A process monitoring system collaborative optimization method, system, device and medium

Abstract

The application discloses a process monitoring system cooperative optimization method, system, device and medium, and belongs to the field of industrial internet and intelligent manufacturing. The method is as follows: obtaining running data of multiple function modules in a monitoring system of a target process, calculating state characteristic values of the function modules and calculating a coupling coordination degree of the monitoring system; when the coupling coordination degree is lower than a preset threshold, identifying a weak function module with the lowest state characteristic value; locating a target process node associated with the weak function module on a pre-constructed process correlation graph and tracing an influence path thereof; and matching a system optimization scheme according to the target process node, the influence path and the weak function module and optimizing the monitoring system. Therefore, the optimization accuracy of the process monitoring system can be improved by implementing the application.

Description

Technical Field

[0001] This application relates to the field of industrial internet and intelligent manufacturing, and in particular to a collaborative optimization method, system, equipment and medium for process monitoring systems. Background Technology

[0002] With the deep integration of the Industrial Internet and intelligent manufacturing, the process monitoring system covers multiple functional modules such as sensor acquisition, data transmission, protocol conversion and intelligent reasoning. The operating status of each module directly affects the monitoring quality and response speed. When the module performance is insufficient or the configuration is incorrect, it will lead to monitoring delay, data loss or abnormal missed reports. It is urgent to systematically optimize the monitoring system.

[0003] Existing technologies for optimizing process monitoring systems typically employ two fragmented approaches: one method monitors the performance of each functional module independently, triggering parameter adjustments or restarts based on preset thresholds. Because these adjustments are independent, local adjustments in one module can conflict with the operational status of others, leading to a decline in overall monitoring effectiveness and a prolonged, unresolved state of fluctuation. The other method collects comprehensive system data for evaluation, outputting a score characterizing system health or synergy. This score serves as statistical information for management reference. However, this approach remains at the evaluation level, failing to correlate the score with specific functional execution steps or explain underlying shortcomings. Optimization decisions still rely on manual module-by-module checks and experience-based judgment, resulting in lengthy optimization response cycles and hindering the formation of an effective system improvement loop. Therefore, when existing technologies identify areas for optimization in the process monitoring system, they fail to define clear optimization directions, preventing optimization measures from precisely targeting system weaknesses and leading to low accuracy in process monitoring system optimization. Summary of the Invention

[0004] This application provides a collaborative optimization method, system, equipment, and medium for a process monitoring system, which can solve the problem in the prior art where the optimization direction cannot be clearly defined, resulting in optimization measures not being able to accurately target the real system shortcomings and the low optimization accuracy of the process monitoring system.

[0005] Some embodiments of this application provide a collaborative optimization method for a process monitoring system, including: The system acquires the operating data of multiple functional modules in the monitoring system of the target process, calculates the state characteristic value of each functional module based on the operating data, and calculates the coupling coordination degree of the monitoring system based on each state characteristic value. When the coupling coordination degree is lower than a preset threshold, weak functional modules are identified from the functional modules based on the state characteristic values ​​of each functional module; Obtain a pre-constructed process association graph of the target process, locate the target process node associated with the weak functional module on the process association graph, and trace the influence path of the target process node along the process association graph; Based on the target process node, the impact path, and the weak functional module, a system optimization scheme is matched, and the monitoring system is optimized according to the system optimization scheme.

[0006] Compared with existing technologies, the above embodiments have the following beneficial effects: By acquiring the operating data of multiple functional modules in the monitoring system of the target process, the state characteristic values ​​of each functional module are first calculated, and the overall coupling coordination degree of the monitoring system is calculated accordingly. When the coupling coordination degree is lower than a preset threshold, instead of directly adjusting the system as a whole or checking each module individually, the weak functional module with the lowest state characteristic value is identified from multiple functional modules based on the state characteristic values ​​of each functional module. Then, a pre-constructed process association map is acquired, the target process node associated with the weak functional module is located on the map, and the influence path of the target process node is traced, thereby clarifying which process links and the scope of influence the weak functional module specifically affects. Finally, a system optimization scheme is matched based on the weak functional module, the target process node, and the influence path. Compared to existing technologies that rely on independent adjustments to each module, leading to repeated fluctuations or simply outputting evaluation scores without pinpointing weaknesses and relying on manual investigation, this application determines whether the overall system needs optimization by assessing the degree of coupling coordination. When optimization is required, it further utilizes the ranking results of the state characteristic values ​​of each module to accurately locate the weak module. Then, it combines the process correlation diagram to determine the influence boundary of the weak module, enabling the optimization scheme to target the real system weaknesses without exceeding the necessary range. This improves the optimization accuracy of the process monitoring system and avoids the repeated effects caused by blind adjustments.

[0007] Further, the step of calculating the state characteristic values ​​of each functional module based on the operational data, and calculating the coupling coordination degree of the monitoring system based on each state characteristic value, includes: Based on the operational data, several status indicators corresponding to each functional module are determined, and the first weight coefficient of each status indicator corresponding to each functional module is determined by the entropy weight method. Based on the first weighting coefficient of each state indicator, the state indicators of each functional module are weighted and calculated to obtain the state feature value of each functional module. Based on the state characteristic values, calculate the system coupling degree, which characterizes the interaction strength between the functional modules, and calculate the coordination index, which characterizes the balance of the development level of the functional modules. The coupling and coordination degree of the monitoring system is calculated based on the system coupling degree and the coordination degree index.

[0008] Compared with existing technologies, the above embodiments have the following beneficial effects: When calculating the state characteristic values ​​of each functional module, several corresponding state indicators are first determined based on the operating data of each functional module. The first weight coefficient of each state indicator corresponding to each functional module is determined using the entropy weight method. Then, the state indicator is weighted according to each first weight coefficient to obtain the state characteristic value of each functional module. Compared with the method of simply averaging or assigning weights based on experience, the entropy weight method determines the weight based on the degree of variation of each state indicator among different samples. The greater the degree of variation of the indicator, the higher the weight, reflecting the more distinguishing information the indicator carries in actual operation. This makes the calculated state characteristic values ​​more objectively reflect the actual operating level of each functional module, avoiding the problem of inaccurate identification of weak functional modules due to subjective weighting bias, and providing a reliable data foundation for accurate identification of weak modules in the future.

