generated through a mapping rule; and the mirror image construction module is used for constructing a state mirror image template according to the process path parameters, the target size parameters, the target pose parameters and the process step numbers, and the state mirror image template represents the target state in the form of a structured parameter set. According to the invention, information fusion and control logic closed loop of multiple key dimensions in the manufacturing execution process are realized; particularly, the index generation module is combined with multi-source information such as a control residual error, a time synchronization deviation and an abnormal record frequency to form a residual error factor group and generate a control credible index.

Description

Technical Field

[0001] This invention relates to the field of furniture manufacturing management technology, specifically to an Internet of Things-based data sensing and execution system for office furniture manufacturing. Background Technology

[0002] With the widespread application of IoT technology in the industrial manufacturing sector, data-driven manufacturing process perception and decision control systems have become an important component of intelligent manufacturing. In furniture manufacturing, especially in the processing of office furniture, there are numerous scenarios involving multi-process collaboration, complex path execution, and high-precision status monitoring, which place higher demands on the real-time data acquisition and intelligent response capabilities of the manufacturing process. By building an interconnection mechanism between manufacturing equipment, sensing terminals, and control units, closed-loop data control can be achieved throughout the entire process from raw material processing to product assembly, thereby improving product quality and process stability. In the existing technology, there are several partially modular design schemes for state perception, process control and command execution in industrial processes, such as sensor-based data acquisition systems, process path-driven control command issuance platforms and data comparison-based anomaly detection mechanisms. However, these systems are mostly single-function oriented. Especially in scenarios involving multi-device linkage and complex process paths, issues such as data redundancy, timing errors and feedback delays between systems can easily lead to control decision deviations, reducing the stability of process execution and the accuracy of response. Therefore, there is a need for an intelligent manufacturing system that integrates data perception, process modeling, state comparison, execution verification, and instruction control to achieve dynamic perception, abnormal response, and control action freezing mechanism for each link in the complex office furniture manufacturing process, thereby enhancing the application value and industrial adaptability of intelligent manufacturing in the office furniture processing field. Therefore, this invention proposes an Internet of Things-based data perception and execution system for office furniture manufacturing. Summary of the Invention

[0003] The purpose of this invention is to provide an Internet of Things-based data sensing and execution system for office furniture manufacturing to solve the problems mentioned in the background art.

[0004] This invention can be achieved through the following technical solution: an Internet of Things-based data sensing and execution system for office furniture manufacturing, comprising: The index generation module is used to collect control command parameters, state awareness data and historical execution data, construct a residual factor set including control residuals, time synchronization deviations and abnormal record frequencies, and generate a control reliability index through mapping rules. The mirror construction module is used to construct a state mirror template based on process path parameters, target size parameters, target pose parameters and process step numbers. The state mirror template represents the target state in the form of a structured parameter set, including size parameters, pose parameters and process attribute parameters. The comparison module is used to compare the state-aware data with the state image template by corresponding parameter fields, calculate the field difference based on the preset tolerance parameters and field weight coefficients, and generate the state comparison result. The verification module is used to generate a reverse state path by calling the preset process reverse path generation rules after the control command is executed, based on the control command parameters, post-execution state perception data and process step number, and calculate the difference between the reverse state path and the state perception data before the control command is executed to generate a reversibility determination result. The control module receives the control reliability index, state comparison results, and reversibility determination results, and makes a joint judgment according to the control threshold rules. When any indicator fails to meet the control requirements, it prevents the issuance of control commands and sets the current process state to a frozen state.

[0005] A further technical improvement of the present invention is that the step of the index generation module generating a control credibility index includes: Collect control command parameters, status awareness data, and historical execution data for the current process, and construct control residuals based on the difference between control command parameters and status awareness data; Obtain the timestamp associated with the state-aware data and calculate the difference with the system's unified time base to construct the time synchronization deviation; Extract control anomaly marker records from historical execution data and count the frequency of anomaly records per unit time. The control residual, time synchronization deviation and anomaly recording frequency are combined with multiple dimensions of data, including at least the historical drift of the sensor, the real-time load rate of the equipment and the environmental temperature and humidity parameters, to form a residual factor group. The residual factor group is input into the preset mapping rule, which is constructed in the form of linear weighted summation. By setting corresponding weight coefficients for each residual factor, the weighting process is performed, and the final output is the control confidence index.

[0006] A further technical improvement of the present invention is that the method for constructing a state image template by the image construction module includes: The process field template is retrieved based on the current process step number. The process field template predefines the logical order and naming rules of the target size parameter field, target pose parameter field, and process attribute parameter field corresponding to each process step. Read the target size parameters and target pose parameters corresponding to the current process path parameters, and collect the material number, tool type and equipment number of the current processing object to form an initial parameter set; The initial parameter set is input into the process rule library, which includes a material-process parameter mapping table, a set of equipment constraints and tool adaptation rules. Based on this rule library, process attribute parameters that conform to the current processing conditions are generated. Based on the field order and field name defined in the process field template, the target size parameters, target pose parameters and process attribute parameters are merged and organized into a structured parameter set. The structured parameter set is constructed in the form of a table, where the fields are represented in the form of key-value pairs and can be indexed by a unique process step number. The structured parameter set is stored as a state image template for use in subsequent processes for state comparison and consistency verification.

