A method for monitoring and warning of mechanical and electrical equipment failures

Abstract

The present invention discloses a method for fault monitoring and early warning of electromechanical equipment, including: collecting the high-voltage power data and low-voltage power data of subway electromechanical equipment, respectively preprocessing and time-synchronizing the high-voltage power data and the low-voltage power data to obtain a high-voltage power data sequence and a low-voltage power data sequence; performing attention cross-fusion on the high-voltage power data sequence and the low-voltage power data sequence and then inputting them into a fault diagnosis model to obtain the fault type and its confidence level; inputting the fault type into a pre-constructed fault propagation knowledge graph to query and obtain the potential fault propagation path; inputting the high-voltage power data sequence, the low-voltage power data sequence, the fault type and its confidence level, and the potential fault propagation path into a Bayesian network to dynamically calculate the fault propagation probability and generate an early warning level.

Description

Technical Field

[0001] The present invention relates to the technical field of electromechanical equipment fault monitoring, and particularly to a method for monitoring and warning of electromechanical equipment faults. Background Art

[0002] Subway electromechanical equipment is an important infrastructure for ensuring the safe operation of the subway. Subway electromechanical equipment is divided into a strong electricity part and a weak electricity part. The strong electricity part (such as 1500V DC traction power supply, 380V power distribution, etc.) undertakes high-power loads and needs to meet the high-energy consumption requirements of train traction, environmental control ventilation, escalators, etc.; the weak electricity part (signal control, communication transmission, BAS environmental monitoring, etc.) covers the equipment interaction network throughout the line and has the characteristics of low voltage, wide coverage, and high sensitivity.

[0003] Traditional manual inspection of electromechanical equipment is basically in a passive state of receiving abnormalities. The existing technology obtains image status information through image recognition of images taken of these electromechanical equipment, that is, specific alarm indicator lights on the subway electromechanical equipment light up or the dial pointers are abnormal to discover abnormalities. In this case, it is very difficult to predict potential fault hazards in advance. Once an alarm occurs, it means that the situation is relatively serious and countermeasures must be taken within a short time. If there is no immediate response, it is very likely to affect the safety of subway operation.

[0004] Furthermore, the existing inspection methods do not distinguish the characteristics of the strong and weak electricity systems, mix the strong electricity part (such as power supply equipment) and the weak electricity part (such as signal control equipment) with the same detection logic, and ignore the characteristics of the time-series data of strong electricity monitoring and the discrete data characteristics of weak electricity monitoring, resulting in a high false alarm and missed alarm rate. In addition, there is a deep coupling between the strong electricity and weak electricity of subway electromechanical equipment (such as traction power supply fluctuations triggering signal system error codes). The existing technology lacks the ability to model the fault propagation path across systems and cannot warn of derivative faults (such as voltage dips causing communication interruptions).

[0005] Designing a method for monitoring and warning of electromechanical equipment faults to address the above problems existing in the prior art is the purpose of the research of the present invention. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to propose a method for monitoring and warning of electromechanical equipment faults, which can solve the above problems.

[0007] The present invention provides a method for monitoring and warning of electromechanical equipment faults, including:

[0008] Collecting the strong electricity data and weak electricity data of subway electromechanical equipment, respectively preprocessing and time-synchronizing the strong electricity data and weak electricity data to obtain a strong electricity data sequence and a weak electricity data sequence;

[0009] After performing attention cross - fusion on the strong - current data sequence and the weak - current data sequence, input them into the fault diagnosis model to obtain the fault type and its confidence level;

[0010] Input the fault type into the pre - constructed fault propagation knowledge graph to query and obtain the potential fault propagation paths;

[0011] Input the strong - current data sequence, the weak - current data sequence, the fault type and its confidence level, and the potential fault propagation paths into the Bayesian network to dynamically calculate the fault propagation probability and generate a warning level.

[0012] Furthermore, the acquisition of the strong - current data and weak - current data of subway electromechanical equipment, and the pre - processing and time synchronization of the strong - current data and weak - current data respectively to obtain the strong - current data sequence and the weak - current data sequence include:

[0013] Collect the voltage data and current data of subway electromechanical equipment, filter the voltage data and current data to obtain the strong - current data sequence;

[0014] Collect the communication status data and control instruction data of subway electromechanical equipment through network logs, perform error checking and cleaning on the communication status data, and record the time stamps for the control instruction data to obtain the weak - current data sequence.

