Distributed factory intelligent inspection path dynamic planning and fault linkage processing system

Abstract

The application relates to the technical field of factory intelligent inspection, and discloses a distributed factory intelligent inspection path dynamic planning and fault linkage processing system. The system comprises four core modules of partition monitoring, path planning, fault analysis and linkage execution. The partition monitoring module obtains current operation data of each production area, and provides basic data for subsequent processes; the path planning module generates an initial inspection path according to the data, breaks the limitation of traditional fixed paths, and optimizes the allocation of inspection resources; the fault analysis module calculates path deviation values according to real-time operation state data of path nodes, captures node state changes to identify fault risks; and the linkage execution module triggers corresponding area linkage processing instructions according to the deviation values, reduces information transmission links, and realizes rapid fault response. The system forms a complete closed loop of "data monitoring-path planning-deviation analysis-linkage processing", and guarantees the stability and safety of distributed factory production.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of factory intelligent inspection, in particular to a distributed factory intelligent inspection path dynamic planning and fault linkage processing system. BACKGROUND

[0002] In the operation management of a distributed factory, a production area usually has characteristics such as a wide land occupation, a scattered equipment distribution, and a closely related production process. As an important link to ensure the stable operation of production, the efficiency and accuracy of inspection directly affect the overall production rhythm and safety level of the factory. Traditional distributed factory inspection relies on manual formulation of fixed inspection paths. Such path planning methods are often based on experience or preset processes and do not fully consider the real-time running state of each production area. For example, when a production area has abnormal equipment load or parameter fluctuation, the fixed inspection path cannot be adjusted in time, resulting in repeated inspection of normal areas by the inspection personnel according to the original plan, while the abnormal area cannot be given priority attention, which easily delays the opportunity for fault discovery and processing.

[0003] In the traditional inspection process, the monitoring data of each production area is in a relatively independent state, lacking an effective data integration and sharing mechanism. After obtaining the running data of each area, the inspection personnel needs to manually organize and analyze the data, which not only consumes time and effort, but also easily leads to data deviation due to human operation errors, thereby affecting the accurate judgment of the running state of the production area. In addition, when a fault is found during the inspection process, the processing flow needs to be started through manual reporting and multi-level approval in the traditional mode, and the information transmission between the relevant departments is lagging, making it difficult to achieve rapid response and linkage processing of faults.

[0004] With the large-scale development and the improvement of the intelligent level of the distributed factory, the complexity and quantity of production equipment are increasing, and the limitations of the traditional inspection method are becoming more and more prominent. The fixed inspection path is difficult to adapt to the dynamic changes of the running demand of each production area, resulting in unreasonable allocation of inspection resources, over-inspection of some areas, and insufficient inspection frequency of some key areas; the lag of fault processing may cause a chain reaction, affecting the production continuity of the entire factory, and causing unnecessary economic losses. Therefore, the current distributed factory inspection field needs a solution that can dynamically plan the inspection path combined with real-time running data and realize rapid linkage processing of faults to improve the inspection efficiency and fault processing capability and ensure the stability and safety of factory production. SUMMARY

[0005] The purpose of the present application is to provide a distributed factory intelligent inspection path dynamic planning and fault linkage processing system to solve the problems raised in the background.

[0006] To achieve the above object, the application provides a distributed factory intelligent inspection path dynamic planning and fault linkage processing system, which comprises:

[0007] a partition monitoring module, a path planning module, a fault analysis module and a linkage execution module;

[0008] The partition monitoring module acquires current production area operation data of each production area of the distributed factory;

[0009] The path planning module generates an initial inspection path based on the current production area operation data;

[0010] The fault analysis module calculates a path deviation value according to real-time node operation state data of each path node on the initial inspection path;

[0011] The linkage execution module triggers linkage processing instructions of the corresponding production area according to the path deviation value.

[0012] Preferably, the partition monitoring module specifically performs the following operations: dividing the production areas of the distributed factory into high-risk production areas, medium-risk production areas and low-risk production areas;

[0013] collecting real-time high-risk equipment vibration values, high-risk environment temperature values and high-risk energy consumption values of the high-risk production areas;

[0014] collecting real-time medium-risk equipment current values and medium-risk material flow values of the medium-risk production areas;

[0015] collecting real-time low-risk environment humidity values and low-risk equipment start-stop frequencies of the low-risk production areas.

[0016] Preferably, the path planning module generates an initial inspection path and a path node sequence covering the high-risk production areas, the medium-risk production areas and the low-risk production areas through a regional risk weight distribution algorithm based on the real-time high-risk equipment vibration values, the high-risk environment temperature values and the high-risk energy consumption values of the high-risk production areas, in combination with the real-time medium-risk equipment current values and the medium-risk material flow values of the medium-risk production areas, and in association with the real-time low-risk environment humidity values and the low-risk equipment start-stop frequencies of the low-risk production areas.

[0017] Preferably, the fault analysis module acquires real-time node operation state data of each path node in the path node sequence during execution of the initial inspection path;

[0018] The path deviation value is calculated based on real-time node operation state data and preset node operation standard value, wherein the path deviation value of the path node in the high-risk production area is associated with the real-time high-risk equipment vibration value deviation, the high-risk environment temperature value deviation and the high-risk energy consumption value deviation, the path deviation value of the path node in the medium-risk production area is associated with the real-time medium-risk equipment current value fluctuation coefficient and the medium-risk material flow value fluctuation coefficient, and the path deviation value of the path node in the low-risk production area is associated with the real-time low-risk environment humidity value change rate and the low-risk equipment start-stop frequency change rate.