[0007] A further technical improvement of the present invention is that the comparison module performs a state comparison step, including: The state-aware data fields are matched one-to-one with the state mirror template fields according to the field names to form a state parameter field mapping set; Based on the state parameter field type in the state parameter field lookup set, the tolerance parameter corresponding to each state parameter field is retrieved. Among them, the size-type state parameter field adopts static fixed tolerance, and the pose-type state parameter field adopts dynamic adaptive tolerance. The dynamic adaptive tolerance is adjusted in real time according to the control reliability index and equipment vibration data. Based on the historical error statistics of each state parameter field and the predefined criticality information of the state parameter field, a weight coefficient is assigned to each state parameter field. The criticality information is predefined by the process field template and associated with the corresponding field name. The difference between the values ​​of each state-aware data field and the corresponding state mirror template field is calculated, and a weighted comparison is performed by combining the tolerance parameter and the weight coefficient. The field-level state comparison result field and the overall state comparison result field are output.

[0008] A further technical improvement of the present invention is that: after generating the state comparison results, when the comparison module detects that any field of the state comparison result exceeds the corresponding tolerance parameter threshold, it performs the following response processing steps: Based on the keyness information of the fields corresponding to the state comparison result fields, the abnormal state comparison result fields are classified and marked. The marking level includes abnormality of key state comparison result fields and abnormality of non-key state comparison result fields. For cases where the key status comparison result fields are marked as abnormal, a freeze suggestion object is generated. The freeze suggestion object includes the status comparison result field that triggers the freeze, the difference between the corresponding status perception data fields, and the abnormality level information. The freeze suggestion object is then sent to the control module. Generate an anomaly event object for state comparison. The anomaly event object for state comparison automatically associates the state perception data field, state image template field and control confidence index corresponding to the anomaly state comparison result field, and establishes a path for tracing the source of the comparison failure. Based on the tracing path, a suspend command is sent to the upstream state perception module and the image construction module to suspend the data refresh of the data acquisition channel of the state perception data field corresponding to the abnormal state comparison result field, until the unfreeze command issued by the control module is received, and then the data refresh process of the corresponding state perception data field is resumed.

[0009] A further technical improvement of the present invention is that the step of the verification module generating the reverse state path includes: Obtain the current process step number and read the equipment shaft structure parameters and tool installation parameters corresponding to the process step number from the equipment parameter library; Based on the equipment shaft system structural parameters and control command parameters, the inverse kinematics solution rules for the equipment are constructed, and the theoretical inverse trajectory parameter set is calculated using the inverse kinematics analytical method. The difference between the theoretical inverse trajectory parameter set and the post-execution state perception data is calculated, and the theoretical inverse trajectory parameter set is matched and corrected based on the least squares error fitting method to form the inverse state parameter set; The reverse state parameter set is organized into a reverse state path according to the field order defined in the process field template.

[0010] A further technical improvement of the present invention is that the step of the verification module generating the reversibility determination result includes: Extract the values ​​of each state parameter field in the reverse state path and the corresponding field values ​​in the state-aware data before the execution of control instructions; Calculate the difference for each corresponding field value to obtain the field-level reversible error value; Each field-level reversible error value is compared with a preset reversible error threshold. When all field-level reversible error values ​​are less than the corresponding reversible error threshold, the reversibility determination result is generated as reversible; otherwise, the reversibility determination result is generated as irreversible.

[0011] Compared with the prior art, the present invention has the following beneficial effects: This invention realizes information fusion and control logic closed loop in multiple key dimensions of the manufacturing execution process; in particular, the index generation module combines multi-source information such as control residuals, time synchronization deviations and abnormal record frequency to form residual factor groups and generate control reliability index, which helps to perceive the stability and reliability of system operation in advance, provides quantitative reference for subsequent judgment, and significantly enhances the predictability and robustness before the control command is issued. Furthermore, this invention achieves accurate mapping and real-time comparison between the current process state and the target process template through the cooperation of the mirror construction module and the comparison module; using the state mirror template represented in the form of a structured parameter set, it can achieve field-level consistency comparison, and combine dynamic adaptive tolerance and field weight to achieve sensitive detection of differences; once the deviation of a key field is found to exceed the set threshold, the system can generate a freeze suggestion object and trigger a suspension mechanism to ensure that abnormal data no longer interferes with subsequent judgments, thereby improving the efficiency of abnormal response and data accuracy in the process flow. On the other hand, after receiving the judgment results from multiple dimensions, the control module in this invention can make a unified decision on the issuance of control commands and the adjustment of process status based on the joint control threshold rule of "freezing if any condition is not met", thus avoiding process risks caused by errors in judgment of a single dimension. Combined with the analysis of the reversibility of control actions by the verification module, the system not only supports static comparison judgment, but also makes a safety assessment on whether the process after execution has the ability to recover, effectively improving the intelligence level, safety assurance capability and operational stability of the overall manufacturing process. Attached Figure Description

