An intelligent electromechanical comprehensive monitoring system based on an industrial internet

Abstract

The application relates to the technical field of predictive maintenance and health management of industrial equipment, in particular to an intelligent electromechanical comprehensive monitoring system based on an industrial internet, which comprises the following steps: a data acquisition module acquires equipment dynamic operation parameter sets, inherent physical attribute parameters and historical maintenance data of target electromechanical equipment, and calls preset process topology data, economic loss data and maintenance cost data; a state evaluation module iteratively calculates a cumulative degradation index representing the health condition of the equipment; an importance quantification module calculates a process importance coefficient representing the key degree of the equipment in a production line based on the process topology data; a risk calculation module generates a production risk degree quantifying the influence of shutdown; and a decision generation module generates a dynamic maintenance decision instruction; the application realizes systematic risk cognition from equipment individuals to the whole production line, breaks away from the extensive management of treating all the equipment equally, and makes the allocation of maintenance resources more strategic and targeted.

Description

Technical Field

[0001] This invention relates to the field of predictive maintenance and health management technology for industrial equipment, specifically to an intelligent electromechanical integrated monitoring system based on the Industrial Internet. Background Technology

[0002] In industrial production environments, electromechanical equipment is the core of ensuring the continuous operation of production lines. To prevent equipment failures, existing technologies typically acquire equipment operating data through sensors and manage it in conjunction with pre-set maintenance plans.

[0003] Traditional maintenance strategies often assess the physical condition of individual devices in isolation, lacking a quantitative evaluation of the criticality of equipment based on the entire production line's process topology. They also fail to dynamically link the real-time health status of equipment with the potential economic losses caused by equipment failure. This assessment method results in an incomplete basis for maintenance decisions, making it difficult to scientifically prioritize maintenance tasks among numerous devices. Consequently, the allocation of maintenance resources lacks optimal guidance, posing a risk of unplanned downtime due to critical equipment failure, which threatens production continuity and economic efficiency. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides an intelligent electromechanical integrated monitoring system based on the Industrial Internet. Specifically, the technical solution of this invention is as follows:

[0005] An intelligent electromechanical integrated monitoring system based on the Industrial Internet includes:

[0006] The data acquisition module is used to acquire the set of dynamic operating parameters, inherent physical attribute parameters and historical maintenance data of the target electromechanical equipment, and to retrieve preset process topology data, economic loss data and maintenance cost data.

[0007] The status assessment module is used to iteratively calculate the cumulative degradation index, which characterizes the health status of the equipment, based on the set of dynamic operating parameters, inherent physical attribute parameters and historical maintenance data of the equipment obtained by the data acquisition module.

[0008] The importance quantification module is used to calculate the process importance coefficient, which characterizes the criticality of equipment in the production line, based on process topology data.

[0009] The risk calculation module is used to determine the probability of equipment failure based on the cumulative degradation index, and combine the process importance coefficient and economic loss data to generate a quantitative production risk level of the impact of downtime.

[0010] The decision generation module is used to determine the decision priority of maintenance tasks based on production risk and maintenance cost data, and generate dynamic maintenance decision instructions.

[0011] Preferably, the state assessment module is specifically used for:

[0012] Determine the baseline degradation amount based on inherent physical property parameters;

[0013] The overall disturbance is determined based on the set of dynamic operating parameters of the equipment;

[0014] The cumulative degradation index is generated by combining the baseline degradation amount with the comprehensive disturbance amount.

[0015] Preferably, the risk calculation module is used to determine the probability of equipment failure, including:

[0016] The cumulative degradation index is input into the preset logistic function model to calculate the probability of equipment failure.

[0017] Preferably, the importance quantification module is specifically used for:

[0018] Based on process topology data, determine the bottleneck index;

[0019] Determine the redundancy coefficient based on process topology data;

[0020] Based on process topology data, downstream influencing factors are determined;

[0021] By combining the bottleneck index, redundancy coefficient, and downstream influencing factors, a process importance coefficient is generated.

[0022] Preferably, the risk calculation module is used to generate a production risk level, including:

[0023] The production risk level is determined by multiplying the equipment failure probability, the process importance coefficient, and the economic loss data.

[0024] Preferably, the decision generation module is specifically used for:

[0025] Based on the preset degradation improvement coefficient and cumulative degradation index of the maintenance measures, the cumulative degradation index after maintenance is determined;

[0026] The cumulative degradation index after maintenance is input into the logistic function model to determine the failure probability after maintenance;

[0027] The residual production risk level is determined by combining the failure probability after maintenance, the process importance coefficient, and economic loss data.