[0019] Preferably, the path planning module performs dynamic adjustment of the path according to the path deviation value: when the path deviation value exceeds a preset deviation threshold, the path node sequence priority of the high-risk production area, the medium-risk production area and the low-risk production area is re-allocated.

[0020] When the path deviation value of the path node in the high-risk production area continuously increases, the inspection interval length of the path node in the high-risk production area is shortened.

[0021] When the path deviation value of the path node in the medium-risk production area exceeds a preset fluctuation range, the redundant inspection frequency of the path node in the medium-risk production area is increased.

[0022] Preferably, the fault analysis module generates a fault level evaluation result based on the path deviation value: if the real-time high-risk equipment vibration value deviation, the high-risk environment temperature value deviation and the high-risk energy consumption value deviation associated with the path deviation value all exceed a high-risk threshold, the path node in the high-risk production area is marked as a first-level fault node.

[0023] If the real-time medium-risk equipment current value fluctuation coefficient and the medium-risk material flow value fluctuation coefficient associated with the path deviation value exceed a medium-risk threshold, the path node in the medium-risk production area is marked as a second-level fault node.

[0024] If the real-time low-risk environment humidity value change rate and the low-risk equipment start-stop frequency change rate associated with the path deviation value exceed a low-risk threshold, the path node in the low-risk production area is marked as a third-level fault node.

[0025] Preferably, the linkage execution module generates a linkage processing instruction according to the first-level fault node, the second-level fault node and the third-level fault node: a high-risk production area equipment speed reduction instruction and an environment cooling instruction are triggered for the first-level fault node.

[0026] A medium-risk production area current stabilizing instruction and a material flow calibration instruction are triggered for the second-level fault node.

[0027] A low-risk production area humidity adjusting instruction and an equipment start-stop frequency limiting instruction are triggered for the third-level fault node.

[0028] Preferably, the partition monitoring module collects high-risk production area update high-risk equipment vibration value, update high-risk environment temperature value and update high-risk energy consumption value after the linkage processing instruction is executed, the medium-risk production area updates the medium-risk equipment current value and the medium-risk material flow value, the low-risk production area updates the low-risk environment humidity value and the low-risk equipment start-stop frequency;

[0029] The fault analysis module calculates the linkage path deviation value based on the updated high-risk equipment vibration value, the updated high-risk environment temperature value, the updated high-risk energy consumption value, the updated medium-risk equipment current value, the updated medium-risk material flow value, the updated low-risk environment humidity value and the updated low-risk equipment start-stop frequency.

[0030] Preferably, the path planning module executes path re-planning according to the linkage path deviation value: if the high-risk production area linkage path deviation value does not decrease to a preset safety threshold, a high-risk production area path node bypass path is generated;

[0031] If the medium-risk production area linkage path deviation value continuously exceeds a preset fluctuation range, a medium-risk production area path node double-track inspection path is generated;

[0032] If the low-risk production area linkage path deviation value exceeds a low-risk threshold, a low-risk production area path node fast inspection path is generated.

[0033] Preferably, the linkage execution module integrates the high-risk production area path node bypass path, the medium-risk production area path node double-track inspection path and the low-risk production area path node fast inspection path into a final inspection path, and updates the path node sequence based on the final inspection path;

[0034] The partition monitoring module starts a new round of current production area running data collection based on the updated path node sequence.

[0035] Compared with the prior art, the present application has the following advantages:

[0036] 1. Through the cooperative action of the partition monitoring module, the path planning module, the fault analysis module and the linkage execution module, a more efficient and flexible solution for distributed factory inspection work is provided. The partition monitoring module can comprehensively obtain the current production area running data of each production area, breaking the situation of isolated data in each area in traditional inspection, so that the system can master the key information such as equipment running parameters and environmental conditions of each production area in real time, providing a comprehensive and accurate data basis for subsequent path planning and fault analysis, and avoiding the deviation of inspection decision caused by data missing or lag.

[0037] 2. The path planning module generates an initial inspection path based on the current production area operation data obtained by the partition monitoring module. Unlike the traditional fixed path mode, the initial inspection path can fully reflect the real-time operation state of each production area. For example, when the operation data of a certain production area shows that the equipment is in a high load state, the path planning module can prioritize the inspection task of this area in the initial path, ensuring that the key area can be paid attention to in time, achieving reasonable allocation of inspection resources, avoiding waste of inspection resources in areas with stable operation state, and also preventing potential faults in key areas due to delayed inspection.

[0038] 3. The fault analysis module calculates the path deviation value based on the real-time node operation state data of each path node in the initial inspection path, which can capture the dynamic changes of the operation state of each node in the inspection process in real time. During the inspection process, the operation state of each path node may change. Through the analysis of these real-time data, the fault analysis module accurately calculates the deviation between the actual operation state and the expected state of the initial path planning, thereby discovering potential fault risks or unreasonable path planning in time. This real-time analysis mechanism avoids the problem that problems can only be found after completing the entire inspection process in traditional inspection, allowing the system to dynamically adjust the strategy during the inspection process and identify potential problems in advance.