[0015] Furthermore, the communication status data includes: packet loss rate, latency, bit error rate;

[0016] The control instruction data includes: equipment operation start - stop signal, parameter adjustment signal.

[0017] Furthermore, the process of performing attention cross - fusion on the strong - current data sequence and the weak - current data sequence and then inputting them into the fault diagnosis model to obtain the fault type and its confidence level includes:

[0018] Align the strong - current data sequence and the weak - current data sequence according to the time stamp, and perform normalization processing on the strong - current data sequence and the weak - current data sequence;

[0019] Fuse the normalized strong - current data sequence and weak - current data sequence through the multi - head attention mechanism to obtain a fused data sequence;

[0020] Input the fused data sequence into the fault classification model, aggregate the fused data sequence through the fully - connected layer of the fault classification model, and use the Softmax activation function to output the fault type and its confidence level.

[0021] Furthermore, the process of fusing the normalized strong - current data sequence and weak - current data sequence through the multi - head attention mechanism to obtain a fused data sequence includes:

[0022] Extract the high-dimensional feature representations of the strong electricity data sequence and the weak electricity data sequence respectively through a convolutional neural network to obtain the strong electricity high-dimensional feature representation and the weak electricity high-dimensional feature representation ;

[0023] Convert the strong electricity high-dimensional feature representation and the weak electricity high-dimensional feature representation into a strong electricity query matrix Q, a weak electricity index matrix K, and a weak electricity value matrix V respectively. The calculation formulas are as follows:

[0024] ,

[0025] ,

[0026] ,

[0027] where is the weight of the strong electricity query matrix Q, is the weight of the weak electricity index matrix K, is the weight of the weak electricity value matrix V;

[0028] Calculate the attention weights through the strong electricity query matrix Q, the weak electricity index matrix K, and the weak electricity value matrix V. The calculation formulas are as follows:

[0029] ,

[0030] where is the scaling factor, and softmax is the activation function to ensure weight normalization;

[0031] Fuse the strong electricity high-dimensional feature representation and the weak electricity high-dimensional feature representation through the attention weights to obtain the fused data sequence.

[0032] Furthermore, the fault propagation knowledge graph is constructed through the following steps:

[0033] Collect the historical strong electricity data of subway electromechanical equipment and their fault types, weak electricity data and their fault types, and expert knowledge and experience data;

[0034] Define the subway electromechanical equipment as equipment nodes, the corresponding strong electricity data and weak electricity data of the subway electromechanical equipment as state nodes, and the corresponding fault types of the subway electromechanical equipment as fault nodes;

[0035] Obtain the causal relationships and dependency relationships between the equipment nodes, state nodes, and fault nodes through the expert knowledge and experience data, and use the causal relationships and dependency relationships between the equipment nodes, state nodes, and fault nodes as edge relationships;

[0036] Set propagation weights for edge relationships according to expert knowledge and experience, and construct a fault propagation knowledge graph.

[0037] Furthermore, inputting the fault type into the pre-constructed fault propagation knowledge graph and querying to obtain the potential fault propagation paths includes:

[0038] Take the fault type as the query starting point of the knowledge graph, and use the depth-first search algorithm to start from the query starting point to find the associated propagation paths;

[0039] Select the associated propagation paths with propagation weights greater than the weight threshold as the potential fault propagation paths.

[0040] Furthermore, the Bayesian network is obtained through the following steps:

[0041] Extract the fault nodes as device status nodes through the potential propagation paths, and take the intermediate status nodes obtained from the potential propagation paths as path propagation nodes;

[0042] Based on the propagation path weights of the potential fault propagation paths, define the causal relationship edges from the parent nodes to the child nodes, initialize the conditional probability table, and construct the Bayesian network.

[0043] Furthermore, inputting the strong electrical data sequence, weak electrical data sequence, fault type and its confidence level, and potential fault propagation paths into the Bayesian network, dynamically calculating the fault propagation probability and generating a warning level includes:

[0044] Set the nodes corresponding to the strong electrical data sequence and weak electrical data sequence as Boolean observed evidences, and input the fault confidence level as the prior probability of the parent nodes into the Bayesian network;

[0045] Traverse the paths in the Bayesian network, calculate the fault propagation probability of the parent nodes to the child nodes level by level, and generate a warning level according to the fault propagation probability.