[0028] Decision priorities are generated based on production risk level, residual production risk level, and maintenance cost data.

[0029] Preferably, the decision generation module is further used for:

[0030] Set a first response threshold and a second response threshold, wherein the second response threshold is greater than the first response threshold;

[0031] An emergency repair work order is generated if the decision priority is greater than the second response threshold.

[0032] In response to decisions with a priority greater than the first response threshold and less than or equal to the second response threshold, a planned maintenance recommendation is generated.

[0033] If the decision priority is less than or equal to the first response threshold, the repair task will be included in the routine maintenance plan.

[0034] Preferred options also include:

[0035] The feedback loop module is used to record the implementation status of maintenance activities and use it to update historical maintenance data.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. This technical solution establishes a deep quantitative assessment capability for equipment health status. Traditional technologies typically rely on isolated operating parameters or fixed time periods to determine maintenance needs, lacking precise insight into the true degradation state of the equipment. The status assessment module in this solution integrates the inherent physical attribute parameters of the equipment with a set of dynamic operating parameters, and takes into account the impact of historical maintenance data to iteratively calculate a cumulative degradation index characterizing the health status of the equipment. This index is not a simple parameter summation, but distinguishes and combines the baseline degradation caused by natural aging with the comprehensive disturbance caused by real-time operating condition fluctuations. This allows for a more physically interpretable and accurate quantification of the performance degradation process of the equipment throughout its entire life cycle from new to failure, laying a solid foundation for subsequent accurate prediction.

[0038] 2. This technical solution achieves a systematic risk assessment from individual equipment to the entire production line. Existing technologies often overlook the topological location and functional criticality of equipment in the entire production process when making maintenance decisions. This solution innovatively introduces an importance quantification module. Based on process topology data, this module generates a process importance coefficient by comprehensively evaluating the bottleneck index, redundancy coefficient, and downstream influencing factors of the equipment. This enables the system to objectively distinguish critical equipment with similar health conditions but vastly different downtime consequences, moving away from the extensive management that treats all equipment the same and making the allocation of maintenance resources more strategic and targeted.

[0039] 3. This technical solution seamlessly couples technical-level status monitoring with operational-level economic benefits, achieving economic optimization in decision-making. Traditional maintenance decisions are mostly based on technical indicators, lacking direct consideration of cost-effectiveness. This solution's risk calculation module integrates the probability of equipment failure representing physical status, the process importance coefficient reflecting the production line's role, and economic loss data directly related to operations to generate a quantified production risk level of downtime impact. Furthermore, the decision generation module determines the decision priority of maintenance tasks based on this production risk level and maintenance cost data. This design transforms complex maintenance problems into a clear risk reduction return on investment assessment, ensuring that limited maintenance resources are prioritized for investment in areas that can minimize overall production line risk and maximize economic benefits.

[0040] 4. This technical solution introduces a feedback closed-loop module to construct an intelligent system with adaptive and continuous evolution capabilities. Static monitoring and decision-making models become inaccurate over time and with changes in operating conditions. This solution can record the implementation of maintenance activities and use it to update historical maintenance data, continuously iterating and optimizing the relevant parameters of the internal state assessment model, risk calculation model, and decision generation model. This allows the system to learn from actual maintenance activities, and the accuracy of its predictions and the rationality of its decisions will continuously improve with the accumulation of data, transforming the entire monitoring system from a one-way analysis tool into a dynamic, high-efficiency intelligent entity that can maintain high performance over the long term. Attached Figure Description

[0041] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0042] Figure 1 This is a structural block diagram of an intelligent electromechanical integrated monitoring system based on the Industrial Internet of Things according to the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0044] Example 1:

[0045] Please see Figure 1 An intelligent electromechanical integrated monitoring system based on the Industrial Internet includes:

[0046] The data acquisition module is used to acquire the set of dynamic operating parameters, inherent physical attribute parameters and historical maintenance data of the target electromechanical equipment, and to retrieve preset process topology data, economic loss data and maintenance cost data.

[0047] The status assessment module is used to iteratively calculate the cumulative degradation index, which characterizes the health status of the equipment, based on the set of dynamic operating parameters, inherent physical attribute parameters and historical maintenance data of the equipment obtained by the data acquisition module.

[0048] The importance quantification module is used to calculate the process importance coefficient, which characterizes the criticality of equipment in the production line, based on process topology data.

