Intelligent equipment state monitoring and operation and maintenance management system based on Internet of Things

Abstract

The invention relates to the technical field of equipment management, and discloses an intelligent equipment state monitoring and operation and maintenance management system based on the Internet of Things, which comprises a system module assembly, and the system module assembly comprises an edge data acquisition unit, a fault tree analysis engine, an edge intelligent reasoning node, a distributed collaborative analysis platform and a cloud edge collaborative operation and maintenance center. The edge data is deployed in a sensor network of an equipment end, collects operation data in real time, supports multi-protocol access and adapts to heterogeneous equipment, and the fault tree analysis engine constructs a dynamic fault tree model based on historical fault data, optimizes an event association rule through machine learning, calculates the probability of a top event in real time, and analyzes the fault data in real time. And risk levels are divided. According to the method, a dynamic fault tree evolution mechanism is adopted, a cooperative fault is combined to predict the equipment degradation trend, the fault tree structure is adaptively adjusted, and the multi-modal data is fused with the vibration signal, the infrared thermal imaging and the current harmonic data to improve the diagnosis precision.

Description

Technical Field

[0001] This invention relates to the field of equipment management technology, and in particular to an intelligent equipment status monitoring and maintenance management system based on the Internet of Things. Background Technology

[0002] With the rapid development of the Industrial Internet and intelligent manufacturing, IoT devices are widely used in key sectors such as energy, manufacturing, and transportation. These devices typically operate in complex environments, requiring real-time sensing of environmental parameters (such as temperature, vibration, and current) and autonomous decision-making. However, existing technologies still face the following technical bottlenecks in equipment condition monitoring and operation and maintenance management:

[0003] 1. Lack of static risk assessment and regulation of low-risk equipment

[0004] Existing solutions typically rely on fixed thresholds to determine equipment status (e.g., triggering an alarm when the temperature exceeds 80°C). However, equipment health status changes dynamically with operating conditions (e.g., the temperature tolerance threshold decreases in older equipment). Furthermore, low-risk equipment, lacking continuous monitoring, may experience systemic risks due to cascading failures (e.g., sensor drift → controller misjudgment → actuator failure).

[0005] 2. Limited level of intelligence in operation and maintenance decision-making.

[0006] Existing systems largely rely on manual experience to set thresholds and rules, making it impossible to use machine learning to automatically discover fault modes and optimize operation and maintenance strategies. For example, they cannot use historical fault data to train models to predict the evolution path of "bearing wear → vibration spectrum shift → gearbox failure". Summary of the Invention

[0007] (a) Technical problems to be solved

[0008] To address the shortcomings of existing technologies, this invention provides an IoT-based intelligent device status monitoring and maintenance management system. It combines LSTM to predict device degradation trends, adaptively adjusts the fault tree structure, and integrates vibration signal (time domain), infrared thermal imaging (spatial), and current harmonic (frequency domain) data to improve diagnostic accuracy. Through IoT technology, it realizes the transformation from passive response to proactive prevention, which can reduce unplanned downtime by 30% and is suitable for scenarios with high reliability requirements such as wind farms and semiconductor production lines.

[0009] (II) Technical Solution

[0010] To achieve the above objectives, the present invention adopts the following technical solution:

[0011] An IoT-based intelligent device status monitoring and operation and maintenance management system includes a system module assembly. This assembly comprises an edge data acquisition unit, a fault tree analysis engine, an edge intelligent inference node, a distributed collaborative analysis platform, and a cloud-edge collaborative operation and maintenance hub. The edge data acquisition unit is deployed on the sensor network at the device end to collect operational data in real time. It supports multi-protocol access and adapts to heterogeneous devices. The fault tree analysis engine constructs a dynamic fault tree model based on historical fault data, optimizes event association rules through machine learning, calculates the probability of top events in real time, and classifies risk levels. The edge intelligent inference node performs feature extraction on high-risk devices, identifies key fault characteristics, and triggers localized operation and maintenance instructions through the rule engine. The distributed collaborative analysis platform performs cross-device association analysis on low-risk devices to identify potential cascading fault risks. The cloud-edge collaborative operation and maintenance hub uniformly manages device lifecycle data and provides predictive maintenance suggestions.