[0046] Furthermore, generating a warning level according to the fault propagation probability includes:

[0047] If the fault propagation probability ≤ 0.5, it is defined that the fault influence range is limited, and a low-level warning is generated;

[0048] If 0.5 < the fault propagation probability ≤ 0.8, it is defined that the fault may affect some devices, and a medium-level warning is generated;

[0049] If the fault propagation probability > 0.8, it is defined that the fault may cause a system-level fault, and a high-level warning is generated.

[0050] The beneficial effects of the present invention:

[0051] First, by collecting the strong and weak power data of subway electromechanical equipment and through processing such as filtering, verification, and cleaning, the quality of the original data is significantly improved, and the interference of noise and error codes is reduced. Appropriate preprocessing means are adopted for different signals to provide the optimal input for subsequent multi-source fusion, feature extraction, and analysis.

[0052] Second, the high-order features of strong and weak power are respectively extracted through a convolutional neural network, making full use of the spatial and internal structure of the signals. The multi-head attention mechanism realizes the dynamic association of heterogeneous feature points, automatically captures the key fault symptoms and the diagnostic clues of multi-factor coupling, and greatly improves the fault recognition accuracy and generalization ability of the model. The attention distribution weights reflect the importance of strong and weak power data in the diagnosis conclusion during a fault, which helps to improve interpretability. The fault type and accurate confidence level are output through the fault classification model to achieve soft decision-making and uncertainty quantification, facilitating subsequent probability modeling and hierarchical early warning.

[0053] Third, by converting the expert experience and historical events accumulated offline into a machine-readable knowledge graph, the causal relationship modeling of compound, hierarchical, and cross-device is realized. The depth-first search and weight screening mechanism can quickly discover the fault propagation paths that may have a wide impact / cross the system, filling the shortcoming that traditional systems cannot capture complex chain risks, realizing the screening of possible fault propagation paths, and effectively controlling the reasoning complexity and false positives / missed reports.

[0054] Fourth, use the knowledge graph to guide the automatic modeling of the Bayesian network to solve the problems of expandable and interpretable modeling of the fault propagation chain. The conditional probability table (CPT) realizes the risk quantification under multi-factor coupling, allows dynamic input of real-time confidence, and converts qualitative knowledge into quantitative analysis. The probability calculation is propagated layer by layer, and the "low / medium / high" early warning is output hierarchically to realize the risk stratification prompt of dynamic cross-strong and weak power systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0056] Figure 1 It is the method flow chart of this embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] For the convenience of those skilled in the art to understand, the embodiments will now be further described in detail in conjunction with the drawings for the structure of the present invention. It should be understood that the steps mentioned in this embodiment, unless specifically stating their order, can be adjusted according to actual needs in their front and back order, and even can be executed simultaneously or partially simultaneously.

[0058] As shown Figure 1 in the figure, an embodiment of the present invention provides a method for monitoring and warning of mechanical and electrical equipment failures, including:

[0059] S1 Collect the strong current data and weak current data of subway mechanical and electrical equipment, preprocess and synchronize the time of the strong current data and weak current data respectively to obtain a strong current data sequence and a weak current data sequence;

[0060] S101 Collect the voltage data and current data of subway mechanical and electrical equipment, filter the voltage data and current data to obtain a strong current data sequence;

[0061] S102 Collect the communication status data and control instruction data of subway mechanical and electrical equipment through network logs, perform error checking and cleaning processing on the communication status data, and record the time stamp of the control instruction data to obtain a weak current data sequence.

[0062] Furthermore, the communication status data includes: packet loss rate, delay, bit error rate;

[0063] The control instruction data includes: equipment operation start / stop signal, parameter adjustment signal.

[0064] In this step, subway mechanical and electrical equipment usually operates coordinately by a strong current system (drive, power supply) and a weak current system (control, communication). Abnormal strong current may affect the weak current. For example, voltage fluctuations may cause the control instruction to fail to execute. Abnormal weak current may cause strong current faults. For example, the loss of control signals may cause the motor to be overloaded or even trip.