[0039] 4. The linkage execution module triggers the corresponding production area linkage processing instruction according to the path deviation value, achieving rapid response and multi-department cooperation of fault handling. When the fault analysis module calculates a large path deviation value, indicating that a fault may exist in a certain production area, the linkage execution module can directly trigger the corresponding linkage processing instruction according to the preset rules without going through the cumbersome manual reporting and approval process, and synchronize the fault information to the maintenance department, production scheduling department and other related links responsible for the area. Each department can obtain fault information in the first time and start the processing process, reducing the intermediate links of information transmission, shortening the fault handling time, and avoiding the expansion of faults to cause greater impact on factory production.

[0040] 5.The whole system realizes the dynamic linkage of inspection path planning and fault handling, and forms a closed-loop process of "data monitoring-path planning-deviation analysis-linkage processing". During the inspection process, if the fault analysis module finds that the path deviation value exceeds the preset range, the linkage execution module can trigger the fault handling instruction and also feedback the deviation information to the path planning module. The path planning module can adjust the subsequent inspection path in time according to the feedback, so as to avoid the inspection personnel to continue to go to the area where the fault has occurred and is being handled, and further optimize the inspection efficiency. This dynamic adjustment mechanism enables the inspection work to continuously adapt to the operation changes of the production area, ensures that the inspection work always has high pertinence and effectiveness, and also makes the fault handling no longer limited to a single area or a single department, forms efficient cooperation between modules and departments, and comprehensively improves the overall operation and management level of the distributed factory, and ensures the continuity and safety of the production process. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 a timing diagram of the distributed factory intelligent inspection path dynamic planning and fault linkage processing system described in the present application;

[0042] Figure 2 a flowchart for the partition monitoring module to execute;

[0043] Figure 3 a flowchart for the fault analysis module to calculate the path deviation value;

[0044] Figure 4 a flowchart for data acquisition and deviation calculation after linkage;

[0045] Figure 5 a flowchart for the linkage execution module to integrate the final path. DETAILED DESCRIPTION

[0046] The technical solutions of the present application will be described in detail below with reference to the drawings of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0047] Please refer to Figure 1 , the present application provides a distributed factory intelligent inspection path dynamic planning and fault linkage processing system, which comprises:

[0048] The system realizes intelligent inspection path dynamic planning and fault linkage processing of distributed factory through integrated module design. The partition monitoring module is responsible for real-time data acquisition of each production area of the factory, obtaining current production area operation data including device operation parameters and environmental indicators. The path planning module generates an initial inspection path based on the collected data using a risk assessment algorithm, which covers the key areas that need to be monitored in the factory. The fault analysis module continuously acquires real-time node operation state data of the path nodes during the inspection process, and obtains path deviation values by calculating the deviation from the preset standard value, which are used to quantify the abnormal conditions of each node. The linkage execution module triggers corresponding linkage processing instructions according to the size and type of the path deviation value, realizing automatic regulation and control and fault response of the production area. The modules work together to form a closed loop from data acquisition, path generation, state analysis to instruction execution, effectively improving the intelligent level and fault handling efficiency of factory inspection.

[0049] Embodiment 1: see Figure 2 The implementation of the partition monitoring module involves systematic division of the production areas of the distributed factory, which is based on the characteristics of the production process, the criticality of the equipment, and the risk patterns reflected in the historical operation data. The identification of high-risk production areas is mainly based on the severity of potential fault consequences. For example, in a chemical enterprise, synthesis workshops involving high-temperature and high-pressure reactions, tank areas storing toxic or flammable raw materials, and workshops with high-speed compressors are usually classified as high-risk. The monitoring of these areas focuses on parameters that can directly reflect the mechanical state of the equipment and the safety of the environment. Piezoelectric acceleration sensors are used to collect vibration data, and their installation positions are determined through finite element analysis. They are usually installed on bearing seats, machine flanges, or equipment foundations, which can transmit core vibration signals. The sampling rate of the sensor is set to cover more than ten times the characteristic frequency of the equipment to capture possible high-frequency fault components. After anti-aliasing filtering and digitization, the original vibration signal is transmitted to the data acquisition station through the industrial field bus. The environmental temperature is monitored using distributed K-type or T-type thermocouples. The measurement points are arranged in key areas for equipment heat dissipation and personnel operation areas. Data is collected every second and subjected to cold end compensation processing. Energy consumption monitoring is achieved through smart meters and power transducers, which simultaneously measure the current, voltage, and power factor of the main drive motor and heating unit to calculate real-time energy consumption indicators.

[0050] The medium-risk production area covers those links with relatively limited failure impact but still possible to cause production interruption or quality problems, such as material conveying system, mixing stirring section or semi-finished product buffer area. The monitoring core of these areas lies in process stability and equipment load state. The collection of equipment current value utilizes Hall effect current sensor or Rogowski coil, and the non-contact measurement ensures safety and avoids interference to the original circuit. The sensor output signal is converted into a standard current signal after amplification and linearization correction. The material flow monitoring selects different flow meters according to the material properties. For clean liquid, electromagnetic flow meter is used, and its measurement is not affected by temperature and pressure changes. For slurry or powder, Coriolis mass flow meter is used to directly obtain mass flow data. The flow meter installation position requires sufficient straight pipe section before and after to ensure stable flow field, and the data output is processed by temperature and pressure compensation algorithm to improve accuracy.