[0009] Further, the calculation of the system coupling degree, which characterizes the strength of interaction between the functional modules, and the coordination index, which characterizes the degree of balance in the development level of the functional modules, based on the state characteristic values, includes: Calculate the geometric mean and arithmetic mean of each state characteristic value, and determine the system coupling degree based on the ratio of the geometric mean and the arithmetic mean; Obtain the second weight coefficient of each functional module, and weight each state feature value according to the second weight coefficient to obtain the coordination index.

[0010] Compared with existing technologies, the above embodiments have the following beneficial effects: When calculating the system coupling degree, the geometric mean and arithmetic mean of each state characteristic value are calculated, and the system coupling degree is determined based on the ratio of the geometric mean to the arithmetic mean; when calculating the coordination degree index, the second weight coefficient of each functional module is obtained, and the coordination degree index is obtained by weighted summation of each state characteristic value based on each second weight coefficient. The ratio of the geometric mean to the arithmetic mean can reflect the degree of dispersion between the state characteristic values ​​of each functional module. When the state characteristic values ​​of each module tend to be balanced, the ratio is close to 1, and when the difference is large, the ratio is significantly less than 1, thereby quantitatively characterizing the interaction strength between modules; while the coordination degree index reflects the balance of the overall development level of each module through weighted summation; both provide a clear quantitative basis for the subsequent calculation of coupling coordination degree from the two dimensions of balance and wholeness, respectively, so that the coupling coordination degree can comprehensively reflect the collaborative state between the modules of the monitoring system.

[0011] Furthermore, the functional modules include: an infrastructure module, a data resource module, a protocol standard module, and an intelligent algorithm module; the step of determining several status indicators corresponding to each of the functional modules based on the operational data includes: Acquire sensor deployment data and edge node operation data of the infrastructure module, determine sensor coverage status indicators based on the sensor deployment data, and determine edge computing capability status indicators based on the edge node operation data; Acquire multi-source data collection data and data transmission data from the data resource module, determine the data fusion rate status index based on the multi-source data collection data, and determine the latency status index based on the data transmission data; Obtain the interface call data and protocol conversion data of the protocol standard module, determine the interface standardization status index based on the interface call data, and determine the protocol conversion efficiency status index based on the protocol conversion data. The model inference data and model update data of the intelligent algorithm module are obtained. The detection accuracy status index is determined based on the model inference data, and the model update frequency status index is determined based on the model update data.

[0012] Compared to existing technologies, the above embodiments have the following advantages: This application divides the functional modules into four dimensions: infrastructure module, data resource module, protocol standard module, and intelligent algorithm module. For each dimension, corresponding operational data is acquired and corresponding status indicators are determined. The infrastructure module corresponds to sensor coverage and edge computing capabilities; the data resource module corresponds to data fusion rate and latency; the protocol standard module corresponds to interface standardization and protocol conversion efficiency; and the intelligent algorithm module corresponds to detection accuracy and model update frequency. This four-dimensional division covers the entire chain of the process monitoring system, from underlying hardware deployment, data acquisition and transmission, protocol interface interoperability to upper-layer intelligent inference. The status indicators for each dimension are selected from quantifiable parameters that reflect the core operational capabilities of that dimension, ensuring that the status evaluation of each functional module has a clear data source and calculation basis, avoiding evaluation bias caused by incomplete dimension division or inappropriate indicator selection.

[0013] Further, the process association graph includes: several process nodes, process dependencies between the process nodes, and support relationships between the process nodes and the functional modules; locating the target process node associated with the weak functional module on the process association graph, and tracing the influence path of the target process node along the process association graph, includes: Based on the support relationship, query the target process node supported by the weak functional module from the process nodes; Following the process dependency relationship, trace downstream process nodes from the target process node towards the output direction to obtain the downstream influence range of the target process node; Following the process dependency relationship, trace upstream process nodes from the target process node towards the input direction to obtain the upstream cause range of the target process node; The impact path of the target process node is determined based on the downstream impact range and the upstream cause range.

[0014] Compared to existing technologies, the above embodiments have the following beneficial effects: When tracing the impact path, the target process node supported by the weak functional module is first queried from the process nodes based on the support relationship. Then, the downstream and upstream process nodes are traced from the target process node in the output and input directions respectively along the process dependency relationship to obtain the downstream impact range and the upstream cause range, thereby determining the impact path. The above method explicitly models the support relationship between functional modules and process nodes through a graph structure, so that once the weak functional module is identified, the specific process link supported by the module can be quickly located. Then, based on the dependency relationship between processes, the upstream and downstream are traced to clarify the impact range of the weak module on subsequent processes and the upstream process cause that leads to the weakness. This fully presents the actual impact boundary of the weak functional module on the process chain, providing a clear basis for the scope of action for subsequent optimization scheme matching.

[0015] Furthermore, the step of matching a system optimization scheme based on the target process node, the impact path, and the weak functional module, and optimizing the monitoring system according to the system optimization scheme, includes: Based on the weak functional module, query the candidate optimization scheme corresponding to the weak functional module from the preset optimization scheme library; Based on the target process node and the influence path, determine the scope of application of the candidate optimization scheme; Based on the scope of application, a target optimization scheme is determined from the candidate optimization schemes; The target optimization scheme is distributed to edge computing nodes or cloud servers within the scope of the system to adjust the operating parameters of the monitoring system.

[0016] Compared to existing technologies, the above embodiments have the following beneficial effects: When matching optimization schemes, firstly, candidate optimization schemes are queried from a preset optimization scheme library based on the weak functional modules. Then, the scope of application of the candidate optimization schemes is determined based on the target process node and the influence path. Based on this scope, the target optimization scheme is determined from the candidate optimization schemes. Finally, the target optimization scheme is distributed to edge computing nodes or cloud servers within the scope of application to adjust the operating parameters of the monitoring system. This method narrows the matching range of schemes by the type of the weak functional module, and then determines the process boundaries that the optimization measures need to cover by combining the influence path. This ensures that the final distributed optimization scheme is both targeted at the type of weak module and has a clearly defined execution scope, avoiding the problem of insufficient effect or over-adjustment caused by a mismatch between the optimization scheme and the scope of influence.