[0049] The risk calculation module is used to determine the probability of equipment failure based on the cumulative degradation index, and combine the process importance coefficient and economic loss data to generate a quantitative production risk level of the impact of downtime.

[0050] The decision generation module is used to determine the decision priority of maintenance tasks based on production risk and maintenance cost data, and generate dynamic maintenance decision instructions.

[0051] This embodiment provides an intelligent electromechanical integrated monitoring system based on the Industrial Internet. The system constructs a complete technical closed loop from data acquisition, status assessment, risk calculation to decision generation, aiming to achieve accurate and forward-looking maintenance of electromechanical equipment in industrial production lines.

[0052] The system includes a data acquisition module, the purpose of which is to provide a comprehensive, multi-dimensional data foundation for all subsequent analysis and decision-making. In this embodiment, this module is implemented through an industrial internet platform. It utilizes sensors such as vibration, temperature, and pressure deployed on the target electromechanical equipment to acquire, in real time, a set of dynamic operating parameters constituting the equipment. It obtains multivariate time-series data; simultaneously, it retrieves inherent physical attribute parameters of the equipment from the equipment management system or digital ledger, such as the design reference life provided by the equipment manufacturer. This module also accesses and integrates static information such as material fatigue limits; furthermore, it retrieves and integrates historical maintenance data from the maintenance management system. It records past maintenance activities, replaced parts, and malfunctions of the equipment; to conduct system-level risk assessments and decisions, this module also retrieves preset process topology data from the process planning database and obtains economic loss data related to downtime of each piece of equipment from the enterprise resource planning system. And maintenance cost data required to implement different maintenance plans To make risk assessment more timely, the system can also be linked with the Manufacturing Execution System (MES) to analyze economic loss data based on dynamic factors such as current production line load and order urgency. Perform real-time corrections;

[0053] The system also includes a status assessment module, the core purpose of which is to transform multi-source raw data into a unified quantitative indicator that can accurately characterize the current health status of the equipment. In this embodiment, this module is based on a set of dynamic operating parameters of the equipment obtained by the data acquisition module. Inherent physical property parameters and historical maintenance data Through an evolutionary model that integrates physical mechanisms and data-driven approaches, the cumulative degradation index characterizing the health status of equipment is calculated iteratively. The design philosophy of this index is that the overall performance degradation of equipment is the result of the combined effect of its inherent chronic damage and acute impacts caused by external operating conditions.

[0054] The system also includes an importance quantification module, which aims to objectively assess the criticality of each piece of equipment in the production network from a global production line perspective. In this embodiment, based on process topology data, this module analyzes the network position, redundancy configuration, and impact range of equipment on downstream processes in the process flow diagram to calculate the process importance coefficient, which characterizes the criticality of equipment in the production line. This coefficient enables the system to distinguish between devices in the same health condition but with drastically different consequences of downtime.

[0055] The system also includes a risk calculation module, the purpose of which is to effectively integrate the physical state of the equipment, its technological role, and its economic consequences to generate a quantitative risk indicator that directly guides business decisions. In this embodiment, the first step of this module is to calculate the cumulative degradation index output by the condition assessment module. Determine the real-time failure probability of the equipment The process importance coefficient is calculated by combining the importance quantification module. Economic loss data obtained by the data acquisition module This generated a quantitative assessment of the production risk level caused by downtime. ;

[0056] The system includes a decision generation module, the purpose of which is to automatically generate maintenance instructions that are both technically sound and economically optimal based on quantified risk assessment results. In this embodiment, this module is based on the calculated production risk level. Compared with the estimated maintenance cost data By evaluating the risk reduction benefits of different maintenance measures, the decision-making priority of maintenance tasks can be determined. Based on this, dynamic maintenance decision instructions are generated;

[0057] The system disclosed in this embodiment, through the organic combination of the above modules, constructs a complete closed loop from bottom-level data perception to top-level intelligent decision-making. It can not only predict the potential failure of a single device, but also quantify the production risks that the failure may bring from the perspective of the entire production line, and propose the optimal maintenance decision based on a comprehensive consideration of risk and cost. This enables enterprises to leap from traditional passive response or planned prevention to precise intervention based on risk prediction, thereby maximizing the utilization efficiency of maintenance resources while ensuring production continuity, and effectively reducing the probability of unplanned downtime events and the economic losses they cause.

[0058] Example 2:

[0059] The status assessment module is specifically used for:

[0060] Determine the baseline degradation amount based on inherent physical property parameters;

[0061] The overall disturbance is determined based on the set of dynamic operating parameters of the equipment;

[0062] The cumulative degradation index is generated by combining the baseline degradation amount with the comprehensive disturbance amount.