[0051] The low-risk production area includes general assembly area, packaging line, raw material warehouse, etc., and its running state has less direct impact on the overall production safety, but long-term abnormalities may indirectly lead to efficiency decline or equipment wear. The environmental humidity monitoring uses capacitive or resistive humidity sensors, and the sensor has a self-cleaning function to prevent dust accumulation from affecting the reading. The data collection interval can be adjusted according to seasonal changes or production process requirements. The statistics of equipment start-stop frequency is realized by monitoring the relay state in the control loop or the digital output signal of PLC, and records the number of starts and stops of the equipment within a unit time. This data reflects the equipment operation intensity and potential wear trend.

[0052] All the raw data collected from various sensors are fed into the regional data concentrator for preliminary data cleaning and verification, including removing abnormal values obviously beyond the physical range, handling transient loss problems in signal transmission, and aligning the time stamp. The pre-processed data stream is uploaded to the central monitoring platform through the factory backbone network and enters the real-time database of the partition monitoring module. The database is stored in layers according to area level and data type, supports high-speed query and historical backtracking, and provides a complete and consistent factory running state information basis for subsequent path planning and risk analysis modules.

[0053] Example 2: see Figure 3, the path planning module receives standardized data streams from the zoning monitoring module, which are organized according to the classification structure of high-risk, medium-risk, and low-risk production zones. Internally, the module maintains a dynamically updated risk weight configuration table that assigns corresponding weight coefficients to each monitoring parameter under each zone type. These coefficients are not fixed but are adjusted through machine learning algorithms analyzing historical operational data, combined with expert experience, to reflect the relative importance of different parameters on the overall risk. For example, in high-risk zones, the weight of equipment vibration values is usually the highest because it is directly related to the integrity of mechanical structures; environmental temperature comes second, related to production safety; and energy consumption values more reflect operational economy. The weight configuration table also considers the coupling relationship between parameters, such as in some cases where temperature rise may lead to increased vibration, which is reflected in the weight allocation.

[0054] Based on real-time data and corresponding weights, the module calculates the real-time risk score of each production zone. This score is a comprehensive indicator that quantifies the current degree of abnormality in the region. The calculation process not only considers the absolute value of the parameter, but also focuses on its deviation from the historical baseline or preset standard value. For each region i, its risk score R i is calculated as follows:

[0055] ;

[0056] In this formula: n represents the total number of parameters monitored in the region. w j is the dynamic weight coefficient assigned to the jth parameter, with a value range of 0 to 1, and the sum of the weights of all parameters is 1. v j represents the current real-time measurement value of the jth parameter. r j is the reference value or baseline value of the parameter under normal operating conditions, usually determined by the statistical average of historical data. s j is a scaling factor used to standardize the deviation of the parameter, which is usually half of the allowed fluctuation range of the parameter or calculated based on its historical standard deviation. The entire fraction essentially calculates the standardized deviation of the current value relative to the reference value. Through weighted summation, the final R i is a dimensionless value whose size directly reflects the comprehensive risk level of region i.

[0057] After calculating the real-time risk scores of all regions, the path planning module uses these scores to sort the regions. High-risk regions, i.e. R iRegions with higher values are assigned higher inspection priority. The core task of generating an initial inspection path is transformed into a path optimization problem with priority constraints. The goal of this problem is to find a path that enables an inspection unit (such as a robot or personnel) to visit all points that need inspection with the highest efficiency, while ensuring that high-priority points are visited earlier and more frequently. This is similar to a variation of the classic Traveling Salesman Problem (TSP) or Vehicle Routing Problem (VRP), but with the added constraint of dynamic priority. The module employs heuristic algorithms, such as genetic algorithms or ant colony algorithms, to solve this optimization problem. The algorithm takes as input the coordinates of all region nodes that need to be visited, the movement cost (such as distance or time) between regions, and the real-time risk score R i of each node. The optimization objective function is usually set to minimize the total path length or total inspection time, while ensuring that high-priority nodes are positioned early in the path sequence through a penalty function mechanism. After the solution is complete, an ordered path node sequence is output, which explicitly defines the order in which the inspection unit visits each production region.

[0058] The fault analysis module starts its monitoring function immediately after the initial inspection path begins execution, continuously and nearly in real-time acquiring the latest operating state data of each node (i.e., production region) in the path node sequence through the factory's sensor network and Supervisory Control and Data Acquisition System (SCADA). The types of data acquired are consistent with those collected by the partition monitoring module, but the frequency is higher, and it is a focused data stream for specific path nodes. For each high-risk production region node in the sequence, the module calculates three key deviation indicators: real-time high-risk equipment vibration value deviation, real-time high-risk environment temperature value deviation, and real-time high-risk energy consumption value deviation. Each deviation is calculated by taking the absolute difference between the current sampling value and the standard operating value of the parameter, and then dividing by the standard value, presented in percentage form, thereby quantifying the degree of abnormality.

[0059] For medium-risk production region path nodes, the focus of analysis is on parameter stability, and the module calculates the real-time medium-risk equipment current value fluctuation coefficient. This coefficient is obtained by calculating the standard deviation of current sampling values within a recent time window (e.g., the past 5 minutes), and then dividing by the average value within that time period. Similarly, the real-time medium-risk material flow value fluctuation coefficient is calculated using the same statistical method. The fluctuation coefficient is a dimensionless quantity that effectively characterizes the degree of volatility of a parameter around its mean value. The larger the coefficient value, the more unstable the operation.