[0017] Further, the step of identifying weak functional modules from the functional modules based on the state characteristic values ​​of each functional module includes: The state feature values ​​of each functional module are sorted to obtain a state feature value sequence; The target functional module with the lowest state feature value is selected from the sequence of state feature values; The target functional module is identified as the weak functional module.

[0018] Compared with existing technologies, the above embodiments have the following advantages: When identifying weak functional modules, the state feature values ​​of each functional module are sorted to obtain a sequence of state feature values. The target functional module with the lowest state feature value is then selected from the sequence and identified as the weak functional module. This method directly sorts and compares the calculated state feature values ​​of each functional module. The functional module with the lowest state feature value is the link that has the greatest constraint on the overall system coordination level under the condition of low coupling coordination. The identification process does not require additional judgment models or manual intervention, has low computational load, and responds quickly. It is suitable for real-time requirements in process monitoring scenarios, enabling the system to quickly locate modules that need priority improvement after detecting insufficient coupling coordination.

[0019] Another embodiment of this application provides a collaborative optimization system for a process monitoring system, which is used to implement the steps of the collaborative optimization method for a process monitoring system as described in this application.

[0020] Another embodiment of this application also provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps of the collaborative optimization method of the process monitoring system of this application.

[0021] Another embodiment of this application also provides a computer-readable storage medium item, including: a stored computer program, which, when the computer program is running, controls the device where the computer-readable storage medium is located to perform the steps of the collaborative optimization method of the process monitoring system of this application. Attached Figure Description

[0022] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 This is a flowchart illustrating a collaborative optimization method for a process monitoring system provided in some embodiments of this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0026] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0027] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0028] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0029] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0030] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0031] Existing technologies for optimizing process monitoring systems typically employ two fragmented approaches: one method monitors the performance of each functional module independently, triggering parameter adjustments or restarts based on preset thresholds. Because these adjustments are independent, local adjustments in one module can conflict with the operational status of others, leading to a decline in overall monitoring effectiveness and a prolonged, unresolved system fluctuation. The other method collects comprehensive system operation data for evaluation, outputting a score characterizing system health or synergy, which serves as statistical information for management. However, this method remains merely an evaluation, failing to correlate scores with specific functional execution steps or explain underlying shortcomings. Optimization decisions still rely on manual module-by-module checks and experience-based judgment, resulting in lengthy optimization response cycles and hindering the formation of an effective system improvement loop. Therefore, when existing technologies identify areas for optimization in the process monitoring system, they fail to define clear optimization directions, preventing optimization measures from precisely targeting system weaknesses and resulting in low accuracy in process monitoring system optimization.

[0032] Please refer to Figure 1 To address the problem in existing technologies where, when an optimization of a process monitoring system is identified as needing improvement, the direction of optimization cannot be clearly defined, leading to inaccurate targeting of optimization measures to the actual system weaknesses and low optimization precision in the process monitoring system, this application provides a collaborative optimization method for a process monitoring system, comprising the following steps S101 to S104: S101: Obtain the operating data of multiple functional modules in the monitoring system of the target process, calculate the state characteristic value of each functional module based on the operating data, and calculate the coupling coordination degree of the monitoring system based on each state characteristic value.

[0033] Furthermore, in some embodiments of this application, the functional modules include: an infrastructure module, a data resource module, a protocol standard module, and an intelligent algorithm module.

[0034] Furthermore, in some embodiments of this application, the step of calculating the state characteristic values ​​of each functional module based on the operating data, and calculating the coupling coordination degree of the monitoring system based on each state characteristic value, includes the following steps S1011 to S1014: S1011: Based on the operational data, determine several status indicators corresponding to each functional module, and determine the first weight coefficient of each status indicator corresponding to each functional module using the entropy weight method.

[0035] Furthermore, in some embodiments of this application, determining several status indicators corresponding to each functional module based on the respective operational data includes: Acquire sensor deployment data and edge node operation data of the infrastructure module, determine sensor coverage status indicators based on the sensor deployment data, and determine edge computing capability status indicators based on the edge node operation data; Acquire multi-source data collection data and data transmission data from the data resource module, determine the data fusion rate status index based on the multi-source data collection data, and determine the latency status index based on the data transmission data; Obtain the interface call data and protocol conversion data of the protocol standard module, determine the interface standardization status index based on the interface call data, and determine the protocol conversion efficiency status index based on the protocol conversion data. The model inference data and model update data of the intelligent algorithm module are obtained. The detection accuracy status index is determined based on the model inference data, and the model update frequency status index is determined based on the model update data.

[0036] In one implementation, taking the body assembly welding process in the welding workshop as an example, the aforementioned operational data refers to the original records actually generated by each functional module during process monitoring. Specifically, for the infrastructure module, its operational data includes sensor deployment data and edge node operational data. The sensor deployment data comes from the on-site sensor installation log, recording the number, type, and number of processes covered by the deployed sensors at each process node, from which the sensor coverage rate can be calculated. The edge node operational data comes from the hardware configuration and operation logs of the edge computing device, recording the processor specifications, memory capacity, actual computing throughput, and end-to-end response latency of each node. For the data resource module, its operational data includes multi-source data acquisition data and data transmission data. The multi-source data acquisition data comes from the data source access registration form of each process link, recording the number of data sources such as sensors, PLCs (Programmable Logic Controllers), and MES (Manufacturing Execution Systems) that have been connected, as well as the total number that should be connected. The data transmission data comes from the network communication log, recording the timestamp deviation, synchronization delay, and packet loss during the transmission of multi-source data. For the protocol standard module, its operational data includes interface call data and protocol conversion data. Interface call data comes from API (Application Programming Interface) call logs between subsystems, recording the number of calls and response statuses of standardized and non-standardized interfaces. Protocol conversion data comes from cross-protocol communication logs, recording the number of direct interoperability calls, the number of protocol conversion requests, and conversion failure records. For the intelligent algorithm module, its operational data includes model inference data and model update data. Model inference data comes from the algorithm model's actual inference logs of process data, recording input features, output results, confidence levels, and comparison results with real labels. Model update data comes from the model version management library, recording the actual number of updates per unit time, the preset number of updates, and changes in verification indicators.