[0065] The voltage and current of subway mechanical and electrical equipment are the core energy part of the equipment operation, and are usually used to directly monitor whether the equipment is in normal operation. Filtering is a key step to remove noise and outliers in the data to ensure the accuracy of the data.

[0066] The control instruction data (equipment operation start / stop signal, parameter adjustment signal) and communication status data (packet loss rate, delay, bit error rate) of subway mechanical and electrical equipment reflect the control logic of the equipment, the signal transmission quality, and whether the operation instructions are normal. If the communication signal is abnormal (such as an increase in the packet loss rate) or the control instruction cannot be executed correctly, it may cause problems in the operation of the strong current equipment.

[0067] The communication link will be affected by various interferences during data transmission, resulting in errors in the communication data. For example, due to network congestion or link failure, some data packets may be lost. Through verification means (such as CRC verification), errors in the data transmission process can be quickly detected to prevent incorrect data from entering the subsequent analysis process. Eliminate invalid information to avoid misleading the analysis results.

[0068] S2 inputs the strongly - electrified data sequence and the weakly - electrified data sequence after attention cross - fusion into the fault diagnosis model to obtain the fault type and its confidence level;

[0069] S201 aligns the strongly - electrified data sequence and the weakly - electrified data sequence according to the time stamp, and normalizes the strongly - electrified data sequence and the weakly - electrified data sequence;

[0070] In this step, the strongly - electrified data sequence and the weakly - electrified data sequence are normalized to the same range (such as [0, 1]) for subsequent fusion.

[0071] S202 fuses the normalized strongly - electrified data sequence and the weakly - electrified data sequence through the multi - head attention mechanism to obtain the fused data sequence;

[0072] S2021 respectively extracts the high - dimensional feature representations of the strongly - electrified data sequence and the weakly - electrified data sequence through the convolutional neural network to obtain the strongly - electrified high - dimensional feature representation and the weakly - electrified high - dimensional feature representation ;

[0073] S2022 respectively transforms the strongly - electrified high - dimensional feature representation and the weakly - electrified high - dimensional feature representation into the strongly - electrified query matrix Q, the weakly - electrified index matrix K, and the weakly - electrified value matrix V. The calculation formulas are as follows:

[0074] ,

[0075] ,

[0076] ,

[0077] Among them, is the weight of the strongly - electrified query matrix Q, is the weight of the weakly - electrified index matrix K, is the weight of the weakly - electrified value matrix V;

[0078] In this step, the query matrix Q usually represents the focus or query requirements for specific target data. The strong electrical data sequence is closely related to the direct operating state of the electromechanical equipment, and the changes in it are often the direct manifestation of the faults of the electromechanical equipment. The goal of the coordinated diagnosis of strong and weak electricity is to find the root cause or propagation path of the fault through the interactive analysis of strong and weak electrical data. When the strong electrical data shows anomalies (such as excessive current fluctuations), it is necessary to find which weak electrical signals (indices) and eigenvalues can explain this anomaly. Therefore, the strong electrical data is used as the query matrix Q to query the weak electrical data. The weak electrical data itself does not directly represent the operating state of the equipment, but is a partial reflection of the communication and control links. If the weak electricity is used as the query first, the analysis results obtained may only describe the communication state itself and cannot reflect the actual impact on the strong electrical state.

[0079] The index K of the weak electrical data (such as the time series of communication status and control instructions) can be regarded as an identifier describing the health of device control and communication. It is used to find the weak electrical information most relevant to the focus in the strong electricity. For example, when an anomaly occurs in the strong electricity at a certain time point, the weak electrical data may show the loss of control signals or the deterioration of communication quality at the corresponding time point, thus indicating the source of the problem. The value V in the weak electrical data is a possible explanation for the strong electrical state. For example, the packet loss rate can indicate whether the communication is interrupted, and the control instruction data can indicate whether an incorrect instruction has been sent to the device.

[0080] S2023 calculates the attention weights through the strong electrical query matrix Q, the weak electrical index matrix K, and the weak electrical value matrix V. The calculation formula is as follows:

[0081] ,

[0082] where is the scaling factor, and softmax is the activation function to ensure the normalization of the weights;

[0083] S2024 performs attention vector fusion on the strong electrical high-dimensional feature representation and the weak electrical high-dimensional feature representation to obtain the fused data sequence.