[0060] For low-risk production area path nodes, the module focuses on the trend of the parameter, it calculates the change rate of the real-time low-risk environment humidity value, that is, the difference between the current humidity value and the humidity value in the last sampling period, divided by the sampling interval time. Similarly, it calculates the change rate of the real-time low-risk equipment start-stop frequency, that is, the difference between the current start-stop frequency per unit time and the frequency in the last statistical period, divided by the period length. The change rate reflects the speed and direction of the dynamic change of the parameter.

[0061] All these calculated indicators, deviation degree, fluctuation coefficient and change rate, are normalized to the interval [0, 1] respectively for comprehensive comparison. Subsequently, they are input into a weighted fusion model similar to the calculation of risk score in the path planning module, but the weight configuration here may focus more on the instantaneous abnormal characteristics of the parameter. Through weighted fusion, a comprehensive path deviation value is finally calculated for each path node. This path deviation value is an aggregate indicator that comprehensively reflects the overall deviation of the node from its normal standard operating state at the current time, providing a quantitative decision basis for subsequent path dynamic adjustment and fault level assessment.

[0062] Example 3: The path planning module is internally set with a dynamically updated preset deviation threshold, which is not a fixed value, but is calculated comprehensively according to the historical operation performance of the equipment, the allowed tolerance range provided by the manufacturer, and the actual safety specifications of the factory. When the path deviation value of a certain path node reported by the fault analysis module continuously exceeds this threshold, the decision engine of the path planning module will immediately start the re-evaluation program. The program first retrieves the historical path deviation data of the production area where the node is located, analyzes whether the trend is a temporary spike or a continuous deterioration, and cross-comparison with the current state of the associated area. Based on this comprehensive assessment, the module recalculates the real-time risk ranking of all online production areas in the factory. The original path node sequence priority is refreshed, and the position of high-risk areas in the sequence is greatly advanced, or even additional temporary inspection points may be inserted. For high-risk production area nodes showing a continuously increasing path deviation value, the module will dynamically compress its inspection cycle. The system automatically adjusts the scheduling algorithm to allocate more inspection time windows for such nodes. For example, a core compressor node that was originally inspected every two hours may have its inspection interval shortened to thirty minutes if its vibration and temperature deviation continues to rise, and a higher level of data acquisition mode may be triggered, such as increasing the frequency of vibration spectrum analysis.

[0063] The high-risk threshold, medium-risk threshold, and low-risk threshold relied on by the fault analysis module for fault level assessment are a set of multi-dimensional parameters accumulated and verified over a long period of data. These thresholds not only include the absolute limit of a single parameter, but more importantly, define the judgment boundary when multiple parameters are simultaneously abnormal. The assessment logic is based on a hierarchical decision tree model. For high-risk production area path nodes, the system will not easily determine a first-level fault simply because a single parameter exceeds the limit. The determination condition is usually that the vibration deviation, temperature deviation, and energy consumption deviation must all exceed the high-risk threshold set for them. The design of this "and" logic relationship effectively avoids excessive alarms caused by false reports or temporary disturbances of a single sensor. Only when multiple key indicators consistently point to a serious abnormality will the module mark the node as a first-level fault node. This means that there is a high probability of a chain reaction failure at this location, which requires the highest level of attention and intervention.

[0064] The assessment of medium-risk production area path nodes focuses on the loss of process stability. The marking of a second-level fault node does not depend on the instantaneous value of the flow or current exceeding the limit, but rather on whether its fluctuation coefficient has continuously and significantly exceeded the medium-risk threshold established based on the historical normal fluctuation model. For example, the current fluctuation threshold of a certain material pump is set based on the statistical fluctuation range during its stable operation in the past three months. If the currently calculated fluctuation coefficient not only exceeds this threshold, but also remains high for a period of time, it indicates that the equipment may be in an abnormal wear, unstable load, or control failure state, and is therefore marked as a second-level fault node. This marks that the production process in this area has deviated from the stable state, and there is a risk of quality deterioration or equipment downtime.

[0065] The marking of a third-level fault node for a low-risk production area focuses on the trend of abnormal changes in parameters, and the low-risk threshold is usually set as the allowed upper limit of the change rate. For example, the change rate threshold of warehouse environment humidity takes into account the characteristics of building materials, the hygroscopicity of stored goods, and the regular ventilation and dehumidification capacity. If the humidity change rate is abnormally high, it means that the temperature control system may be malfunctioning or there may be unexpected situations. Similarly, an abnormally high change rate in the frequency of device start-stop may indicate that the automatic control logic is disordered or manual operation is incorrect. When both of these change rate indicators exceed their low-risk thresholds, it indicates that although the current direct risk is not high, the development trend is unfavorable, and intervention is needed to prevent gradual deterioration. All fault level marking information, including fault type, level, associated parameters, and specific abnormal values, are packaged into structured data objects and transmitted in real time to the linkage execution module to provide decision-making basis for precise automated response.