[0037] Preferably, when determining the first weight coefficient, the entropy weight method described above processes the preprocessed set of valid evaluation samples. Before entering the entropy weight method calculation, time alignment, missing data completion, and validity verification are performed on the multi-source heterogeneous operating data: for short-term missing data, interpolation of adjacent time windows or historical averages of similar devices are used for completion; for data windows with continuously missing key fields that cannot be completed, they are marked as invalid samples and removed, and do not participate in subsequent calculations, thus obtaining the set of valid evaluation samples. Based on the operating data in the set of valid samples, each evaluation index is calculated to form a valid evaluation sample matrix. Let the valid evaluation sample matrix be X, where row vectors correspond to samples, column vectors correspond to state indicators, the total number of samples is n, and the total number of state indicators is m.

[0038] Preferably, in some embodiments of this application, the range standardization of each state index is first performed: for positive indices, the standardized value is the difference between the original value of the index and the minimum value of all samples divided by the difference between the maximum value and the minimum value of all samples; for negative indices, the standardized value is the difference between the maximum value of all samples and the original value divided by the difference between the maximum value and the minimum value; after standardization, a preset constant is uniformly shifted to eliminate the influence of zero values. Then, the proportion of each sample under each state index is calculated, that is, the ratio of the standardized value of the sample to the sum of the standardized values ​​of all samples. Next, the information entropy of each state index is calculated, which is the negative of the sum of the products of the proportion of each sample and its natural logarithm, multiplied by a constant k, where k is the reciprocal of the natural logarithm of the total number of samples; the smaller the information entropy, the greater the difference between samples and the more information the state index contains. Then, the difference coefficient is calculated, that is, the difference between 1 and the information entropy; the larger the difference coefficient, the greater the variation of the state index among samples and the stronger its discriminative ability. Finally, the difference coefficient of each state index is divided by the sum of the difference coefficients of all state indices to obtain the first weight coefficient of the corresponding state index.

[0039] This application divides the functional modules into four dimensions: infrastructure module, data resource module, protocol standard module, and intelligent algorithm module. For each dimension, corresponding operational data is acquired and relevant status indicators are determined. The infrastructure module corresponds to sensor coverage and edge computing capabilities; the data resource module corresponds to data fusion rate and latency; the protocol standard module corresponds to interface standardization and protocol conversion efficiency; and the intelligent algorithm module corresponds to detection accuracy and model update frequency. This four-dimensional division covers the entire chain of the process monitoring system, from underlying hardware deployment, data acquisition and transmission, protocol interface interoperability to upper-layer intelligent inference. The status indicators for each dimension are selected from quantifiable parameters that reflect the core operational capabilities of that dimension. This ensures that the status assessment of each functional module has a clear data source and calculation basis, avoiding assessment biases caused by incomplete dimension division or inappropriate indicator selection.

[0040] S1012: Based on the first weighting coefficient of each state indicator, the state indicators of each functional module are weighted and calculated to obtain the state characteristic value of each functional module.

[0041] For example, in one implementation, the weighted calculation is categorized according to the functional module dimension. Taking the infrastructure module as an example, its corresponding status indicators include sensor coverage, edge computing capability, and response latency. Each status indicator is calculated using the entropy weight method and corresponds to a first weight coefficient, such as sensor coverage 0.22, edge computing capability 0.25, and response latency 0.20. The standardized values ​​of each status indicator under this module are multiplied by their corresponding first weight coefficients and then summed to obtain the status feature value of the infrastructure module. Similarly, the data resource module multiplies the three status indicators of data fusion rate, synchronization latency, and data quality by their corresponding first weight coefficients and then sums them to obtain the status feature value of the module; the protocol standard module multiplies the protocol coverage, interface standardization degree, and protocol conversion efficiency by their corresponding weights and then sums them; the intelligent algorithm module multiplies the detection accuracy, model update frequency, and human-computer interaction efficiency by their corresponding weights and then sums them. Each functional module completes the weighted calculation in this way, and finally obtains the state characteristic values ​​of the infrastructure module, data resource module, protocol standard module and intelligent algorithm module respectively, which are used as inputs for subsequent calculation of the system coupling degree and coordination degree index.

[0042] S1013: Based on the state characteristic values, calculate the system coupling degree, which characterizes the interaction strength between the functional modules, and calculate the coordination index, which characterizes the balance of the development level of the functional modules.

[0043] Furthermore, in some embodiments of this application, the step of calculating the system coupling degree, which characterizes the strength of interaction between the functional modules, and the coordination index, which characterizes the degree of balance in the development level of the functional modules, based on the state characteristic values, includes: Calculate the geometric mean and arithmetic mean of each state characteristic value, and determine the system coupling degree based on the ratio of the geometric mean and the arithmetic mean; Obtain the second weight coefficient of each functional module, and weight each state feature value according to the second weight coefficient to obtain the coordination index.

[0044] For example, in one implementation, the system coupling degree reflects the dispersion and interaction strength between the state characteristic values ​​of each functional module. Specifically, the arithmetic mean of the state characteristic values ​​of each functional module is first calculated, i.e., the sum of all state characteristic values ​​divided by the total number of functional modules; then, the geometric mean of the state characteristic values ​​of each functional module is calculated, i.e., the nth root of the product of all state characteristic values, where n is the number of functional modules. The geometric mean is then compared with the arithmetic mean. When the state characteristic values ​​of each module tend to be balanced, the geometric mean is close to the arithmetic mean, and the system coupling degree approaches 1, indicating strong interaction and high coordination between modules. When the state characteristic values ​​of each module differ significantly, the geometric mean is significantly less than the arithmetic mean, and the system coupling degree is significantly lower than 1, indicating uneven development and weak interaction between modules. Taking four functional modules as an example, if the state characteristic value of the infrastructure module is 0.82, the data resource module is 0.78, the protocol standard module is 0.85, and the intelligent algorithm module is 0.80, then the arithmetic mean is 0.8125 and the geometric mean is 0.8123. The ratio of the two is normalized by the fourth root, and the system coupling degree is about 0.97, indicating that there is a high degree of coupling between the four modules.