[0012] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0013] Figure 1 This is a schematic diagram of the system logic of the present invention. Detailed Implementation

[0014] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0015] Please see Figure 1 As shown, this invention provides an Internet of Things-based data sensing and execution system for office furniture manufacturing, comprising: The index generation module is used to collect control command parameters, state awareness data and historical execution data, construct a residual factor set including control residuals, time synchronization deviations and abnormal record frequencies, and generate a control reliability index through mapping rules. The index generation module assesses the reliability of the current control state by combining the deviations before and after the execution of the current control command (control residuals), the synchronization accuracy of sensor sampling time (time synchronization deviation), and the statistical frequency of historical control anomalies (anomaly record frequency), forming a multi-dimensional residual factor set. This module quantifies each factor using predefined mapping rules (such as linear weighting, fuzzy rule sets, or lookup table functions), ultimately outputting a control reliability index that can be used for subsequent control decisions, serving as a fundamental reference indicator for reliable control throughout the entire process.

[0016] The steps for the index generation module to generate a control credibility index include: Collect control command parameters, status awareness data, and historical execution data for the current process, and construct control residuals based on the difference between control command parameters and status awareness data; Obtain the timestamp associated with the state-aware data and calculate the difference with the system's unified time base to construct the time synchronization deviation; Extract control anomaly marker records from historical execution data and count the frequency of anomaly records per unit time. The control residual, time synchronization deviation and anomaly recording frequency are combined with multiple dimensions of data, including at least the historical drift of the sensor, the real-time load rate of the equipment and the environmental temperature and humidity parameters, to form a residual factor group. The residual factor group is input into the preset mapping rule, which is constructed in the form of linear weighted summation. By setting corresponding weight coefficients for each residual factor, the weighting process is performed, and the final output is the control confidence index.

[0017] In the specific implementation process, when the index generation module performs the operation of generating the control reliability index, it first collects the control command parameters issued by the process control unit under the current process, as well as the status perception data obtained through the status perception device. The status perception data includes the actual measured size parameters, pose parameters and real-time process parameters after the workpiece is processed. At the same time, it calls the historical execution data recorded in the manufacturing execution system. The historical execution data includes the control command records, execution feedback information and abnormal marking logs generated in the past execution of this process.

[0018] Based on the one-to-one mapping relationship between the collected control command parameters and the corresponding state sensing data, the actual execution deviation of each corresponding field is calculated to form the control residual. The control residual is calculated by subtracting the sensing field value from the control command field value and taking its absolute value or square value for unified dimension processing.

[0019] The system calls the acquisition timestamp attached to the status perception data and calculates the difference with the unified time base set by the system to construct the time synchronization deviation. The unified time base comes from the standard clock signal inside the main controller. All sensing devices must complete clock synchronization within the set period. If the deviation exceeds the set threshold, it is included as an abnormal factor.

[0020] Extract control events with clear anomaly markers from historical execution data, count the number of anomalies in the past 10 similar processes, divide by the time window length to obtain the anomaly record frequency, which serves as a negative indicator for assessing control stability.

[0021] Read the additional environmental factors corresponding to the current sensor number, equipment number, and process environment number, including the sensor's historical drift (expressed as the maximum zero drift within the calibration period), the equipment's real-time load rate (taken from the current power percentage of the servo control unit), and the environmental temperature and humidity parameters (from the field monitoring nodes). All of the above factors, together with the control residual, time synchronization deviation, and frequency of abnormal records, constitute the residual factor group.

[0022] Finally, the residual factor group is input into the preset mapping rule. The mapping rule adopts a linear weighted summation form. The factor weight coefficients are set in advance through process experience or obtained based on regression modeling of a large number of historical samples. Different weight configuration tables can be called for different process scenarios.

[0023] During the weighting process, the product of each residual factor is calculated according to the weight coefficient, and all product values ​​are added together to form a cumulative value. Then, after standardization (such as normalization to the 0-1 range), a control confidence index is generated. The closer the control confidence index value is to 1, the more reliable the current process control status is; otherwise, it indicates that there is a greater risk of process deviation.

[0024] The mirror construction module is used to build a state mirror template based on process path parameters, target size parameters, target pose parameters and process step numbers. The state mirror template represents the target state in the form of a structured parameter set, including size parameters, pose parameters and process attribute parameters. The core of the mirroring module lies in constructing standardized templates of ideal process states for subsequent comparison and verification. Based on the specific process path, combined with the target product's geometric dimensions, installation orientation requirements, and current process step, this module extracts preset parameter field templates from the database to generate a structured set of target states. This mirroring module not only includes geometric shapes but also covers key process attributes such as pressure, speed, and temperature, enabling multi-dimensional consistency comparisons across different process stages.