[0084] In this step, the attention mechanism can dynamically assign weights to different parts of the strong electrical data and the weak electrical data, capturing the important correlation relationships between them. For example: If there are abnormal fluctuations in the current in the strong electrical sequence and there are no changes in the control instructions in the weak electrical sequence, it may be a physical fault of the strong electrical equipment itself. If the communication status data (such as the packet loss rate) in the weak electrical sequence abnormally increases, it may be that the weak electrical control system has caused a fault in the strong electrical equipment.

[0085] S203 inputs the fused data sequence into the fault classification model, aggregates the fused data sequence through the fully connected layer of the fault classification model, and uses the Softmax activation function to output the fault type and its confidence level.

[0086] In this step, the fault classification model can be constructed using a neural network model. Historical high-voltage data sequences and corresponding timestamp low-voltage data sequences of subway electromechanical equipment are collected as sample inputs, and the manually labeled fault types and their confidence levels are used as sample outputs to train this fault classification model. The category with the highest probability is selected as the fault type, and this probability is output as the confidence level.

[0087] S3 inputs the fault type into the pre-constructed fault propagation knowledge graph to query and obtain the potential fault propagation paths;

[0088] Specifically, the fault propagation knowledge graph is constructed through the following steps:

[0089] Collect historical high-voltage data of subway electromechanical equipment and their fault types, low-voltage data and their fault types, and expert knowledge and experience data;

[0090] Define subway electromechanical equipment as equipment nodes, the corresponding high-voltage and low-voltage data of subway electromechanical equipment as state nodes, and the corresponding fault types of subway electromechanical equipment as fault nodes;

[0091] Obtain the causal relationships and dependency relationships between equipment nodes, state nodes, and fault nodes through expert knowledge and experience data, and use the causal relationships and dependency relationships between equipment nodes, state nodes, and fault nodes as edge relationships;

[0092] Set propagation weights for the edge relationships according to expert knowledge and experience to construct the fault propagation knowledge graph.

[0093] In this step, subway electromechanical equipment is jointly coordinated by a high-voltage system (such as drive and power supply) and a low-voltage system (such as control and communication). There are complex interactions between the high-voltage system and the low-voltage system. High-voltage faults (such as voltage fluctuations or current anomalies) will affect low-voltage functions (such as signal transmission or control logic errors). Low-voltage anomalies (such as signal packet loss or control command loss) may further trigger high-voltage system faults (such as motor overload or equipment tripping). The complexity between high-voltage and low-voltage is high, and traditional single-domain fault modeling methods cannot effectively capture the correlation relationships between them. In addition, electromechanical equipment faults may experience multi-hop propagation in multiple subsystems. For example: current fluctuation → control signal error → motor overload → tripping.

[0094] The current single-factor analysis method cannot effectively infer potential multi-step propagation paths and the root causes of faults. Traditional fault diagnosis methods, such as rule-based expert systems or statistical models, are often static and cannot dynamically adjust predictions according to changes in equipment status, and perform poorly on unknown faults or unseen correlation relationships.

[0095] Map strong electricity (such as current, voltage) data to state nodes, and also model relevant weak electricity data (control instruction data (equipment operation start-stop signals, parameter adjustment signals), communication status data (packet loss rate, delay, bit error rate)) as state nodes, and connect these nodes through causal relationships. For example: current fluctuation → voltage instability → control signal failure → motor trip. Using the graph structure of the knowledge graph, multi-step relationships from strong electricity problems to weak electricity problems can be represented, and the associated points can be inferred. Map electromechanical equipment to equipment nodes, and establish functional dependencies between equipment and other nodes, such as controller B → motor A. The knowledge graph combines domain expert knowledge to construct mapping relationships and propagation weights. At the same time, newly mined causal relationships (such as new equipment, new faults) from historical data can be added in batches to provide basic data for subsequent queries of potential fault propagation paths.

[0096] S301 Use the fault type as the query starting point of the knowledge graph, and start from the query starting point using the depth-first search algorithm to find the associated propagation path;

[0097] S302 Screen the associated propagation paths with propagation weights greater than the weight threshold as potential fault propagation paths.