[0045] Furthermore, the aforementioned coordination index reflects the weighted balance of the overall development level of each functional module. In the specific calculation, the second weight coefficient of each functional module is first obtained. This coefficient can be pre-set according to the importance of each module in process monitoring, such as 0.22 for the infrastructure module, 0.21 for the data resource module, 0.22 for the protocol standard module, and 0.21 for the intelligent algorithm module. Then, the state characteristic value of each functional module is multiplied by its corresponding second weight coefficient and summed to obtain the coordination index. Taking the aforementioned state characteristic value as an example, the coordination index is calculated as follows: The results show that the development levels of each module are relatively balanced. In practical applications, a sliding window approach can also be used for dynamic evaluation. For example, the window length can be set to 24 hours and the sliding step to 1 hour. The continuously collected operating data can be calculated in a rolling manner to track the evolution trend of the system coupling degree and coordination degree index in real time, thereby dynamically grasping the changing patterns of the collaborative state of the monitoring system.

[0046] When calculating the system coupling degree, the geometric mean and arithmetic mean of each state characteristic value are calculated. The system coupling degree is determined based on the ratio of the geometric mean to the arithmetic mean. When calculating the coordination index, the second weight coefficient of each functional module is obtained. The coordination index is obtained by weighting and summing the state characteristic values ​​based on the second weight coefficient. The ratio of the geometric mean to the arithmetic mean reflects the degree of dispersion among the state characteristic values ​​of each functional module. When the state characteristic values ​​of each module tend to be balanced, the ratio is close to 1. When the differences are large, the ratio is significantly less than 1, thus quantitatively characterizing the interaction strength between modules. The coordination index reflects the balance of the overall development level of each module through weighted summation. Both provide a clear quantitative basis for the subsequent calculation of coupling coordination degree from the two dimensions of balance and wholeness, respectively, so that the coupling coordination degree can comprehensively reflect the collaborative state between the modules of the monitoring system.

[0047] S1014: Calculate the coupling and coordination degree of the monitoring system based on the system coupling degree and the coordination degree index.

[0048] Preferably, in some embodiments of this application, the aforementioned coupling coordination degree is determined by the product of the system coupling degree and the coordination degree index. Specifically, the coupling coordination degree is obtained by multiplying the system coupling degree and the coordination degree index and taking the square root. This value comprehensively considers two dimensions: the strength of interaction between modules and the overall balance of development level. When both the system coupling degree and the coordination degree index are high, the coupling coordination degree approaches 1, indicating that the functional modules of the monitoring system are in a highly coordinated state. When either dimension is low, the coupling coordination degree decreases accordingly, reflecting the existence of coordination obstacles or development imbalances between modules. Taking the verification data from the welding workshop as an example, the system coupling degree is 0.97 and the coordination degree index is 0.81. After substituting into the calculation, the coupling coordination degree is approximately 0.89, which is higher than the preset threshold, indicating that the current monitoring system is in a highly coordinated state. In actual operation, the coupling coordination degree can be divided into different coordination levels according to numerical ranges, such as 0.0 to 0.2 corresponding to severe misalignment, 0.2 to 0.4 corresponding to mild misalignment, 0.4 to 0.6 corresponding to basic coordination, 0.6 to 0.8 corresponding to good coordination, and 0.8 to 1.0 corresponding to high coordination. This allows for the determination of the collaborative operation status level of the monitoring system and provides a basis for subsequent identification of weak functional modules and matching of optimization schemes. Furthermore, by setting a sliding window to continuously track the coupling coordination degree (e.g., a window length of 24 hours and a sliding step of 1 hour), its phased evolution from an initial value of 0.36, through an acceleration period of 0.41, a peak value of 0.72, to a convergence period of 0.89, can be observed, thereby dynamically evaluating the actual effect of optimization measures.

[0049] When calculating the state characteristic values ​​of each functional module, several corresponding state indicators are first determined based on the operational data of each functional module. The entropy weight method is then used to determine the first weight coefficient for each state indicator corresponding to each functional module. Finally, the state characteristic values ​​of each functional module are calculated by weighting each state indicator according to its first weight coefficient. Compared to simply averaging or assigning weights based on experience, the entropy weight method determines the weights based on the degree of variation of each state indicator among different samples. The greater the degree of variation of the indicator, the higher the weight, reflecting the more discriminative information the indicator carries in actual operation. This makes the calculated state characteristic values ​​more objectively reflect the actual operational level of each functional module, avoiding the problem of inaccurate identification of weak functional modules due to subjective weighting bias, and providing a reliable data foundation for accurate identification of weak modules in the future.

[0050] S102: When the coupling coordination degree is lower than a preset threshold, weak functional modules are identified from the functional modules based on the state characteristic values ​​of each functional module.

[0051] Preferably, in some embodiments of this application, the determination process when the coupling coordination degree is lower than a preset threshold is divided into two scenarios: static evaluation and dynamic evaluation. In the static evaluation scenario, the coupling coordination degree of the monitoring system is calculated periodically according to a preset evaluation cycle, such as performing a complete evaluation once per shift, daily, or weekly. The calculated coupling coordination degree is directly compared with the preset threshold. If it is lower than the threshold, the subsequent weak functional module identification process is triggered. In the dynamic evaluation scenario, a sliding window method is used to perform real-time rolling calculations on the running data. The window length and sliding step size are set, such as a window length of 24 hours and a sliding step size of 1 hour. Each time a window calculation is completed, a coupling coordination degree value is obtained. When the decrease in the coupling coordination degree obtained from two consecutive calculations exceeds a preset fluctuation range, such as 0.05, or when the coupling coordination degree of the current window is directly lower than the preset threshold, the triggering condition is determined to be met. In addition, dynamic evaluation can also be combined with the real-time requirements of process monitoring. Data of key processes can be collected according to a fixed sampling period, such as a sampling period of 200 milliseconds and a sliding window size of 50. When the number of consecutive triggers reaches the preset number, such as more than 3 times, and the current coupling coordination degree is lower than the preset threshold, it is also determined that the triggering condition is met, so as to promptly start the identification and optimization scheme matching of weak functional modules.