[0025] The image build module provides methods for building state image templates, including: The process field template is retrieved based on the current process step number. The process field template predefines the logical order and naming rules of the target size parameter field, target pose parameter field, and process attribute parameter field corresponding to each process step. Read the target size parameters and target pose parameters corresponding to the current process path parameters, and collect the material number, tool type and equipment number of the current processing object to form an initial parameter set; The initial parameter set is input into the process rule library, which includes a material-process parameter mapping table, a set of equipment constraints and tool adaptation rules. Based on this rule library, process attribute parameters that conform to the current processing conditions are generated. Based on the field order and field name defined in the process field template, the target size parameters, target pose parameters and process attribute parameters are merged and organized into a structured parameter set. The structured parameter set is constructed in the form of a table, where the fields are represented in the form of key-value pairs and can be indexed by a unique process step number. The structured parameter set is stored as a state image template for use in subsequent processes for state comparison and consistency verification.

[0026] Specifically, based on the current process step number, the process field template is retrieved. This process field template is a pre-defined structured description file, which defines target dimension parameter fields, target pose parameter fields, and process attribute parameter fields that correspond one-to-one with each process step number, and sets the logical order, data type, and field naming rules for each type of field. The process field template is stored in the form of a structured parameter file and supports indexing and retrieval by process step number.

[0027] The target dimension and target pose parameters for the current step are obtained by parsing the current process path parameters. These, along with the material identification number, tool model parameters, and manufacturing equipment identification number collected during the manufacturing process, form the initial parameter set. Taking the edge banding process of an office desk as an example, the target dimension parameters might be "edge banding length = 1200mm" and "edge banding width = 18mm," while the target pose parameters might be "angle = 90°" and "position vector = (x, y, z)." Simultaneously, the material number of the processed board "PB18-A1," the currently installed tool "Spiral-TCT-12," and the edge banding equipment number "EDG-03" are collected.

[0028] The initial parameter set is input into the process rule base for attribute parameter calculation. The process rule base contains multiple independent rule sets, including: a material-process parameter mapping table (e.g., the heat sealing temperature range corresponding to material number "PB18-A1" is 180–200℃), a set of equipment constraints (e.g., the maximum feed rate of equipment number "EDG-03" is 30m / min), and tool adaptation rules (e.g., the optimal edge sealing pressure time corresponding to tool model "Spiral-TCT-12" is 1.8–2.3 seconds). The mirror construction module performs joint matching on the input initial parameter set. Under the premise of satisfying all rule constraints, a priority-driven parameter interpolation mechanism is used to complete the parameters. Specifically, when the target value is an interval, the median value with the lowest risk is selected (e.g., the median value of 190℃ is taken for 180–200℃); when the parameter depends on multiple condition variables, the material priority is used for priority determination before the equipment restricts the cutting.

[0029] Based on the field order and names defined in the aforementioned process field template, the generated target dimension parameters, target pose parameters, and process attribute parameters are merged to construct a structured parameter set. The structured parameter set is represented in a table structure, with each field organized as a key-value pair, such as "edge sealing temperature" = 190.0℃, "feed speed" = 25.0m / min, and "pressure application time" = 2.1s. This structured parameter set supports indexing and retrieval via a unique process step number, and each field is bound to a unique field identification tag and a data source identifier.

[0030] The constructed structured parameter set is encapsulated and stored as a state image template. During subsequent process execution, the state image template will be called by the comparison module for field-level parameter consistency comparison with real-time state-aware data, and also serves as a reference benchmark for determining the reversibility of control commands in the verification module. The state image template is stored using a distributed key-value database and equipped with field-level version control and access restriction mechanisms to ensure its traceability and real-time performance in complex multi-device collaborative manufacturing environments.

[0031] The comparison module is used to compare the state-aware data with the state image template by corresponding parameter fields, calculate the field difference based on the preset tolerance parameters and field weight coefficients, and generate the state comparison result. The comparison module is used to achieve accurate comparison between the perceived data and the ideal template. Its execution logic includes: first, matching the real-time collected state perception data with the mirror template field by field name or index structure; then, setting tolerance values ​​according to the process tolerance standards of each type of field, and assigning different weight coefficients according to the importance of the process fields; on this basis, performing difference calculation and weighted integration on the field-level data, and finally generating state comparison results that can be used to determine the degree of deviation of the process state.

[0032] The comparison module performs the following steps for state comparison: The state-aware data fields are matched one-to-one with the state mirror template fields according to the field names to form a state parameter field mapping set; Based on the state parameter field type in the state parameter field lookup set, the tolerance parameter corresponding to each state parameter field is retrieved. Among them, the size-type state parameter field adopts static fixed tolerance, and the pose-type state parameter field adopts dynamic adaptive tolerance. The dynamic adaptive tolerance is adjusted in real time according to the control reliability index and equipment vibration data. Based on the historical error statistics of each state parameter field and the predefined criticality information of the state parameter field, a weight coefficient is assigned to each state parameter field. The criticality information is predefined by the process field template and associated with the corresponding field name. The difference between the values ​​of each state-aware data field and the corresponding state mirror template field is calculated, and a weighted comparison is performed by combining the tolerance parameter and the weight coefficient. The field-level state comparison result field and the overall state comparison result field are output.