[0066] Example 4: see Figure 4, the instruction generation engine first parses the fault level, type and specific abnormal parameters. For high-risk production areas marked as first-level fault nodes, the module accesses the pre-set linkage rule library. This rule library stores standardized processing plans for different equipment types under different fault modes. For example, for a large compressor node marked due to simultaneous vibration and temperature exceeding the standard, the module will generate a device speed reduction instruction and an environmental cooling instruction. The speed reduction instruction is directly issued to the frequency converter controller of the compressor through the OPCUA protocol, and the instruction contains a specific target speed value, which is calculated based on the current speed and the deviation degree, aiming to immediately switch the device to a pre-set safe operation mode. The cooling instruction is sent to the intelligent ventilation system of the plant and the cooling unit of the unit itself, and the instruction increases the fan speed or the cooling water flow to quickly remove heat. The instruction issuance is not a simple on-off control, but an analog quantity set value containing specific control parameters.

[0067] For medium-risk production areas marked as second-level fault nodes, such as material conveying pumps marked due to abnormal current fluctuation, the current stabilizing instruction generated by the module will be sent to the stabilizing power supply device or the drive controller of the pump, instructing it to switch to a more stringent voltage-current closed-loop control mode, and may temporarily relax the response time to reduce overshoot, thereby suppressing fluctuations. The material flow calibration instruction is sent to the positioner of the flow control valve, instructing it to fine-tune according to the latest flow sensor reading to eliminate the steady-state error between the set value and the actual value. For low-risk production areas marked as third-level fault nodes, such as a warehouse area marked due to abnormal humidity change rate, the humidity adjustment instruction triggered will be sent to the industrial dehumidifier, instructing it to adjust the compressor operating frequency or damper opening degree according to the deviation between the current humidity and the target value according to a specific algorithm. The device start-stop frequency limitation instruction is issued in the form of logical interlocking to the PLC of the devices in this area, and a time delay lock is temporarily inserted into its control logic to ensure that the number of device starts within a unit time does not exceed a newly set, lower safety limit. All instructions have a timestamp, instruction sequence number and expected response time, and are reliably transmitted to the corresponding actuators or control systems through the industrial network.

[0068] The partition monitoring module initiates high-frequency data collection immediately after the preset instruction execution waiting time. Its data collection strategy focuses on and strengthens the area that has just undergone linkage processing. For high-risk areas, the sampling frequency of vibration sensors and temperature sensors is temporarily increased to capture the transient response after instruction execution. The energy consumption monitoring unit synchronously performs high-precision power collection. For medium-risk areas, the collection period of current and flow data is shortened to observe the changes in fluctuations more finely. The humidity sensors and equipment status monitoring points of low-risk areas also enter high-frequency sampling mode. These updated data are labeled with the "after linkage" label, distinguished from the data before linkage, and transmitted to the central database.

[0069] The fault analysis module then processes this batch of "after linkage" data to calculate the post-linkage path deviation value. Its calculation logic remains consistent with the calculation of the original path deviation value, using the same weight coefficients, reference values, and standardization methods. This ensures the comparability of the two calculation results. The module compares the calculated new deviation value with the pre-linkage value and compares it again with various thresholds set internally, thereby making a quantitative evaluation of the effectiveness of the linkage processing measures. The evaluation result reveals whether the state of the production area has returned to the safe range, whether it has been partially improved, or whether the measures have been ineffective or even worsened.

[0070] Table 1: Linkage processing instruction execution effect evaluation summary.

[0071] Fault node area Fault level Trigger instruction type Main monitoring parameter Pre-linkage deviation value Post-linkage deviation value Deviation change trend Synthetic workshop A line Primary Equipment speed reduction, environmental temperature reduction Vibration value, temperature value 0.87 0.45 Down Raw material pump station B unit Secondary Current stabilization, flow calibration Current value, flow value 0.62 0.59 Slight decrease Finished product storage area C Tertiary Humidity adjustment, start-stop limitation Humidity value, start-stop frequency 0.41 0.38 Slight decrease

[0072] Referring to Table 1, the evaluation summary summarizes the key information in a structured manner, intuitively showing the change in deviation values of core operating parameters of different levels of fault nodes after a specific linkage intervention. The deviation change trend column reflects the immediate effect of the intervention measures, providing a clear overview for operations and maintenance personnel to quickly understand the overall processing situation and providing direct input basis for the next decision of the path planning module. The entire process from instruction generation, execution to effect verification forms a tightly coupled closed loop, enabling the system to continuously adjust strategies based on actual feedback data rather than relying on one-way open-loop control.

[0073] Example 5: Referring to Figure 5The path planning module receives the post-intervention path deviation value calculated by the fault analysis module. This value is a core quantitative indicator for evaluating the actual effect of the previously executed inter-processing commands. The module internally sets preset safety thresholds and preset fluctuation ranges. These thresholds and ranges are dynamic boundaries determined jointly by equipment design specifications, historical safe operation data statistics, and process requirements, rather than fixed values. If, after executing equipment deceleration and environmental cooling commands, the calculated post-intervention path deviation value for a high-risk production area still fails to decrease below the preset safety threshold set for its category, this indicates that the initial inter-processing intervention measures have not completely resolved the inherent risks in the area. This location may harbor undiscovered deep-seated faults or irreversible degradation of the equipment's inherent performance. In this situation, the module's path replanning algorithm will initiate special processing logic for this high-risk node. Generating a detour path for the high-risk production area becomes the preferred strategy. This detour path does not simply completely avoid the area where the node is located geographically, but rather involves a refined risk avoidance and monitoring trade-off. Path planning guides inspection units (such as robots) to alter their routes within a node, bypassing identified faulty equipment or high-risk sub-areas. For example, a robot might pass through a safe passage on the other side of a compressor unit instead of directly traversing its violently vibrating base. Simultaneously, the planning algorithm ensures that the new route still utilizes surrounding environmental sensors, cameras, or other non-contact monitoring devices to indirectly but effectively monitor the overall status of the faulty node continuously. This redirects inspection resources to other adjacent, critical points within the area, maximizing information acquisition while maintaining risk isolation.