[0052] Furthermore, in some embodiments of this application, the step of identifying weak functional modules from the functional modules based on the state characteristic values ​​of each functional module includes: The state feature values ​​of each functional module are sorted to obtain a state feature value sequence; The target functional module with the lowest state feature value is selected from the sequence of state feature values; The target functional module is identified as the weak functional module.

[0053] In some embodiments of this application, the above sorting is performed by arranging the state feature values ​​of each functional module in descending or ascending order according to their numerical values, resulting in a sequence of state feature values. Specifically, when arranged in ascending order, the functional module with the lowest state feature value is located at the beginning of the sequence; when arranged in descending order, the functional module with the lowest state feature value is located at the end of the sequence. Then, the target functional module in an extreme position, i.e., the one with the lowest state feature value, is directly selected from this sequence. This target functional module is the link that has the greatest constraint on the overall coupling and coordination of the monitoring system, and is identified as a weak functional module. Taking welding workshop verification data as an example, the state feature value of the infrastructure module is 0.82, the data resource module is 0.78, the protocol standard module is 0.85, and the intelligent algorithm module is 0.80. After sorting, the state feature value sequence is 0.78, 0.80, 0.82, and 0.85. The data resource module located at the beginning of the sequence is selected as the target functional module and identified as the weak functional module that needs to be prioritized for optimization, thereby facilitating the subsequent location of affected process nodes and matching of targeted optimization schemes on the process association graph.

[0054] When identifying weak functional modules, the state feature values ​​of each functional module are sorted to obtain a sequence of state feature values. The target functional module with the lowest state feature value is then selected from the sequence and identified as the weak functional module. This method directly sorts and compares the calculated state feature values ​​of each functional module. The functional module with the lowest state feature value is the link with the lowest coupling coordination degree that has the greatest impact on the overall system coordination level. The identification process requires no additional judgment model or manual intervention, has low computational load, and responds quickly. It is suitable for real-time requirements in process monitoring scenarios, enabling the system to quickly locate modules that need priority improvement after detecting insufficient coupling coordination degree.

[0055] S103: Obtain the pre-constructed process association map of the target process, locate the target process node associated with the weak functional module on the process association map, and trace the influence path of the target process node along the process association map.

[0056] Furthermore, in some embodiments of this application, the process association graph includes: a number of process nodes, process dependencies between the process nodes, and support relationships between the process nodes and the functional modules.

[0057] Preferably, in some embodiments of this application, the construction of the process association graph is based on the process route file and the process configuration in the Manufacturing Execution System (MES). Information on each sub-process and workstation is extracted from the process route file as process nodes. For example, the body assembly welding process can be decomposed into process nodes such as spot welding, arc welding, adhesive application, and inspection. Input-output relationships between process nodes are extracted from the MES process constraint file to establish process dependencies, including the flow of workpieces or data from preceding processes to subsequent processes. Simultaneously, information on the functional modules that each process node depends on is extracted from the monitoring system configuration ledger to establish support relationships. For example, the spot welding process node is supported by a specific sensor group from the infrastructure module, the corresponding data source from the data resource module, the interface channel from the protocol standard module, and the detection model from the intelligent algorithm module. Based on this, a mapping is established between process nodes and functional modules. Each process node records the process number, cycle time, pass rate, data source, and sampling frequency as node attributes. Each process dependency records the relationship type, transmission strength, and delay time as edge attributes, thereby completing the construction of the process association graph.

[0058] Furthermore, in some embodiments of this application, the step of locating the target process node associated with the weak functional module on the process association graph and tracing the influence path of the target process node along the process association graph includes the following steps S1031 to S1034: S1031: Based on the support relationship, query the target process node supported by the weak functional module from the process nodes.

[0059] Preferably, in some embodiments of this application, the above-described positioning process first filters based on support relationships. Specifically, all process nodes in the process association graph are traversed, the support relationship attributes of each process node are queried, and nodes whose support relationships include the weak functional module are selected as target process nodes. Taking the body assembly welding process in the welding workshop as an example, if the data resource module is identified as a weak functional module, then all process nodes whose support relationships include the data resource module are queried, such as spot welding process nodes, arc welding process nodes, etc., and these nodes are determined as target process nodes.

[0060] S1032: Following the process dependency relationship, trace downstream process nodes from the target process node towards the output direction to obtain the downstream influence range of the target process node.

[0061] Preferably, in some embodiments of this application, after determining the target process node, a breadth-first traversal is performed along the process dependency relationship towards its output direction. The subsequent process nodes directly output by the target process node, as well as the indirect process nodes passed downstream by these subsequent process nodes, are searched layer by layer. Simultaneously, the relationship type and transmission strength during each layer's transmission process are recorded until a terminal process node with no subsequent output is reached. This yields the downstream influence range of the target process node. Taking the aforementioned spot welding process node as an example, tracing along the process dependency relationship towards the output direction, its downstream process nodes include arc welding process nodes, adhesive application process nodes, and inspection process nodes. These nodes constitute the downstream influence range of the spot welding process node.

[0062] S1033: Following the process dependency relationship, trace upstream process nodes from the target process node towards the input direction to obtain the upstream cause range of the target process node.

[0063] Preferably, in some embodiments of this application, the upstream cause range of the target process node is then traced backward along the process dependency relationship towards the input direction. The preceding process nodes that provide input to the target process node are searched layer by layer, as well as the indirect process nodes traced upward from the preceding process nodes. Simultaneously, the delay time and process constraints in each layer of input relationship are recorded until a source process node without external input is reached. This yields the upstream cause range of the target process node. Taking the spot welding process node as an example, its upstream process nodes include the body stamping part supply node and the fixture positioning node. These nodes constitute the upstream cause range of the spot welding process node.

[0064] S1034: Determine the impact path of the target process node based on the downstream impact range and the upstream cause range.

[0065] Preferably, in some embodiments of this application, the downstream impact range and the upstream cause range are finally merged and integrated according to the connection order and transmission direction in the process dependency relationship to form a complete path from the upstream cause range through the target process node to the downstream impact range, i.e., the impact path of the target process node. This path clarifies the specific position of the weak functional module in the process chain, the root cause of the abnormality spreading upstream, and the impact boundary transmitted downstream, providing a direct basis for determining the scope of action when matching and optimizing the scheme.