[0033] After generating the state comparison results, if the comparison module detects that any field of the state comparison result exceeds the corresponding tolerance parameter threshold, it will execute the following response processing steps: Based on the keyness information of the fields corresponding to the state comparison result fields, the abnormal state comparison result fields are classified and marked. The marking level includes abnormality of key state comparison result fields and abnormality of non-key state comparison result fields. For cases where the key status comparison result fields are marked as abnormal, a freeze suggestion object is generated. The freeze suggestion object includes the status comparison result field that triggers the freeze, the difference between the corresponding status perception data fields, and the abnormality level information. The freeze suggestion object is then sent to the control module. Generate an anomaly event object for state comparison. The anomaly event object for state comparison automatically associates the state perception data field, state image template field and control confidence index corresponding to the anomaly state comparison result field, and establishes a path for tracing the source of the comparison failure. Based on the tracing path, a suspend command is sent to the upstream state perception module and the image construction module to suspend the data refresh of the data acquisition channel of the state perception data field corresponding to the abnormal state comparison result field, until the unfreeze command issued by the control module is received, and then the data refresh process of the corresponding state perception data field is resumed.

[0034] Specifically, the comparison module matches the state-aware data fields with the state mirror template fields one by one according to field name, forming a state parameter field comparison set. During the matching process, not only are field names required to be consistent, but also the engineering units and data precision levels must be consistent. Field engineering unit consistency verification is implemented through field metadata, which records the engineering unit information corresponding to each field. The comparison module reads the field metadata during the matching phase; if a unit inconsistency is detected, it calls the unit conversion rules to convert the state-aware data field to an engineering unit consistent with the state mirror template field before including it in the comparison calculation, ensuring the accuracy and physical consistency of the calculation results.

[0035] Secondly, the system retrieves the corresponding tolerance parameter based on the field type of each state parameter field in the field comparison set. For size-related state parameter fields, the system uses a preset static fixed tolerance; while for pose-related state parameter fields, the system uses a dynamic adaptive tolerance. The calculation of this dynamic adaptive tolerance is based on the following functional relationship: the dynamic adaptive tolerance is determined by the weighted relationship between the base tolerance, the control confidence index, and the current vibration amplitude. When the control confidence index decreases or the vibration amplitude increases, the dynamic adaptive tolerance decreases according to a preset proportional function to improve comparison sensitivity. The specific formula is as follows: Dynamic tolerance = Basic tolerance × (α × Control reliability index + β × (1 − Maximum permissible vibration amplitude and current vibration amplitude)); Wherein, α and β are the system preset adjustment coefficients, the control reliability index ranges from 0 to 1, and the vibration amplitude is collected in real time by the vibration sensor at the device end.

[0036] The system further assigns weight coefficients to each field. These weight coefficients are composed of two factors: the historical error variance of the field and the criticality level defined in the process field template. Critical fields (such as drilling depth and splicing position) typically have higher weights. The system calculates the mean squared error of each field using historical process data and combines this with the criticality level defined in the process template to construct weight calculation rules, ultimately deriving the field-level weight coefficients.

[0037] Then, the comparison module performs a weighted comparison operation on each field. Specifically, this includes: calculating the difference between the field value of the state-aware data and the corresponding state mirror template field value, dividing it by the corresponding tolerance parameter to obtain the field normalized difference, and then multiplying this field normalized difference by the field weight coefficient to obtain the field-level weighted deviation value; finally, aggregating all field-level weighted deviation values ​​to form the overall state comparison result. The overall state comparison result is compared with the system's preset overall tolerance threshold. If the overall state comparison result exceeds the overall tolerance threshold, it is determined that the current process state does not meet the process requirements. The overall tolerance threshold is usually dynamically set based on historical data training models, product risk levels, or engineering standards, with a typical configuration of 1.0.

[0038] When any field-level comparison result exceeds the corresponding tolerance parameter threshold, the comparison module enters the anomaly response phase. The system classifies and marks abnormal fields based on their criticality information, with marking levels including critical field anomalies and non-critical field anomalies. Critical field anomalies indicate that the field plays a decisive role in the process flow; anomalies in this field will severely impact product quality or subsequent steps.

[0039] For cases marked as critical field anomalies, the comparison module generates a freeze suggestion object. The freeze suggestion object contains the following information: the name of the field triggering the freeze, the calculated difference between the fields, the criticality level, and the tolerance factor. This freeze suggestion object is then sent to the control module to support subsequent freeze determinations in the control process.

[0040] The system encapsulates this comparison anomaly event as a state comparison anomaly event object. This object automatically associates the state-aware data of the current field, the corresponding mirror template field value, and the control confidence index at the time of generation, thereby establishing a complete path for tracing the comparison failure. This path facilitates subsequent error analysis and process optimization.