[0063] The condition assessment module enables more accurate calculation of the cumulative degradation index. The degradation process of the equipment is broken down into two core parts and quantified separately. The specific implementation method is described below:

[0064] This module determines the baseline degradation amount based on inherent physical property parameters; baseline degradation amount This refers to the chronic, irreversible performance degradation of equipment under ideal operating conditions, caused solely by the passage of time and natural aging of materials. Its function is to provide a stable, growing baseline for the equipment's degradation model, reflecting its inherent lifespan loss. In this embodiment, its calculation method is as follows:

[0065]

[0066] in, The baseline degradation amount for the current calculation period is dimensionless and is calculated using this formula.

[0067] The inherent degradation coefficient is dimensionless and is calibrated based on the inherent properties of the equipment, such as materials and manufacturing processes, through factory testing or by referring to historical data of similar equipment.

[0068] The duration of the current calculation cycle compared to the previous cycle, in time units, is obtained from the system clock;

[0069] The design reference lifespan of the equipment, in units of time, is obtained from the equipment ledger by the data acquisition module.

[0070] This formula is derived from the classic fatigue cumulative damage theory, which ensures that even in an ideal state without any external disturbances, the health of the equipment will still decrease monotonically with the increase of operating time, which is in line with the laws of physics.

[0071] This module determines the overall disturbance quantity based on the set of dynamic operating parameters of the equipment; the overall disturbance quantity This refers to the additional, acute damage to equipment caused by real-time operating conditions deviating from the normal range; its function is to quantify the accelerated wear and tear on equipment lifespan due to fluctuations in the operating environment and load; in this embodiment, its calculation method is as follows:

[0072]

[0073] in, The total disturbance amount for the current calculation period is dimensionless and is calculated using this formula.

[0074] : Total number of sensor measurement points, integer, preset parameters;

[0075] :No. The weighting coefficients of each measurement point data are dimensionless and are determined based on expert experience or through data-driven methods such as principal component analysis.

[0076] :No. The current real-time reading of each measuring point corresponds to the physical dimension, and its source is the real-time data acquisition module.

[0077] and The historical mean and standard deviation of this measuring point under normal operating conditions, corresponding to physical dimensions, are obtained through statistical analysis of historical normal operating data; among them, It is a preset, extremely small positive number, used as a stabilizing term to prevent the standard deviation from being affected by the extremely stable readings of the sensor under normal operating conditions. When the value is zero or close to zero, a division-by-zero error occurs, thereby enhancing the computational robustness of the model;

[0078] This formula uses the Z-score standardization approach to normalize sensor data from multiple sensors with different dimensions, and then performs a weighted summation to form a unified disturbance index that comprehensively reflects the degree of deviation from the current operating condition. It should be noted that the current model assumes that the impact of each disturbance factor on equipment damage is linearly additive. In practical applications, to capture potential nonlinear coupling effects between physical quantities, this model can be upgraded to a nonlinear function, for example... ,in It is a nonlinear mapping model trained through machine learning methods such as neural networks, in order to more comprehensively characterize complex damage mechanisms;

[0079] This module combines the baseline degradation amount and the comprehensive disturbance amount to generate a cumulative degradation index. In this embodiment, the cumulative degradation index is updated iteratively by weighting and fusing the two components. :

[0080]

[0081] in, The cumulative degradation exponent at time t is dimensionless and is calculated iteratively using this formula.

[0082] The cumulative degradation exponent at the previous moment is dimensionless and originates from the output of the system's previous calculation cycle. Its initial value is... It can be set to 0;

[0083] The weighting coefficients are dimensionless, and their calibration process aims to ensure that the degradation trajectory predicted by the model best matches the actual observed degradation process of the equipment. Specifically, a calibration dataset containing multiple sets of equipment from operation to failure can be prepared. For any sample in this dataset, its actual degradation degree at any given time can be determined by an observable physical quantity. To characterize; by employing optimization algorithms such as the least squares method, to find an optimal set of... This makes the formula calculated as follows: Sequence and corresponding actual degradation level The sum of squared errors between sequences is minimized;

[0084] By clearly distinguishing between baseline degradation and comprehensive disturbance, the state assessment model in this embodiment has stronger physical interpretability. It not only considers the inevitable aging of equipment over time but also accurately quantifies the dynamic impact of real-time operating conditions, resulting in a more accurate cumulative degradation index. It can more accurately and robustly reflect the true health status of the equipment, laying a solid foundation for the accurate prediction of subsequent failure probability.