[0098] In this step, depth-first search (DFS) is an algorithm used to traverse or search graph or tree data structures. It starts from the starting point (or root node), explores a branch as far as possible until it reaches the end point or there are no more unvisited nodes, and then backtracks to continue searching other branches. In the fault propagation knowledge graph, a certain node (such as a device or state in the knowledge graph) is often connected to multiple nodes through causal relationships, dependency relationships, etc. Depth-first search can start from a certain node and find all possible paths related to it.

[0099] Assume that the following propagation relationships are included in the knowledge graph: motor overcurrent → rectifier module failure (propagation probability 0.8), rectifier module failure → communication signal interruption (propagation probability 0.7). When "motor overcurrent" is input, the query results may be: Path 1: motor overcurrent → rectifier module failure (probability 0.8), Path 2: motor overcurrent → rectifier module failure → communication signal interruption (probability 0.56). Relevant propagation paths are obtained through a search algorithm, and then invalid paths with low propagation possibilities (low-weight paths) are removed through a weight threshold to focus on the key propagation chains. Necessary paths are retained to improve the computational efficiency of the subsequent Bayesian network.

[0100] S4 inputs the strong electricity data sequence, weak electricity data sequence, fault type and its confidence level, and the potential fault propagation path into the Bayesian network, dynamically calculates the fault propagation probability, and generates a warning level.

[0101] Specifically, the Bayesian network is obtained through the following steps:

[0102] Fault nodes are extracted from the potential propagation path as device status nodes, and intermediate status nodes obtained from the potential propagation path are used as path propagation nodes;

[0103] Based on the propagation path weights of the potential fault propagation path, the causal relationship edges from the parent node to the child node are defined, the conditional probability table is initialized, and the Bayesian network is constructed.

[0104] In this step, the fault propagation of subway electromechanical equipment involves multi-level dependency relationships. Traditional rule engines or static fault trees are difficult to quantify the dynamic propagation risk. The fault propagation is affected by multiple factors such as the environment, device status, and historical data, and there is a lot of uncertainty. In addition, there are various types of data signals in subway electromechanical equipment. The multi-level causal relationship can be intuitively represented through the directed acyclic graph (DAG) structure of the Bayesian network. The Bayesian network maps all data into nodes and uniformly models the dependency relationships through the conditional probability table (CPT). The Bayesian network quantifies the uncertainty through the probability distribution and transforms qualitative knowledge into quantitative probabilities.

[0105] The device status node, representing a specific fault (such as "motor overcurrent"), is the starting point and core of the reasoning. The path propagation node, representing the intermediate state (such as "rectifier module failure"), quantifies the probability attenuation in the propagation process and can decompose the complex propagation path into a computable probability chain. The propagation weight from expert experience or historical statistics (such as the weight of "motor overcurrent → rectifier module failure" being 0.8) is transformed into the conditional probability P(rectifier module failure | motor overcurrent)=0.8 to make the logical relationship quantifiable. The probability of the parent node is initialized using the fault confidence level output by S2 (such as P(motor overcurrent)=0.85) to ensure that the starting point of the reasoning is consistent with the actual diagnosis result.

[0106] S401 sets the corresponding nodes of the strong power data sequence and the weak power data sequence as Boolean observation evidences, and inputs the fault confidence as the prior probability of the parent node into the Bayesian network;

[0107] S402 traverses the paths in the Bayesian network, calculates the fault propagation probability of the parent node to the child node level by level, and generates a warning level according to the fault propagation probability.

[0108] S4021 If the fault propagation probability ≤ 0.5, it is defined that the fault influence range is limited, and a low-level warning is generated;

[0109] S4022 If 0.5 < the fault propagation probability ≤ 0.8, it is defined that the fault may affect some devices, and a medium-level warning is generated;

[0110] S4023 If the fault propagation probability > 0.8, it is defined that the fault may trigger a system-level fault, and a high-level warning is generated.

[0111] In this step, existing rule-based or static fault tree technologies often only focus on the fault chain within a single system and are difficult to model the "cross-system" fault propagation path. For example, after a voltage dip occurs in an electrical system, it may cause an interruption in the communication system link, but static analysis methods usually cannot perceive and warn of such derivative faults. Using a Bayesian network, based on the parent node (usually a known and directly observed direct fault), the occurrence probability of its possible downstream faults (child nodes) is inferred. Quantify the risk of a fault propagating from one node to other nodes, not just "whether there is" but "what is the probability of propagation". Realize real-time warning of dynamic and multi-level fault association in complex systems, not just relying on static logic or manual experience. Divide the propagation probability into three levels and give different severity warnings to the management system or operation and maintenance personnel.