[0041] To prevent continuous updates of abnormal data from interfering with system judgment, if a critical field anomaly occurs, the comparison module will send a suspension command to the data acquisition service after encapsulating the anomaly event. This command targets the sensor data channel associated with the abnormal field, suspending data acquisition and uploading for that channel until an unfreeze command is received from the control module, at which point the data refresh process resumes, ensuring data consistency and comparison validity. The suspension command only applies to the data acquisition channel corresponding to the currently anomaly field; data acquisition for other fields remains unchanged. The suspension mechanism pauses the refresh of data for that field until an unfreeze command is received from the control module, at which point the data acquisition process for that field resumes. This mechanism avoids system misjudgments caused by data contamination and helps maintain data consistency.

[0042] The verification module is used to generate a reverse state path by calling the preset process reverse path generation rules after the control command is executed, based on the control command parameters, post-execution state perception data and process step number, and calculate the difference between the reverse state path and the state perception data before the control command is executed to generate a reversibility determination result. The verification module is primarily used to verify the "reversibility" of the current control execution, thereby determining whether the executed action is safe, stable, and controllable. After the control action is executed, this module calls predefined reverse path generation rules (such as those based on inverse kinematics or state backtracking methods) using the process number and control parameters to calculate the theoretical pre-control state, and compares the difference between this reverse state path and the actual state perception data before execution. When the difference falls within the preset reversibility error threshold range, the execution is determined to be "reversible"; otherwise, it is determined to be "irreversible." This result is used for the final control strategy determination.

[0043] Specifically, the verification module reads the corresponding equipment axis structure parameters and tool installation parameters from the equipment parameter library based on the current process step number. The equipment axis structure parameters include the equipment's coordinate system information, kinematic parameters, and joint lengths, while the tool installation parameters include the tool's positioning information and installation method. These parameters provide the foundational data for subsequent inverse kinematics analysis.

[0044] The verification module uses the DH parameter method for inverse kinematics analysis to derive the angles or displacements of each joint. The DH parameter method constructs the kinematic equations of the mechanical system by defining parameters such as link length, torsion angle, and offset for each joint, forming a link transformation matrix (e.g., a 4x4 matrix). This matrix describes the coordinate transformation from the base coordinate system to the end effector. For multi-joint robotic arms or machine tools, the verification module solves for the angles of each joint by giving a target position (e.g., the workpiece pose).

[0045] For complex devices (such as six-axis robots), since their inverse kinematics equations lack a closed-form solution, the verification module employs a numerical iterative method (such as the Newton-Raphson method) to calculate the angle of each joint. Specifically, the error between the current guessed position and the target position is calculated, and the joint angles are adjusted to minimize this error. In each iteration, feedback corrections are used to obtain more accurate joint angles until the error falls below a predetermined threshold.

[0046] Subsequently, the verification module calculates the difference between the calculated theoretical inverse trajectory parameter set and the post-execution state perception data to obtain the error function. The error function is constructed as follows: Error Calculation: The verification module calculates the error between the theoretical trajectory and the actual executed trajectory. The error value includes position error and attitude error. The formula for error calculation is: sum the squares of the differences between each corresponding point between the theoretical trajectory and the actual perceived trajectory.

[0047] Error Function: The error function is a weighted combination of the sum of squares of position errors and the sum of squares of attitude errors. Position error and attitude error describe the deviations in position and attitude between the theoretical trajectory and the actual trajectory, respectively. Weighting coefficients are used to control the relative importance of position error and attitude error in the overall error.

[0048] The verification module optimizes the parameters of the reverse trajectory using the least squares method, as follows: Error calculation: The verification module calculates the error between the theoretical trajectory and the actual execution trajectory and establishes the target error function; Minimizing error: The calibration module optimizes parameters using least squares methods (such as gradient descent or normal equations). Gradient descent calculates the gradient of the error function with respect to the parameters and adjusts the parameters along the negative gradient direction to gradually approach the optimal solution. Normal equations obtain the optimal parameters by solving linear equations. Convergence determination: When the change in the error function is less than the preset threshold, the verification module considers the optimization process to have converged, obtains the optimal DH parameters, and thus corrects the theoretical trajectory to minimize the error between the theoretical trajectory and the actual trajectory.

[0049] Next, the verification module organizes the corrected reverse state parameter set according to the field order of the process field template to form a reverse state path. This path will serve as the basis for subsequent reversibility determination, and the verification module will use the reverse state path to determine whether the control commands meet the reversibility requirements.

[0050] Finally, the verification module compares the generated reverse state path with the state-aware data before the control command execution to determine reversibility. The steps include: Extracting field values: The verification module extracts the field values ​​of each state parameter in the reverse state path and compares them with the corresponding field values ​​in the state perception data before the control command is executed; Difference Calculation: The verification module performs difference calculation on each corresponding field value to obtain the field-level reversible error value.

[0051] Error Threshold Comparison: The verification module compares the reversible error values ​​at each field level with preset reversible error thresholds. These thresholds are typically set dynamically based on product quality requirements, historical data analysis, and process standards. When all field-level reversible error values ​​are less than their corresponding reversible error thresholds, the process step is determined to be reversible; otherwise, it is determined to be irreversible.