[0074] For medium-risk production areas, if the calculated path deviation value after executing current stabilization and material flow calibration commands (especially its core parameters such as current fluctuation coefficient or flow fluctuation coefficient) continuously exceeds the preset fluctuation range calculated based on its historical stable operation data, this indicates that the operational instability of this area is persistent or intermittent, and conventional control methods are unlikely to quickly restore it to stability. To address this persistent fluctuation, the module generates a dual-track inspection path for the path nodes in the medium-risk production area. Dual-track inspection means constructing a parallel, time- or space-complementary verification inspection route in addition to the standard inspection task for that node. This can be achieved in two ways: first, scheduling the same inspection unit to repeatedly visit the node within shorter time intervals, thereby increasing the data acquisition frequency to capture its fluctuation patterns and transient characteristics; second, coordinating multiple inspection units (such as the main robot and backup robots) to conduct cross-inspections of the node from different physical paths or observation angles, using data redundancy to eliminate the possibility of false alarms from a single sensor and to more comprehensively assess the equipment status. This dual-track strategy significantly enhances the reliability of monitoring and diagnostic capabilities for complex fluctuation faults.

[0075] If the linkage path deviation value of a low-risk production area still exceeds the low-risk threshold set for it after the execution of the humidity adjustment instruction and the equipment start-stop frequency limit instruction, although its absolute risk level is not high, it indicates that the initial intervention has not effectively reversed its abnormal trend. The module's response to this is to generate a low-risk production area path node rapid patrol path. The core of planning this path is to optimize the movement efficiency and access frequency to the node. The algorithm recalculates the accessibility model in the factory map to select a path with the lowest time cost to the node, which may involve enabling shortcuts that are not usually used or prioritizing the use of patrol carriers with faster movement speeds. The goal is to increase the patrol frequency of the node with minimal resource consumption and the shortest response delay, so that it can more densely collect its state data and closely monitor its parameter change trend to prevent its abnormality from escalating.

[0076] The linkage execution module receives these three types of special path instructions from the path planning module: bypass path, dual-track patrol path, and rapid patrol path. Its primary task is to globally integrate and coordinate these heterogeneous path schemes tailored to the specific needs of different areas. The integration process is not simply splicing, but a complex optimization process that needs to address possible resource conflicts, time conflicts, and spatial conflicts between different path strategies. For example, to avoid scheduling conflicts caused by assigning a patrol unit with both dual-track patrol and rapid patrol tasks, the module will perform unified resource allocation and time scheduling. Finally, a globally optimized, logically unified final patrol path is generated. This final path macroscopically covers the overall factory patrol needs and microscopically embeds special patrol strategies for specific fault nodes. The module converts this final patrol path into an executable specific instruction sequence, i.e., an updated path node sequence. This sequence explicitly specifies the precise order of patrol unit visits to each node, the estimated arrival time, the length of stay at each node, and the specific set of detection tasks to be performed.

[0077] The partition monitoring module updates the path node sequence based on this, and immediately starts a new round of current production area operation data collection work. Compared with the initial inspection, the data collection strategy of the new round of collection process is more targeted and dynamic. For the nodes subjected to special path strategy, the frequency, type and accuracy of data collection will be adaptively adjusted according to the fault level and path requirements. For example, for the high-risk nodes that are bypassed, the data reading frequency of the surrounding sensors will be improved; for the medium-risk nodes that implement double-track inspection, more types of sensors will be activated simultaneously for collaborative measurement; for the low-risk nodes that are quickly inspected, a high-frequency but lightweight data collection mode is adopted. All the collected data are again fed into the system's data processing pipeline to provide the latest input for the fault analysis module, thereby starting a new "monitoring-analysis-planning-execution" closed loop cycle. This process is repeated, so that the entire system can continuously perceive the changes in the factory state, dynamically adjust the monitoring strategy, and continuously optimize its response through feedback evaluation of the execution effect, forming an intelligent operation and maintenance closed loop with learning and adaptation capabilities.