[0112] For example, when a voltage dip occurs in the subway traction power supply system, the traditional operation and maintenance monitoring system can only issue a power supply fault warning and cannot insight into the further impact of the electrical fault on the communication system (such as the signal network) - for example, the voltage dip causes abnormal communication power supply, which ultimately leads to communication link jitter or even interruption. Existing rules and fault trees often cannot model and timely warn of such "cross-system" derivative risks.

[0113] Through step S3, "voltage dip → abnormal communication power supply → communication interruption" is selected as the potential propagation path. Through step S4, Bayesian network modeling is carried out for node setting. The nodes are voltage dip (equipment status node), abnormal communication power supply (intermediate propagation node), and communication interruption (equipment status node). CPT initialization is performed: P(abnormal communication power supply|voltage dip) = 0.75, P(communication interruption|abnormal communication power supply) = 0.8. The parent node "voltage dip" is input with a prior probability of 0.82 (obtained in step S2). Dynamic propagation probability calculation: P(abnormal communication power supply) = 0.82 × 0.75 = 0.615, P(communication interruption) = 0.615 × 0.8 = 0.492.

[0114] Early warning classification: The propagation probability of abnormal communication power supply is 0.615 (0.5 < 0.615 ≤ 0.8), and a medium-level early warning is generated. The propagation probability of communication interruption is 0.492 (≤ 0.5), and a low-level early warning is generated.

[0115] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0116] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram can be realized by computer program instructions, and the combination of processes and / or blocks in the flowchart and / or block diagram can also be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0117] These computer program instructions can also be stored in a computer-readable memory that can guide the computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including instruction means, and the instruction means realizes the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0118] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 or multiple blocks.

[0119] It should be noted that in the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In a unit claim listing several devices, several of these devices can be embodied by the same item of hardware. The use of the words first, second, and third, etc. does not denote any order. These words can be interpreted as names.

[0120] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn the basic creative concept. Therefore, the appended claims are intended to be construed to cover the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0121] Obviously, those skilled in the art can make various changes and deformations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and deformations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these changes and deformations.

[0122] In the present invention, unless otherwise clearly defined and limited, the terms "mounted", "connected", "coupled", "fixed", etc. shall be construed in a broad sense. For example, it can be a fixed connection, a detachable connection, or integrated; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal communication of two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0123] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms should not be understood as necessarily referring to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

Claims

1. A method for monitoring and warning of mechanical and electrical equipment failures, characterized in that, Including: Collect the strong electricity data and weak electricity data of subway electromechanical equipment, preprocess and time-synchronize the strong electricity data and weak electricity data respectively to obtain a strong electricity data sequence and a weak electricity data sequence; After performing attention cross-fusion on the strong electricity data sequence and the weak electricity data sequence, input them into a fault diagnosis model to obtain the fault type and its confidence level. Specifically: Align the strong electricity data sequence and the weak electricity data sequence according to the time stamp, and perform normalization processing on the strong electricity data sequence and the weak electricity data sequence; Fuse the normalized strong electricity data sequence and weak electricity data sequence through a multi-head attention mechanism to obtain a fused data sequence; Input the fused data sequence into a fault classification model, aggregate the fused data sequence through the fully connected layer of the fault classification model, and use the Softmax activation function to output the fault type and its confidence level; Input the fault type into a pre-constructed fault propagation knowledge graph to query and obtain the potential fault propagation path; Input the strong electricity data sequence, the weak electricity data sequence, the fault type and its confidence level, and the potential fault propagation path into a Bayesian network, dynamically calculate the fault propagation probability and generate a warning level. Specifically: Set the corresponding nodes of the strong electricity data sequence and the weak electricity data sequence as Boolean observation evidences, and input the fault confidence level as the prior probability of the parent node into the Bayesian network; Traverse the paths in the Bayesian network, calculate the fault propagation probability of the parent node to the child node level by level, and generate a warning level according to the fault propagation probability.