[0012] Furthermore, the correlation analysis method and steps of the distributed collaborative analysis platform are as follows:

[0013] 1) Physical topology modeling

[0014] A physical topology model is constructed using equipment installation location diagrams and electrical wiring diagrams.

[0015] 2) Data dependency analysis

[0016] Real-time monitoring of data flow between devices

[0017] The correlation strength of equipment parameters is calculated using the mutual information entropy algorithm.

[0018] 3) Cascade Fault Simulation

[0019] Fault propagation path simulation based on digital twin: using digital twin technology to simulate the operating status of equipment groups;

[0020] 4) Dynamic weight adjustment

[0021] Update association weights using a time decay factor.

[0022] Based on the aforementioned solution, the cloud-edge collaborative operation and maintenance hub mechanism is as follows:

[0023] 1) Operation and maintenance decision support mechanism:

[0024] Decision-making basis: Fault tree risk probability

[0025] Economic analysis model: comparing downtime losses and maintenance costs

[0026] 2) Predictive maintenance suggestion generation process

[0027] 3) Knowledge base-driven optimization, continuously updating the operation and maintenance strategy library.

[0028] Supports AR remote assistance and automatic work order generation.

[0029] As a further embodiment of the present invention, the system's runtime workflow is as follows:

[0030] S1: Data acquisition, edge sensors collect device parameters in real time;

[0031] S2: Preprocessing, filtering out noisy data and compressing the collected parameters before transmission through the edge gateway;

[0032] S3: Dynamic fault tree modeling. The fault tree analysis engine generates an initial fault tree based on historical fault database data and optimizes event weights by combining online learning.

[0033] S4: Real-time risk assessment, edge nodes calculate the probability of top event occurrence, if it is greater than 80%, trigger local alarm and isolate the device;

[0034] S5: Collaborative fault prediction. The distributed platform uses graph neural networks to model the dependencies between devices and identify weak correlation risks.

[0035] S6: Closed-loop operation and maintenance. AR glasses push maintenance guidance videos to on-site personnel, robotic arms automatically replace faulty parts, update the fault tree model, and enhance newly discovered fault modes.

[0036] Furthermore, the low-risk device data identified in the real-time risk assessment during S4 is uploaded to the cloud for multi-dimensional correlation analysis.

[0037] Based on the aforementioned scheme, S5 includes a red line threshold early warning module: when the key parameters of the device deviate from the threshold by more than 3σ and adjacent devices show the same trend, a global early warning is triggered.

[0038] The beneficial effects of this invention are as follows:

[0039] 1. In this invention, a dynamic fault tree evolution mechanism is adopted, which is combined with collaborative fault prediction of equipment degradation trend to adaptively adjust the fault tree structure. Furthermore, multimodal data fusion of vibration signal (time domain), infrared thermal imaging (spatial), and current harmonic (frequency domain) data improves diagnostic accuracy.

[0040] 2. In this invention, risk simulation driven by digital twins is used to simulate the equipment failure propagation path in virtual space and quantitatively assess the probability of "single point failure → system paralysis".

[0041] 3. In this invention, adaptive threshold setting dynamically adjusts the alarm threshold based on the device's health status.

[0042] 4. In this invention, the Internet of Things (IoT) technology enables a shift from passive response to proactive prevention, which can reduce unplanned downtime and is suitable for scenarios with high reliability requirements, such as wind farms and semiconductor production lines. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the system framework of an IoT-based intelligent device status monitoring and maintenance management system proposed in this invention.

[0044] Figure 2 This is a schematic diagram of the system workflow structure of an IoT-based intelligent device status monitoring and maintenance management system proposed in this invention. Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. It should be noted that, unless otherwise expressly specified and limited, the terms "installation", "connection", and "setting" should be interpreted broadly. For those skilled in the art, the specific meaning of the above terms in this patent can be understood according to the specific circumstances.