[0085] Example 3:

[0086] The risk calculation module is used to determine the probability of equipment failure, including:

[0087] The cumulative degradation index is input into the preset logistic function model to calculate the probability of equipment failure.

[0088] When determining the probability of equipment failure, the risk calculation module establishes a cumulative degradation index. With the probability of eventual failure The nonlinear mapping relationship between them is implemented as follows:

[0089] This module inputs the cumulative degradation index into a preset logistic function model to calculate the device failure probability. The logistic function model is an S-shaped curve function that can map any real number input to the (0,1) interval. Its function is to simulate the nonlinear process of a system gradually transitioning from a stable normal state to an unstable failure state as internal degradation accumulates. In this embodiment, the specific mathematical expression of the model is:

[0090]

[0091] in, The failure probability of the i-th device at time t is dimensionless, with a range of (0,1), and is calculated using this formula.

[0092] The cumulative degradation index of the device at time t is dimensionless and is calculated by the state assessment module.

[0093] : Scale parameter, dimensionless, controls the steepness of the probability curve;

[0094] Threshold parameter, dimensionless, represents the threshold value at which the device degrades to a 50% failure probability. The critical point;

[0095] Parameter source: parameters and The calibration is based on a failure dataset containing a large number of historical samples of similar devices; each sample in this dataset contains two types of information: the cumulative degradation index at the time of device failure. And the cumulative degradation index when the equipment has not failed. ; Find an optimal set of values ​​using statistical fitting methods such as maximum likelihood estimation. This makes the failure probability calculated by the model for all failed samples approach 1, while the failure probability calculated for all non-failed samples approaches 0.

[0096] In the real world, the failure process of equipment is often not linear, but rather deteriorates rapidly after its performance degrades to a certain critical point. The S-shaped characteristic of the logistic function can well simulate this nonlinear transformation process from quantitative to qualitative change, and can provide a more realistic failure probability prediction compared to a simple linear model.

[0097] Using the logistic function model, the internal, abstract cumulative degradation exponent can be expressed. Scientifically transformed into an external, statistically significant failure probability. This nonlinear mapping relationship better reflects the actual physical process of failure of electromechanical equipment, avoids prediction bias caused by linear assumptions, and thus significantly improves the accuracy and reliability of risk assessment.

[0098] Example 4:

[0099] The importance quantification module is specifically used for:

[0100] Based on process topology data, determine the bottleneck index;

[0101] Determine the redundancy coefficient based on process topology data;

[0102] Based on process topology data, downstream influencing factors are determined;

[0103] By combining the bottleneck index, redundancy coefficient, and downstream influencing factors, a process importance coefficient is generated.

[0104] The importance quantification module comprehensively evaluates the process importance of equipment by calculating the process importance coefficient from different dimensions of the network structure. The specific implementation method is described below:

[0105] This module determines the bottleneck index based on process topology data; bottleneck index This refers to the number of paths passing through a given equipment node in the entire material flow path of a production network; its function is to measure the equipment's pivotal role in the overall process flow; in this embodiment, the betweenness centrality algorithm, well-known in network science, is used to calculate it. ;

[0106] Simultaneously, this module determines the redundancy coefficient based on process topology data; the redundancy coefficient... This refers to the completeness of the backup or parallel units configured for a device node, which can take over its functions when it fails; its role is to quantify the robustness of the system in the face of single-point failures; in this embodiment, It is a preset parameter with a value range of 0 to 1, and its source is the process design document;

[0107] Furthermore, this module determines downstream influencing factors based on process topology data; downstream influencing factors This refers to the total number of downstream nodes reachable from a given equipment node on the process topology graph; its function is to measure the impact of potential material supply disruptions on downstream processes should the equipment stop. In this embodiment, the number of downstream nodes is counted by performing a depth-first search or breadth-first search on the graph. The total number of reachable nodes from the starting point, thus obtaining ;

[0108] Based on the aforementioned sub-indicators, this module combines the bottleneck index, redundancy coefficient, and downstream influencing factors to generate a process importance coefficient. In this embodiment, the three sub-indicators are fused together to obtain the process importance coefficient. :

[0109] ;

[0110] in, The process importance coefficient of the i-th equipment is dimensionless and is calculated by this formula.

[0111] The bottleneck index of node i, an integer, is calculated using the betweenness centrality algorithm;

[0112] The maximum value of the bottleneck index of all nodes in the network, an integer, which is obtained by traversing the bottleneck index of all nodes in the network;

[0113] The redundancy coefficient of node i is a floating-point number between 0 and 1, which is derived from the preset process design.