[0078] It should be noted that the relational terms herein, such as first and second, are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any such actual relationship or order between or among the entities or actions. The terms "comprises," "comprising," or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0079] Although embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, replacements and variations can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A distributed factory intelligent inspection path dynamic planning and fault linkage processing system, characterized in that, The system comprises a partition monitoring module, a path planning module, a fault analysis module and a linkage execution module. The partition monitoring module acquires current production area operation data of each production area of the distributed factory. The production areas of the distributed factory are divided into high-risk production areas, medium-risk production areas and low-risk production areas. The path planning module generates an initial inspection path based on the current production area operation data. The fault analysis module calculates a path deviation value according to real-time node operation state data of each path node on the initial inspection path. The linkage execution module triggers linkage processing instructions for the corresponding production area according to the path deviation value. The path planning module performs path dynamic adjustment according to the path deviation value: when the path deviation value exceeds a preset deviation threshold, the path node sequence priority of the high-risk production area, the medium-risk production area and the low-risk production area is re-distributed. When the path deviation value of the high-risk production area path node continuously increases, the inspection interval length of the high-risk production area path node is shortened. When the path deviation value of the medium-risk production area path node exceeds a preset fluctuation range, the redundant inspection frequency of the medium-risk production area path node is increased. The fault analysis module generates a fault level evaluation result based on the path deviation value: if the real-time high-risk equipment vibration value deviation degree, the high-risk environment temperature value deviation degree and the high-risk energy consumption value deviation degree associated with the path deviation value all exceed a high-risk threshold, the high-risk production area path node is marked as a first-level fault node. If the real-time medium-risk equipment current value fluctuation coefficient and the medium-risk material flow value fluctuation coefficient associated with the path deviation value exceed a medium-risk threshold, the medium-risk production area path node is marked as a second-level fault node. If the real-time low-risk environment humidity value change rate and the low-risk equipment start-stop frequency change rate associated with the path deviation value exceed a low-risk threshold, the low-risk production area path node is marked as a third-level fault node. The linkage execution module generates linkage processing instructions according to the first-level fault node, the second-level fault node and the third-level fault node: a high-risk production area equipment speed reduction instruction and an environment cooling instruction are triggered for the first-level fault node. A medium-risk production area current stabilizing instruction and a material flow calibration instruction are triggered for the second-level fault node. A low-risk production area humidity adjusting instruction and an equipment start-stop frequency limiting instruction are triggered for the third-level fault node. 2.The distributed plant intelligent inspection path dynamic planning and fault linkage processing system according to claim 1, characterized in that, The partition monitoring module specifically performs the following operations: The real-time high-risk equipment vibration value, the high-risk environment temperature value and the high-risk energy consumption value of the high-risk production area are collected. The real-time medium-risk equipment current value and the medium-risk material flow value of the medium-risk production area are collected. The real-time low-risk environment humidity value and the low-risk equipment start-stop frequency of the low-risk production area are collected. 3.The distributed plant intelligent inspection path dynamic planning and fault linkage processing system according to claim 2, characterized in that, The path planning module generates an initial inspection path and a path node sequence covering the high-risk production area, the medium-risk production area and the low-risk production area through a regional risk weight distribution algorithm based on the real-time high-risk equipment vibration value, the high-risk environment temperature value and the high-risk energy consumption value of the high-risk production area, in combination with the real-time medium-risk equipment current value and the medium-risk material flow value of the medium-risk production area, and in association with the real-time low-risk environment humidity value and the low-risk equipment start-stop frequency of the low-risk production area. 4.The distributed plant intelligent inspection path dynamic planning and fault linkage processing system according to claim 3, characterized in that, The fault analysis module obtains real-time node running state data of each path node in the path node sequence during initial path inspection path execution; The path deviation value is calculated based on the real-time node running state data and the preset node running standard value, wherein the path deviation value of the path node in the high-risk production area is associated with the real-time high-risk equipment vibration value deviation, the high-risk environment temperature value deviation, and the high-risk energy consumption value deviation, the path deviation value of the path node in the medium-risk production area is associated with the real-time medium-risk equipment current value fluctuation coefficient and the medium-risk material flow value fluctuation coefficient, and the path deviation value of the path node in the low-risk production area is associated with the real-time low-risk environment humidity value change rate and the low-risk equipment start-stop frequency change rate. 5.The distributed plant intelligent inspection path dynamic planning and fault linkage processing system according to claim 1, characterized in that, The partition monitoring module updates the high-risk equipment vibration value, the high-risk environment temperature value, and the high-risk energy consumption value in the high-risk production area, updates the medium-risk equipment current value and the medium-risk material flow value in the medium-risk production area, and updates the low-risk environment humidity value and the low-risk equipment start-stop frequency in the low-risk production area after the linkage processing instruction is executed. The fault analysis module calculates the path deviation value after linkage based on the updated high-risk equipment vibration value, the updated high-risk environment temperature value, the updated high-risk energy consumption value, the updated medium-risk equipment current value, the updated medium-risk material flow value, the updated low-risk environment humidity value, and the updated low-risk equipment start-stop frequency. 6.The distributed plant intelligent inspection path dynamic planning and fault linkage processing system according to claim 5, characterized in that, The path planning module executes path re-planning according to the path deviation value after linkage: if the path deviation value after linkage in the high-risk production area does not decrease to the preset safety threshold, a bypass path of the path node in the high-risk production area is generated; if the path deviation value after linkage in the medium-risk production area continuously exceeds the preset fluctuation range, a double-track path inspection path of the path node in the medium-risk production area is generated; if the path deviation value after linkage in the low-risk production area exceeds the low-risk threshold, a fast path inspection path of the path node in the low-risk production area is generated.

7. The distributed factory intelligent patrol path dynamic planning and fault linkage processing system according to claim 6, characterized in that, The linkage execution module integrates the bypass path of the path node in the high-risk production area, the double-track path inspection path of the path node in the medium-risk production area, and the fast path inspection path of the path node in the low-risk production area into a final path inspection path, and updates the path node sequence based on the final path inspection path; The partition monitoring module starts a new round of current production area running data collection based on the updated path node sequence.

Images