[0046] Reference Figure 1-2 An IoT-based intelligent device status monitoring and operation and maintenance management system includes a system module assembly. This assembly comprises an edge data acquisition unit, a fault tree analysis engine, edge intelligent inference nodes, a distributed collaborative analysis platform, and a cloud-edge collaborative operation and maintenance hub. The edge data acquisition unit is deployed on the device's sensor network to collect operational data (such as vibration, temperature, current, voltage, etc.) in real time. It supports multi-protocol access (MQTT / CoAP) and is adaptable to heterogeneous devices. The fault tree analysis engine (FTE) constructs a dynamic fault tree model based on historical fault data and utilizes machine learning... Optimize event association rules, calculate the probability of top events (such as "equipment downtime") in real time, and classify risk levels (high risk / low risk). Edge intelligent inference nodes perform feature extraction (such as frequency domain analysis and time series pattern matching) on ​​high-risk devices to identify key fault characteristics. Trigger localized operation and maintenance instructions (such as shutdown and load reduction) through the rule engine. The distributed collaborative analysis platform performs cross-device association analysis on low-risk devices to identify potential cascading fault risks (such as supply chain disruptions affecting multiple devices). The cloud-edge collaborative operation and maintenance hub uniformly manages equipment lifecycle data and provides predictive maintenance suggestions.

[0047] In this invention, the correlation analysis method and steps of the distributed collaborative analysis platform are as follows:

[0048] 1) Physical topology modeling

[0049] Construct a physical topology model (such as the hierarchical connection relationship between wind turbines, transformers, and the power grid in a wind farm) using equipment installation location diagrams and electrical wiring diagrams.

[0050] 2) Data dependency analysis

[0051] Real-time monitoring of data flow between devices (e.g., current fluctuations in device A → voltage harmonics in device B → temperature rise curve in device C).

[0052] The mutual information entropy algorithm is used to calculate the correlation strength of device parameters (formula: I(X;Y)=Σp(x,y)log(p(x,y) / p(x)p(y))).

[0053] 3) Cascade Fault Simulation

[0054] Fault propagation path simulation based on digital twins: The simulation of the operating status of equipment groups is carried out using digital twin technology; the simulation process includes sensor drift, error accumulation, controller misjudgment, erroneous commands, actuator overload, mechanical stress, and transmission chain breakage.

[0055] 4) Dynamic weight adjustment

[0056] Update the associated weights using the time decay factor: W_t=W_0*e^(-λt) (λ is the equipment aging coefficient).

[0057] 3. A smart device status monitoring and operation and maintenance management system based on the Internet of Things according to claim 2, characterized in that the cloud-edge collaborative operation and maintenance hub 1) operation and maintenance decision support mechanism:

[0058] Decision basis:

[0059] Fault tree risk probability (>80% triggers red alert)

[0060] Equipment Health Index (HI): HI = 1 - Σ(β_i * F_i) (β_i is the fault weight, F_i is the fault probability)

[0061] Economic analysis model: Comparing downtime losses vs. maintenance costs

[0062] 2) Predictive maintenance suggestion generation process, as shown in the following example:

[0063]

[0064] 3) Knowledge base-driven optimization, continuously updating the operation and maintenance strategy library, for example:

[0065] Failure Mode: Bearing Wear → Solution Library:

[0066] Short-term solution: Add grease (85% effectiveness)

[0067] --->Mid-term solution: Reduce load to 80% (extend lifespan by 3 months)

[0068] Fundamental solution: Replace with SKF imported bearings (estimated lifespan of 5 years).

[0069] Supports AR remote assistance and automatic work order generation.

[0070] In this invention, section 4 describes its runtime workflow as follows:

[0071] S1: Data acquisition, edge sensors collect device parameters in real time;

[0072] S2: Preprocessing, filtering out noisy data and compressing the collected parameters before transmission through the edge gateway;

[0073] S3: Dynamic fault tree modeling. The fault tree analysis engine generates an initial fault tree based on historical fault database data and optimizes event weights by combining online learning.

[0074] Example fault tree snippet:

[0075] Top event: Production line shutdown

[0076]

[0077] S4: Real-time risk assessment. Edge nodes calculate the probability of top events. If the probability is greater than 80%, a local alarm is triggered and the device is isolated. Low-risk device data in S4 are uploaded to the cloud for multi-dimensional correlation analysis.