[0066] When tracing the impact path, the process first queries the process nodes supported by the weak functional module based on the support relationship. Then, it traces downstream and upstream process nodes from the target process node in the output and input directions along the process dependencies to obtain the downstream impact range and the upstream cause range, thereby determining the impact path. This method explicitly models the support relationship between functional modules and process nodes using a graph structure. Once a weak functional module is identified, the specific process link it supports can be quickly located. Then, based on the dependencies between processes, tracing upstream and downstream clarifies the impact range of the weak module on subsequent processes and the upstream process causes leading to the weakness. This fully presents the actual impact boundary of the weak functional module on the process chain, providing a clear basis for matching subsequent optimization schemes.

[0067] S104: Based on the target process node, the influence path, and the weak functional module, match a system optimization scheme, and optimize the monitoring system according to the system optimization scheme.

[0068] Furthermore, in some embodiments of this application, the step of matching a system optimization scheme based on the target process node, the influence path, and the weak functional module, and optimizing the monitoring system according to the system optimization scheme, includes the following steps S1041 to S1044: S1041: Based on the weak functional module, query the candidate optimization scheme corresponding to the weak functional module from the preset optimization scheme library.

[0069] Preferably, in some embodiments of this application, a pre-defined optimization scheme library is stored in partitions according to the type of weak functional modules. For scenarios with weak infrastructure modules, the pre-defined candidate optimization schemes in the library include increasing the number of deployed sensors, adjusting the sensor sampling frequency, upgrading the edge node hardware configuration, or optimizing the edge network topology. For scenarios with weak data resource modules, candidate optimization schemes include expanding the number of data source accesses, adjusting multi-source data synchronization strategies, optimizing data caching mechanisms, or increasing transmission link bandwidth. For scenarios with weak protocol standard modules, candidate optimization schemes include unifying interface protocol versions, deploying protocol conversion gateways, optimizing interface call logic, or updating protocol adapters. For scenarios with weak intelligent algorithm modules, candidate optimization schemes include updating the inference model version, adjusting the model inference threshold, expanding the training sample set, or optimizing the model's lightweight parameters. During a query, all candidate optimization schemes are directly extracted from the corresponding partition based on the identified type of weak functional module, avoiding the problem of inapplicable schemes caused by cross-module matching.

[0070] S1042: Determine the scope of the candidate optimization scheme based on the target process node and the influence path.

[0071] S1043: Determine the target optimization scheme from the candidate optimization schemes according to the scope of application.

[0072] Preferably, in some embodiments of this application, after determining the candidate optimization scheme, the actual scope of the optimization scheme to be deployed is determined based on the node number of the target process node in the process association graph and the set of nodes covered by the influence path. Specifically, the location attribute of the target process node is read. If the influence path only covers the node and one or two directly adjacent downstream nodes, the scope of effect is marked as a local scope; if the influence path extends to multiple process branches or cross-regional nodes, the scope of effect is marked as an extended scope. Based on the size of the scope of effect, the target optimization scheme is determined from the candidate optimization schemes: when the scope of effect is a local scope, the scheme that only requires modification of a single point configuration to take effect is preferred; when the scope of effect is an extended scope, the scheme that supports multi-point synchronous deployment is preferred.

[0073] S1044: Distribute the target optimization scheme to the edge computing nodes or cloud servers within the scope of action to adjust the operating parameters of the monitoring system.

[0074] Preferably, in some embodiments of this application, the target optimization scheme is distributed to edge computing nodes or cloud servers within the scope of effect. The specific distribution path is determined according to the hierarchy of the scope of effect: when the scope of effect corresponds to a single workstation or a local monitoring node, the optimization scheme is directly pushed to the edge computing node deployed at that workstation through the edge network. The edge node parses the configuration parameters in the scheme locally and writes them into the running configuration file, adjusting the sensor sampling period, switching the lightweight inference model, or updating the local data caching strategy; when the scope of effect corresponds to a geographical cluster composed of multiple workstations, the optimization scheme is uploaded to a regional-level cloud server, and the regional server batch synchronizes and updates the configuration to each edge node in the cluster; when the scope of effect covers the entire enterprise-level area, the optimization scheme is distributed to a central-level cloud server, and the central server uniformly schedules each regional node to perform parameter adjustments. After the configuration takes effect, each receiving node continues to collect and process process data according to the new running parameters, thereby completing the adjustment of the monitoring system's running parameters.

[0075] When matching optimization solutions, the system first queries the pre-set optimization solution library for corresponding candidate optimization solutions based on the weak functional modules. Then, it determines the scope of application of the candidate optimization solutions based on the target process nodes and the impact path. Based on this scope, it selects the target optimization solution from the candidate solutions. Finally, it distributes the target optimization solution to the edge computing nodes or cloud servers within the scope of application to adjust the operating parameters of the monitoring system. This method narrows the matching range of solutions by identifying the type of weak functional modules and then determines the process boundaries that the optimization measures need to cover by combining the impact path. This ensures that the final optimization solution targets both the type of weak module and has a clearly defined execution scope, avoiding problems such as insufficient effectiveness or over-adjustment caused by a mismatch between the optimization solution and the scope of impact.

[0076] In summary, the collaborative optimization method for a process monitoring system provided in this application has the following advantages compared to the prior art: By acquiring the operating data of multiple functional modules in the monitoring system of the target process, the state characteristic values ​​of each functional module are first calculated, and the overall coupling coordination degree of the monitoring system is calculated accordingly. When the coupling coordination degree is lower than a preset threshold, instead of directly adjusting the system as a whole or checking each module individually, the weak functional module with the lowest state characteristic value is identified from multiple functional modules based on the state characteristic values ​​of each functional module. Then, a pre-constructed process association map is acquired, the target process node associated with the weak functional module is located on the map, and the influence path of the target process node is traced, thereby clarifying which process links and the scope of influence the weak functional module specifically affects. Finally, a system optimization scheme is matched based on the weak functional module, the target process node, and the influence path. Compared to existing technologies that rely on independent adjustments to each module, leading to repeated fluctuations or simply outputting evaluation scores without pinpointing weaknesses and relying on manual investigation, this application determines whether the overall system needs optimization by assessing the degree of coupling coordination. When optimization is required, it further utilizes the ranking results of the state characteristic values ​​of each module to accurately locate the weak module. Then, it combines the process correlation diagram to determine the influence boundary of the weak module, enabling the optimization scheme to target the real system weaknesses without exceeding the necessary range. This improves the optimization accuracy of the process monitoring system and avoids the repeated effects caused by blind adjustments.