[0114] : The downstream influence factor of node i, an integer, which is calculated through a graph search algorithm;

[0115] The total number of nodes in the network, an integer, is obtained from the statistical process topology map;

[0116] This method constructs a more comprehensive and objective quantitative model of importance than a single indicator by comprehensively considering three orthogonal dimensions: the pivotal nature, substitutability, and scope of influence of the equipment. It can accurately identify key equipment that, although not the most complex in itself, has systemic importance due to its special position in the production line, thus making risk assessment and resource allocation more targeted.

[0117] Example 5:

[0118] The risk calculation module is used to generate production risk levels, including:

[0119] The production risk level is determined by multiplying the equipment failure probability, the process importance coefficient, and the economic loss data.

[0120] The risk calculation module generates a quantitative indicator that can directly guide business decisions. The specific method for generating the production risk level is as follows:

[0121] This module determines the production risk level based on the product of equipment failure probability, process importance coefficient, and economic loss data; production risk level The calculation logic originates from the basic definition in the field of risk management, namely, risk = probability of event occurrence × consequence of event, and concretizes it in the technical environment of this invention; its function is to couple the physical health status of equipment, the role of the process network, and the financial impact to form a unified risk measure measured in monetary terms; in this embodiment, its calculation formula is:

[0122]

[0123] in, : The production risk level of the i-th device at time t, in monetary units, calculated by this formula;

[0124] The real-time failure probability of this device is dimensionless and is derived from the calculations performed by the risk calculation module in its preceding steps.

[0125] The process importance coefficient of this equipment is dimensionless and is obtained by calculation from the importance quantification module.

[0126] The estimated economic loss caused by a single unplanned downtime of the equipment, expressed in monetary terms, is obtained by the data acquisition module from business management systems such as ERP.

[0127] This embodiment successfully transforms multi-dimensional, abstract technical indicators into a single, specific economic indicator with clear business implications—production risk level—through this calculation method. This allows for direct comparison and ranking of the potential risks of different types and levels of equipment, providing managers with an extremely intuitive and powerful basis for decision-making, thereby seamlessly integrating technical-level equipment monitoring with enterprise-level risk management.

[0128] Example 6:

[0129] The decision generation module is specifically used for:

[0130] Based on the preset degradation improvement coefficient and cumulative degradation index of the maintenance measures, the cumulative degradation index after maintenance is determined;

[0131] The cumulative degradation index after maintenance is input into the logistic function model to determine the failure probability after maintenance;

[0132] The residual production risk level is determined by combining the failure probability after maintenance, the process importance coefficient, and economic loss data.

[0133] Decision priorities are generated based on production risk level, residual production risk level, and maintenance cost data.

[0134] The decision generation module determines decision priorities to optimize the allocation of maintenance resources. The specific implementation method is as follows:

[0135] For subsequent calculations, this module determines the cumulative degradation index after maintenance based on preset maintenance measure degradation improvement coefficients and cumulative degradation indexes; maintenance measure degradation improvement coefficients This refers to the quantification of the improvement effect of a specific maintenance activity on the physical degradation of equipment, with a value ranging from 0 to 1. Its function is to establish a quantitative relationship between maintenance actions and the improvement of equipment health status, representing the cumulative degradation index after maintenance. The calculation formula is:

[0136] ;

[0137] in, The estimated cumulative degradation index after maintenance is dimensionless and is calculated using this formula.

[0138] The current cumulative degradation index before maintenance is dimensionless and is provided by the condition assessment module.

[0139] The degradation improvement coefficient for maintenance measures is a floating-point number between 0 and 1. Its calibration process involves collecting a maintenance dataset containing multiple sets of historical records of similar maintenance activities; each set of records includes the degradation index before maintenance. and the actual degradation index after repair By performing statistical regression on the dataset, we can obtain... ;

[0140] To further improve the physical fidelity of the model, the effects of maintenance measures can be further refined. For example, maintenance activities can be divided into repair-type and preventative-type. Repair-type maintenance can directly reduce the cumulative degradation index, such as... ,in This refers to specific degradation associated with repaired parts; while preventative maintenance primarily reduces the rate of increase in future degradation, thus affecting degradation in subsequent cycles. or In terms of calculation, when generating decisions, the system can call more refined models to predict the maintenance effect based on the type of maintenance work order, thereby making the decision basis more accurate;

[0141] This module inputs the cumulative degradation index after maintenance into the logistic function model to determine the failure probability after maintenance. :

[0142]

[0143] This module combines post-repair failure probability, process importance coefficient, and economic loss data to determine the residual production risk. :

[0144] ;

[0145] This module generates decision priorities based on production risk level, residual production risk level, and maintenance cost data; decision priorities Defined as the reduction in risk that can be achieved per unit of maintenance cost, its calculation formula is as follows:

[0146] ;

[0147] in, The dimensionless ratio for the decision priority of maintenance tasks is calculated using this formula.