2. The method for monitoring and warning of mechanical and electrical equipment failures according to claim 1, characterized in that, The step of collecting the strong electricity data and weak electricity data of subway electromechanical equipment, preprocessing and time-synchronizing the strong electricity data and weak electricity data respectively to obtain a strong electricity data sequence and a weak electricity data sequence includes: Collect the voltage data and current data of subway electromechanical equipment, filter the voltage data and current data to obtain a strong electricity data sequence; Collect the communication status data and control instruction data of subway electromechanical equipment through network logs, perform error checking and cleaning processing on the communication status data, and record the time stamp for the control instruction data to obtain a weak electricity data sequence.

3. The method for monitoring and warning of mechanical and electrical equipment faults according to claim 2, characterized in that, The communication status data includes: packet loss rate, delay, bit error rate; The control instruction data includes: equipment operation start-stop signal, parameter adjustment signal.

4. A method for monitoring and warning of mechanical and electrical equipment failures according to claim 1, characterized in that, The step of fusing the normalized strong electricity data sequence and weak electricity data sequence through a multi-head attention mechanism to obtain a fused data sequence includes: Extract the high-dimensional feature representations of the strong electricity data sequence and the weak electricity data sequence respectively through a convolutional neural network to obtain the high-dimensional feature representation of strong electricity and the high-dimensional feature representation of weak electricity ; Convert the high-dimensional feature representation of strong electricity and the high-dimensional feature representation of weak electricity into a strong electricity query matrix Q, a weak electricity index matrix K, and a weak electricity value matrix V respectively. The calculation formulas are as follows: , , , Among them, is the weight of the strong electricity query matrix Q, is the weight of the weak electricity index matrix K, is the weight of the weak electricity value matrix V; Calculate the attention weights through a strong electricity query matrix Q, a weak electricity index matrix K, and a weak electricity value matrix V. The calculation formula is as follows: ， Among them, is the scaling factor, softmax is the activation function to ensure weight normalization; Represent the high-dimensional features of strong electricity through attention weights and the high-dimensional features of weak electricity Perform attention vector fusion to obtain a fused data sequence.

5. A method for monitoring and warning of mechanical and electrical equipment failures according to claim 1, characterized in that, The fault propagation knowledge graph is constructed through the following steps: Collect the historical strong electricity data of subway electromechanical equipment and its fault types, weak electricity data and its fault types, and expert knowledge and experience data; Define the subway electromechanical equipment as equipment nodes, define the corresponding strong electricity data and weak electricity data of the subway electromechanical equipment as state nodes, and define the corresponding fault types of the subway electromechanical equipment as fault nodes; Obtain the causal relationship and dependency relationship between the equipment nodes, state nodes, and fault nodes through expert knowledge and experience data, and use the causal relationship and dependency relationship between the equipment nodes, state nodes, and fault nodes as edge relationships; Set the propagation weight for the edge relationship according to expert knowledge and experience, and construct a fault propagation knowledge graph.

6. The method for monitoring and warning of mechanical and electrical equipment failures according to claim 5, characterized in that, Input the fault type into the pre-constructed fault propagation knowledge graph, and the obtained potential fault propagation paths include: Use the fault type as the query starting point of the knowledge graph, and start from the query starting point using the depth-first search algorithm to find the associated propagation path; Filter the associated propagation paths with propagation weights greater than the weight threshold as potential fault propagation paths.

7. A method for monitoring and warning of mechanical and electrical equipment failures according to claim 6, characterized in that The Bayesian network is obtained through the following steps: Extract the fault nodes as device status nodes through the potential propagation path, and use the intermediate status nodes obtained from the potential propagation path as path propagation nodes; Based on the propagation path weights of the potential fault propagation paths, define the causal relationship edges from the parent node to the child node, initialize the conditional probability table, and construct a Bayesian network.

8. A method for monitoring and warning of mechanical and electrical equipment failures according to claim 1, characterized in that, The generation of the warning level according to the fault propagation probability includes: If the fault propagation probability ≤ 0.5, it is defined that the fault influence range is limited, and a low-level warning is generated; If 0.5 < the fault propagation probability ≤ 0.8, it is defined that the fault may affect some devices, and a medium-level warning is generated; If the fault propagation probability > 0.8, it is defined that the fault may cause a system-level fault, and a high-level warning is generated.

Images