[0052] The control module is used to receive the control reliability index, state comparison results and reversibility judgment results, and make a joint judgment according to the control threshold rules. When any indicator fails to meet the control requirements, it prevents the issuance of control commands and sets the current process state to a frozen state. As the decision-making center of the entire system, the control module aggregates the core results from various analysis modules and executes unified control decisions accordingly. It first receives the control reliability index from the index generation module, the state comparison results output by the comparison module, and the reversibility judgment results given by the verification module. Then, it performs a joint judgment based on control threshold rules, which may employ a rigid strategy of "freezing if any fails." Once any indicator is detected as not meeting control requirements, the system immediately blocks the issuance of subsequent control commands and triggers the freeze mechanism, marking the current process state as frozen, awaiting anomaly resolution or manual intervention before resuming execution.

[0053] In one implementation, the control module, as the core decision-making unit of the entire manufacturing execution process, primarily performs joint judgment based on preset control threshold rules after receiving multiple key judgment results. This determines whether to allow the normal issuance of control commands or trigger a freeze mechanism to suspend the execution of the current process. The process includes the following steps: The control module receives judgment results from three independent modules, including: a control reliability index output by the index generation module, which reflects the reliability of the matching between the current control command and the process state; a state comparison result output by the comparison module, which is used to evaluate the consistency between the current state perception data and the preset process target; and a reversibility judgment result generated by the verification module, which is used to determine whether the current control action has theoretical reversibility to ensure operational safety.

[0054] The control module performs a joint judgment on the above three indicators based on predefined control threshold rules. These control threshold rules can employ various strategies; in this embodiment, a "freeze if any one fails" judgment logic is used. Specifically, when the control confidence index is detected to be lower than the preset threshold, or when a key field deviation in the state comparison result exceeds the tolerance range, or when the reversibility judgment result is irreversible, the control module considers the current process control state to be unsatisfactory.

[0055] When any of the above indicators fails to meet the control requirements, the control module will execute a freeze response. The freeze response includes the following sub-steps: the control module immediately prevents the control command corresponding to the current process step from being sent to the equipment controller to avoid potentially risky operations; at the same time, the current process status is set to "frozen state" and broadcast to all task scheduling and status processing modules through a unified freeze status identifier, so that the entire data link is aware of the current freeze status.

[0056] Finally, after the freeze state is set, the control module activates the freeze tracking mechanism, recording the core judgment results, abnormal fields, and corresponding values ​​when the freeze occurs, forming a freeze event log for subsequent tracing and fault analysis. When the abnormality release conditions are met (e.g., the new control confidence index rises to the confidence range, the state comparison results return to consistency, the reversibility judgment is passed, etc.), or the freeze is explicitly lifted through manual operation, the control module restores the signal to the control channel, re-allows the issuance of control commands, and restores the process status from "frozen" to "executable," thereby achieving closed-loop safety control of the manufacturing process.

[0057] For example, in the edge banding process of an office desk, the control module receives the following indicators: the control reliability index is 0.72 (higher than the preset threshold of 0.65), the comparison deviation of the "edge banding angle" field in the status comparison result is 1.2° (exceeding the tolerance of 1.0°), and the reversibility judgment result is "reversible". According to the control rule of "freeze if any fails", the control module determines that the status comparison has failed, triggers the freeze mechanism, prevents the edge banding command from being issued, and sends a freeze alarm prompt to the operation panel. Execution can only continue after the edge banding angle is adjusted to the tolerance range and re-verified.

[0058] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0059] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. An Internet of Things-based data sensing and execution system for office furniture manufacturing, characterized in that, include: The index generation module is used to collect control command parameters, state awareness data and historical execution data, construct a residual factor set including control residuals, time synchronization deviations and abnormal record frequencies, and generate a control reliability index through mapping rules. The mirror construction module is used to construct a state mirror template based on process path parameters, target size parameters, target pose parameters and process step numbers. The state mirror template represents the target state in the form of a structured parameter set, including size parameters, pose parameters and process attribute parameters. The comparison module is used to compare the state-aware data with the state image template by corresponding parameter fields, calculate the field difference based on the preset tolerance parameters and field weight coefficients, and generate the state comparison result. The verification module is used to generate a reverse state path by calling the preset process reverse path generation rules after the control command is executed, based on the control command parameters, post-execution state perception data and process step number, and to calculate the difference between the reverse state path and the state perception data before the control command is executed, thereby generating a reversibility determination result. The control module receives the control reliability index, state comparison results, and reversibility determination results, and makes a joint judgment according to the control threshold rules. When any indicator fails to meet the control requirements, it prevents the issuance of control commands and sets the current process state to a frozen state.

2. The Internet of Things-based office furniture manufacturing data sensing and execution system according to claim 1, characterized in that, The steps for the index generation module to generate a control credibility index include: Collect control command parameters, status awareness data, and historical execution data for the current process, and construct control residuals based on the difference between control command parameters and status awareness data; Obtain the timestamp associated with the state-aware data and calculate the difference with the system's unified time base to construct the time synchronization deviation; Extract control anomaly marker records from historical execution data and count the frequency of anomaly records per unit time. The control residual, time synchronization deviation and anomaly recording frequency are combined with multiple dimensions of data, including at least the historical drift of the sensor, the real-time load rate of the equipment and the environmental temperature and humidity parameters, to form a residual factor group. The residual factor set is input into the preset mapping rule, and the final output is the control confidence index.