[0148] The current production risk level before performing maintenance, in monetary units, is provided by the risk calculation module.

[0149] : Estimated residual production risk after maintenance, in monetary units, derived from calculations in the preceding steps of this module;

[0150] The total cost required to perform this maintenance task, expressed in currency, is obtained from the maintenance management system by the data acquisition module; this is to address the cost of certain maintenance tasks. In special cases where the value is extremely low or zero, the system needs to perform boundary handling. When the cost is less than the preset minimum cost threshold, its decision priority can be adjusted. It can be directly set as a special highest priority code, or its cost can be calculated based on the lowest threshold, in order to avoid overflow of the calculation result due to the denominator being zero, and to ensure the logical completeness of priority sorting.

[0151] This decision prioritization method elevates maintenance decision-making from an experience-based or single-risk-indicator-based model to a refined operational model based on risk reduction return on investment. It can scientifically identify high-value maintenance tasks, ensuring that limited maintenance budgets and human resources are prioritized for those links that can minimize the overall risk of the production line, thereby optimizing the economic benefits of maintenance decisions.

[0152] Example 7:

[0153] The decision generation module is also used for:

[0154] Set a first response threshold and a second response threshold, wherein the second response threshold is greater than the first response threshold;

[0155] An emergency repair work order is generated if the decision priority is greater than the second response threshold.

[0156] In response to decisions with a priority greater than the first response threshold and less than or equal to the second response threshold, a planned maintenance recommendation is generated.

[0157] If the decision priority is less than or equal to the first response threshold, the repair task will be included in the routine maintenance plan.

[0158] To transform the quantified decision priorities into specific, executable operational instructions, the decision generation module also employs a hierarchical response mechanism:

[0159] This module sets a first response threshold and a second response threshold; the first response threshold... Second response threshold These are two preset values, among which Their function is to divide a continuous range of decision priority values ​​into three different levels of urgency; these thresholds can be set based on historical maintenance decision priorities. The distribution is determined through statistical analysis; one feasible approach is to analyze all historical tasks... The values ​​are sorted, and the top 10% quantiles are set as the second response threshold. This ensures that the highest-value tasks are identified as urgent; and sets the median value, which falls within the top 50% of the distribution, as the first response threshold. This method is used to distinguish between planned maintenance and routine maintenance that have significant value; it ensures that the threshold setting is supported by objective data and can be dynamically adjusted as data accumulates.

[0160] The module will calculate the decision priorities Compare with these two thresholds and execute the corresponding response action:

[0161] The response is based on a decision priority greater than the second response threshold, i.e. Generate an emergency repair work order;

[0162] The response is based on a decision priority greater than the first response threshold and less than or equal to the second response threshold. Generate planned maintenance recommendations;

[0163] The response is based on a decision priority less than or equal to the first response threshold, i.e. Incorporate repair tasks into the routine maintenance plan;

[0164] This hierarchical response mechanism provides clear logical rules for the automated execution of decisions; it transforms complex and continuous quantitative analysis results into concise, clear, and standardized operating instructions, greatly simplifying the maintenance management process. This not only improves response speed and execution efficiency, but also ensures that different personnel can take consistent and optimal action strategies when facing the same situation, thereby enhancing the standardization and reliability of the entire maintenance system.

[0165] Example 8:

[0166] The feedback loop module aims to enable the entire monitoring and decision-making system to possess self-learning and adaptive capabilities, ensuring continuous performance improvement. In this embodiment, the module's function is to record the implementation status of maintenance activities and use it to update historical maintenance data. Specifically, when a maintenance task is completed, the module captures and records the relevant execution data, stores it in a structured manner, and uses this data to update the historical maintenance data accessed by the data acquisition module. ;

[0167] This data feedback can trigger periodic or event-driven re-optimization of the system's internal models; for example, updated historical maintenance data. It can be used to recalibrate and optimize model parameters in the state assessment module. and This makes the degradation model's predictions more closely reflect the actual situation of the equipment; by comparing the actual performance data before and after maintenance with the model's predictions, the degradation improvement coefficient of the maintenance measures in the decision generation module can be optimized. Long-term accumulation of failure and maintenance data can be used to more accurately calibrate the parameters of the logistic function model in the risk calculation module. and ;

[0168] The introduction of the feedback loop module transforms this system from a one-way prediction-decision model into a dynamic, continuously evolving intelligent system of perception-cognition-action-learning. By continuously absorbing new real-world data to iteratively optimize its internal model, it ensures that the system's prediction accuracy and decision rationality do not diminish over time, but rather become more and more accurate due to the continuous accumulation of data, thereby achieving long-term and sustainable improvement in equipment maintenance efficiency.