[0077] Based on the above-described method embodiments, one embodiment of this application provides a collaborative optimization system for a process monitoring system, which is used to implement the collaborative optimization method for a process monitoring system in any embodiment of this application.

[0078] It should be noted that the system embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0079] Based on the above embodiments of the collaborative optimization method for the process monitoring system, another embodiment of this application provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the collaborative optimization method for the process monitoring system of any embodiment of this application.

[0080] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete this application. The one or more module units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0081] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0082] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0083] Based on the above-described method embodiments, another embodiment of this application provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the collaborative optimization method of the process monitoring system described in any of the above-described method embodiments of this application.

[0084] The modules / units integrated in the aforementioned device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, an electrical signal, and a software distribution medium, etc.

Claims

1. A collaborative optimization method for a process monitoring system, characterized in that, include: The system acquires the operating data of multiple functional modules in the monitoring system of the target process, calculates the state characteristic value of each functional module based on the operating data, and calculates the coupling coordination degree of the monitoring system based on each state characteristic value. When the coupling coordination degree is lower than a preset threshold, weak functional modules are identified from the functional modules based on the state characteristic values ​​of each functional module; Obtain a pre-constructed process association graph of the target process, locate the target process node associated with the weak functional module on the process association graph, and trace the influence path of the target process node along the process association graph; Based on the target process node, the impact path, and the weak functional module, a system optimization scheme is matched, and the monitoring system is optimized according to the system optimization scheme.

2. The collaborative optimization method for a process monitoring system as described in claim 1, characterized in that, The step of calculating the state characteristic values ​​of each functional module based on the operational data, and calculating the coupling coordination degree of the monitoring system based on each state characteristic value, includes: Based on the operational data, several status indicators corresponding to each functional module are determined, and the first weight coefficient of each status indicator corresponding to each functional module is determined by the entropy weight method. Based on the first weighting coefficient of each state indicator, the state indicators of each functional module are weighted and calculated to obtain the state feature value of each functional module. Based on the state characteristic values, calculate the system coupling degree, which characterizes the interaction strength between the functional modules, and calculate the coordination index, which characterizes the balance of the development level of the functional modules. The coupling and coordination degree of the monitoring system is calculated based on the system coupling degree and the coordination degree index.

3. The collaborative optimization method for a process monitoring system as described in claim 2, characterized in that, The calculation of the system coupling degree, which characterizes the interaction strength between the functional modules, and the coordination index, which characterizes the balance of development levels of the functional modules, based on the state characteristic values, includes: Calculate the geometric mean and arithmetic mean of each state characteristic value, and determine the system coupling degree based on the ratio of the geometric mean and the arithmetic mean; Obtain the second weight coefficient of each functional module, and weight each state feature value according to the second weight coefficient to obtain the coordination index.

4. The collaborative optimization method for a process monitoring system as described in claim 2, characterized in that, The functional modules include: an infrastructure module, a data resource module, a protocol standard module, and an intelligent algorithm module; the step of determining several status indicators corresponding to each functional module based on the operational data includes: Acquire sensor deployment data and edge node operation data of the infrastructure module, determine sensor coverage status indicators based on the sensor deployment data, and determine edge computing capability status indicators based on the edge node operation data; Acquire multi-source data collection data and data transmission data from the data resource module, determine the data fusion rate status index based on the multi-source data collection data, and determine the latency status index based on the data transmission data; Obtain the interface call data and protocol conversion data of the protocol standard module, determine the interface standardization status index based on the interface call data, and determine the protocol conversion efficiency status index based on the protocol conversion data. The model inference data and model update data of the intelligent algorithm module are obtained. The detection accuracy status index is determined based on the model inference data, and the model update frequency status index is determined based on the model update data.

5. The collaborative optimization method for a process monitoring system as described in claim 1, characterized in that, The process association graph includes: several process nodes, process dependencies between the process nodes, and support relationships between the process nodes and the functional modules; locating the target process node associated with the weak functional module on the process association graph, and tracing the influence path of the target process node along the process association graph, includes: Based on the support relationship, query the target process node supported by the weak functional module from the process nodes; Following the process dependency relationship, trace downstream process nodes from the target process node towards the output direction to obtain the downstream influence range of the target process node; Following the process dependency relationship, trace upstream process nodes from the target process node towards the input direction to obtain the upstream cause range of the target process node; The impact path of the target process node is determined based on the downstream impact range and the upstream cause range.

6. The collaborative optimization method for a process monitoring system as described in claim 1, characterized in that, The step of matching a system optimization scheme based on the target process node, the impact path, and the weak functional module, and optimizing the monitoring system according to the system optimization scheme, includes: Based on the weak functional module, query the candidate optimization scheme corresponding to the weak functional module from the preset optimization scheme library; Based on the target process node and the influence path, determine the scope of application of the candidate optimization scheme; Based on the scope of application, a target optimization scheme is determined from the candidate optimization schemes; The target optimization scheme is distributed to edge computing nodes or cloud servers within the scope of the system to adjust the operating parameters of the monitoring system.

7. The collaborative optimization method for a process monitoring system as described in claim 1, characterized in that, The step of identifying weak functional modules from the functional modules based on the state feature values ​​of each functional module includes: The state feature values ​​of each functional module are sorted to obtain a state feature value sequence; The target functional module with the lowest state feature value is selected from the sequence of state feature values; The target functional module is identified as the weak functional module.

8. A collaborative optimization system for process monitoring, characterized in that, The process monitoring system collaborative optimization system is used to implement the process monitoring system collaborative optimization method as described in any one of claims 1 to 7.

9. A terminal device, characterized in that, The system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements a collaborative optimization method for a process monitoring system as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform a collaborative optimization method for a process monitoring system as described in any one of claims 1 to 7.

Images