3. The Internet of Things-based office furniture manufacturing data sensing and execution system according to claim 2, characterized in that, The mapping rule is constructed using a linear weighted summation method, and weighting is performed by setting corresponding weight coefficients for each residual factor.

4. The Internet of Things-based office furniture manufacturing data sensing and execution system according to claim 1, characterized in that, The method for constructing a state image template by the image construction module includes: The process field template is retrieved based on the current process step number. The process field template predefines the logical order and naming rules of the target size parameter field, target pose parameter field, and process attribute parameter field corresponding to each process step. Read the target size parameters and target pose parameters corresponding to the current process path parameters, and collect the material number, tool type and equipment number of the current processing object to form an initial parameter set; The initial parameter set is input into the process rule library, which includes a material-process parameter mapping table, a set of equipment constraints and tool adaptation rules. Based on this rule library, process attribute parameters that conform to the current processing conditions are generated. Based on the field order and field name defined in the process field template, the target size parameters, target pose parameters and process attribute parameters are merged and organized into a structured parameter set. The structured parameter set is constructed in the form of a table, where the fields are represented in the form of key-value pairs and can be indexed by a unique process step number. The structured parameter set is stored as a state image template for use in subsequent processes for state comparison and consistency verification.

5. The Internet of Things-based office furniture manufacturing data sensing and execution system according to claim 1, characterized in that, The comparison module performs the state comparison step, including: The state-aware data fields are matched one-to-one with the state mirror template fields according to the field names to form a state parameter field mapping set; Based on the state parameter field type in the state parameter field lookup set, the tolerance parameter corresponding to each state parameter field is retrieved. Among them, the size-type state parameter field adopts static fixed tolerance, and the pose-type state parameter field adopts dynamic adaptive tolerance. The dynamic adaptive tolerance is adjusted in real time according to the control reliability index and equipment vibration data. Based on the historical error statistics of each state parameter field and the predefined criticality information of the state parameter field, a weight coefficient is assigned to each state parameter field. The criticality information is predefined by the process field template and associated with the corresponding field name. The difference between the values ​​of each state-aware data field and the corresponding state mirror template field is calculated, and a weighted comparison is performed by combining the tolerance parameter and the weight coefficient. The field-level state comparison result field and the overall state comparison result field are output.

6. The Internet of Things-based office furniture manufacturing data sensing and execution system according to claim 5, characterized in that, After generating the state comparison results, if the comparison module detects that any field of the state comparison result exceeds the corresponding tolerance parameter threshold, it performs the following response processing steps: Based on the keyness information of the fields corresponding to the state comparison result fields, the abnormal state comparison result fields are classified and marked. The marking level includes abnormality of key state comparison result fields and abnormality of non-key state comparison result fields. For cases where the key status comparison result fields are marked as abnormal, a freeze suggestion object is generated. The freeze suggestion object includes the status comparison result field that triggers the freeze, the difference between the corresponding status perception data fields, and the abnormality level information. The freeze suggestion object is then sent to the control module. Generate an anomaly event object for state comparison. The anomaly event object for state comparison automatically associates the state perception data field, state image template field and control confidence index corresponding to the anomaly state comparison result field, and establishes a path for tracing the source of the comparison failure. Based on the tracing path, a suspend command is sent to the upstream state perception module and the image construction module to suspend the data refresh of the data acquisition channel of the state perception data field corresponding to the abnormal state comparison result field, until the unfreeze command issued by the control module is received, and then the data refresh process of the corresponding state perception data field is resumed.

7. The Internet of Things-based office furniture manufacturing data sensing and execution system according to claim 1, characterized in that, The steps for the verification module to generate the reverse state path include: Obtain the current process step number and read the equipment shaft structure parameters and tool installation parameters corresponding to the process step number from the equipment parameter library; Based on the equipment shaft system structural parameters and control command parameters, the inverse kinematics solution rules for the equipment are constructed, and the theoretical inverse trajectory parameter set is calculated using the inverse kinematics analytical method. The difference between the theoretical inverse trajectory parameter set and the post-execution state perception data is calculated, and the theoretical inverse trajectory parameter set is matched and corrected based on the least squares error fitting method to form the inverse state parameter set; The reverse state parameter set is organized into a reverse state path according to the field order defined in the process field template.

8. The Internet of Things-based office furniture manufacturing data sensing and execution system according to claim 7, characterized in that, The step of the verification module generating the reversibility determination result includes: Extract the values ​​of each state parameter field in the reverse state path and the corresponding field values ​​in the state-aware data before the execution of control instructions; Calculate the difference for each corresponding field value to obtain the field-level reversible error value; Each field-level reversible error value is compared with a preset reversible error threshold. When all field-level reversible error values ​​are less than the corresponding reversible error threshold, the reversibility determination result is generated as reversible; otherwise, the reversibility determination result is generated as irreversible.

Images