[0169] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention; any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0170] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A smart electromechanical integrated monitoring system based on the Industrial Internet, characterized in that, include: The data acquisition module is used to acquire the set of dynamic operating parameters, inherent physical property parameters, and historical maintenance data of the target electromechanical equipment, and to retrieve preset process topology data, economic loss data, and maintenance cost data; the inherent physical property parameters include the design reference life and material fatigue limit provided by the equipment manufacturer; The condition assessment module is used to iteratively calculate the cumulative degradation index, which characterizes the health status of the equipment, based on the set of dynamic operating parameters, inherent physical attribute parameters, and historical maintenance data obtained by the data acquisition module; the historical maintenance data is used to recalibrate and optimize the model parameters for iteratively calculating the cumulative degradation index in the condition assessment module. The importance quantification module is used to calculate the process importance coefficient, which characterizes the criticality of equipment in the production line, based on process topology data. The process topology data characterizes the network position, redundancy configuration, and impact range of equipment on downstream processes in the process flow diagram. The risk calculation module is used to determine the probability of equipment failure based on the cumulative degradation index, and combine the process importance coefficient and economic loss data to generate a quantitative production risk level of the impact of downtime. The decision generation module is used to determine the decision priority of maintenance tasks based on production risk and maintenance cost data, and generate dynamic maintenance decision instructions.

2. The intelligent electromechanical integrated monitoring system based on the Industrial Internet according to claim 1, characterized in that, The status assessment module is specifically used for: Determine the baseline degradation amount based on inherent physical property parameters; The overall disturbance is determined based on the set of dynamic operating parameters of the equipment; The cumulative degradation index is generated by combining the baseline degradation amount with the comprehensive disturbance amount.

3. The intelligent electromechanical integrated monitoring system based on the Industrial Internet according to claim 1, characterized in that, The risk calculation module is used to determine the probability of equipment failure, including: The cumulative degradation index is input into the preset logistic function model to calculate the probability of equipment failure.

4. The intelligent electromechanical integrated monitoring system based on the Industrial Internet according to claim 1, characterized in that, The importance quantification module is specifically used for: Based on process topology data, determine the bottleneck index; Determine the redundancy coefficient based on process topology data; Based on process topology data, downstream influencing factors are determined; By combining the bottleneck index, redundancy coefficient, and downstream influencing factors, a process importance coefficient is generated.

5. The intelligent electromechanical integrated monitoring system based on the Industrial Internet according to claim 1, characterized in that, The risk calculation module is used to generate production risk levels, including: The production risk level is determined by multiplying the equipment failure probability, the process importance coefficient, and the economic loss data.

6. The intelligent electromechanical integrated monitoring system based on the Industrial Internet according to claim 1, characterized in that, The decision generation module is specifically used for: Based on the preset degradation improvement coefficient and cumulative degradation index of the maintenance measures, the cumulative degradation index after maintenance is determined; The cumulative degradation index after maintenance is input into the logistic function model to determine the failure probability after maintenance; The residual production risk level is determined by combining the failure probability after maintenance, the process importance coefficient, and economic loss data. Decision priorities are generated based on production risk level, residual production risk level, and maintenance cost data.

7. The intelligent electromechanical integrated monitoring system based on the Industrial Internet according to claim 1, characterized in that, The decision generation module is also used for: Set a first response threshold and a second response threshold, wherein the second response threshold is greater than the first response threshold; An emergency repair work order is generated if the decision priority is greater than the second response threshold. In response to decisions with a priority greater than the first response threshold and less than or equal to the second response threshold, a planned maintenance recommendation is generated. If the decision priority is less than or equal to the first response threshold, the repair task will be included in the routine maintenance plan.

8. The intelligent electromechanical integrated monitoring system based on the Industrial Internet according to claim 1, characterized in that, Also includes: The feedback loop module is used to record the implementation status of maintenance activities and use it to update historical maintenance data.

Images