[0078] S5: Collaborative fault prediction. The distributed platform uses graph neural networks to model the dependencies between devices and identify weak correlation risks. S5 includes a red line threshold early warning module: when the key parameters of a device deviate from the threshold by more than 3σ and adjacent devices show the same trend, a global early warning is triggered.

[0079] S6: Closed-loop operation and maintenance. AR glasses push maintenance guidance videos to on-site personnel, robotic arms automatically replace faulty parts, update the fault tree model, and enhance newly discovered fault modes.

[0080] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An IoT-based intelligent device status monitoring and operation and maintenance management system, comprising a system module assembly, characterized in that, The system module assembly includes an edge data acquisition unit, a fault tree analysis engine, an edge intelligent inference node, a distributed collaborative analysis platform, and a cloud-edge collaborative operation and maintenance hub. The edge data acquisition unit is deployed on the sensor network at the device end to collect operational data in real time. It supports multi-protocol access and is compatible with heterogeneous devices. The fault tree analysis engine builds a dynamic fault tree model based on historical fault data, optimizes event association rules through machine learning, calculates the probability of top events in real time, and classifies risk levels. The edge intelligent inference node performs feature extraction on high-risk devices, identifies key fault characteristics, and triggers localized operation and maintenance instructions through the rule engine. The distributed collaborative analysis platform performs cross-device association analysis on low-risk devices to identify potential cascading fault risks. The cloud-edge collaborative operation and maintenance hub uniformly manages device lifecycle data and provides predictive maintenance suggestions.

2. The IoT-based intelligent device status monitoring and operation and maintenance management system according to claim 1, characterized in that, The correlation analysis method and steps of the distributed collaborative analysis platform are as follows: 1) Physical topology modeling A physical topology model is constructed using equipment installation location diagrams and electrical wiring diagrams. 2) Data dependency analysis Real-time monitoring of data flow between devices The correlation strength of equipment parameters is calculated using the mutual information entropy algorithm. 3) Cascade Fault Simulation Fault propagation path simulation based on digital twin: using digital twin technology to simulate the operating status of equipment groups; 4) Dynamic weight adjustment Update association weights using a time decay factor.

3. The IoT-based intelligent device status monitoring and maintenance management system according to claim 2, characterized in that, The cloud-edge collaborative operation and maintenance central mechanism is as follows: 1) Operation and maintenance decision support mechanism: Decision-making basis: Fault tree risk probability Economic analysis model: comparing downtime losses and maintenance costs 2) Predictive maintenance suggestion generation process 3) Knowledge base-driven optimization, continuously updating the operation and maintenance strategy library. Supports AR remote assistance and automatic work order generation.

4. The IoT-based intelligent device status monitoring and operation and maintenance management system according to claim 1, characterized in that, The system's runtime workflow is as follows: S1: Data acquisition, edge sensors collect device parameters in real time; S2: Preprocessing, filtering out noisy data and compressing the collected parameters before transmission through the edge gateway; S3: Dynamic fault tree modeling. The fault tree analysis engine generates an initial fault tree based on historical fault database data and optimizes event weights by combining online learning. S4: Real-time risk assessment, edge nodes calculate the probability of top event occurrence, if it is greater than 80%, trigger local alarm and isolate the device; S5: Collaborative fault prediction. The distributed platform uses graph neural networks to model the dependencies between devices and identify weak correlation risks. S6: Closed-loop operation and maintenance. AR glasses push maintenance guidance videos to on-site personnel, robotic arms automatically replace faulty parts, update the fault tree model, and enhance newly discovered fault modes.

5. The IoT-based intelligent device status monitoring and operation and maintenance management system according to claim 4, characterized in that, The low-risk device data identified in S4 during real-time risk assessment is uploaded to the cloud for multi-dimensional correlation analysis.

6. The IoT-based intelligent device status monitoring and operation and maintenance management system according to claim 4, characterized in that, The S5 includes a red line threshold early warning module: when the key parameters of the device deviate from the threshold by more than 3σ and adjacent devices show the same trend, a global early warning is triggered.

Images