Hydraulic hoist fault diagnosis method and system based on multi-modal data fusion

Abstract

The invention discloses a hydraulic hoist fault diagnosis method and system based on multi-modal data fusion, and relates to the technical field of hydraulic hoist fault diagnosis, and the method comprises the steps: installing a sensor for data collection, amplifying the collected data, carrying out the real-time recognition of abnormal sound, and carrying out the data classification; according to the classified data, establishing an association relationship between a historical state and a current state of the hydraulic system, and judging a system fault mode; analyzing a fault propagation path and positioning a fault source according to an association relationship among hydraulic system components, and setting key elements in combination with historical records and an expert database; and full-life-cycle management is constructed for the hydraulic hoist and key elements. The running state of the hydraulic hoist system is comprehensively monitored, the sensitivity of fault detection is improved, the accuracy and efficiency of fault diagnosis are improved, the service life of equipment is prolonged, the maintenance cost is reduced, the risk of accidental shutdown is reduced through preventive maintenance, and technical guarantee is provided for reliable running and maintenance of the hydraulic hoist.

Description

Technical Field

[0001] The present invention relates to the technical field of hydraulic hoist fault diagnosis, and specifically to a hydraulic hoist fault diagnosis method and system based on multimodal data fusion. Background Art

[0002] With the development of science and technology, various complex systems such as aviation, aerospace, electricity, manufacturing and other fields have higher and higher requirements for system reliability and safety. As a cutting-edge technical means, the fault prediction and health management system can predict the occurrence of faults in advance through the analysis and mining of system operation data, and take corresponding measures to intervene, effectively avoiding the impact of faults on energy.

[0003] Therefore, fault prediction and health management systems have been widely studied and have become a hot topic and difficulty in current research. The research on fault prediction and health management systems involves multiple disciplines, including data mining, machine learning, statistics, etc. At present, researchers at home and abroad have proposed various fault prediction and health management system models and algorithms based on the application backgrounds in different fields. At present, a set of intelligent fault diagnosis and health management system for hydraulic gate hoists is being developed to monitor hydraulic gate hoists, and to perform quantitative, qualitative, and positioning diagnosis on the entire set of hydraulic gate hoists, so as to improve the working efficiency of hydraulic gate hoists and the service life of equipment, and ensure the safe and stable operation of the power system.

[0004] Although the existing fault diagnosis technology can detect and identify the abnormal state of the hydraulic system to a certain extent, there are still many shortcomings. First, the existing data acquisition methods often only focus on a single type of signal, ignoring the multi-source nature of hydraulic system faults, resulting in insufficient comprehensiveness of fault diagnosis. Secondly, the traditional fault diagnosis method lacks in-depth analysis of the correlation between historical status and current status, making it difficult to accurately determine the fault mode. In addition, the existing technology is relatively simple in analyzing the fault propagation path, making it difficult to accurately locate the source of the fault, thus affecting the timeliness and effectiveness of fault handling. Furthermore, the existing fault diagnosis system lacks full life cycle management of the hydraulic gate hoist and its key components, resulting in a lack of scientific basis for the formulation of maintenance decisions. The hydraulic gate hoist fault diagnosis method based on multimodal data fusion proposed in the present invention is aimed at the above-mentioned problems. By comprehensively using a variety of sensor data, a comprehensive monitoring system for the hydraulic system is constructed, aiming to improve the accuracy and efficiency of fault diagnosis and provide data support for maintenance decisions of the hydraulic gate hoist. Summary of the invention

[0005] In view of the above-mentioned existing problems, the purpose of the present invention is to solve the problem of how to effectively collect and analyze multi-source data of the hydraulic gate hoist system to achieve early identification and diagnosis of faults; to improve the accuracy and efficiency of fault diagnosis through data fusion technology, and to construct a correlation between the historical state and the current state of the hydraulic system to determine the failure mode of the system, how to analyze the propagation path of the fault between the components of the hydraulic system, and accurately locate the source of the fault; how to manage the hydraulic gate hoist and its key components throughout the life cycle to reduce the risk of failure and maintenance costs.

[0006] In order to solve the above technical problems, a hydraulic hoist fault diagnosis method based on multimodal data fusion is proposed, including:

[0007] Install sensors to collect data, amplify the collected data, and identify abnormal sounds in real time for data classification; build a correlation between the historical state and current state of the hydraulic system based on the classified data to determine the system failure mode; analyze the fault propagation path and locate the fault source through the correlation between the hydraulic system components, and set key components based on historical records and expert database; establish full life cycle management for hydraulic gate hoists and key components.

[0008] As a preferred solution of the hydraulic gate hoist fault diagnosis method based on multimodal data fusion described in the present invention, the data acquisition includes that the installed sensors include sound sensors, pressure sensors, temperature sensors and status monitoring sensors.

[0009] The scope of data collection is: full hydraulic hoist system, hydraulic cylinder, plunger pump, proportional pump, proportional valve, reversing valve, overflow valve, solenoid valve, filter, oil, oil temperature, and hoist on-site control cabinet.

[0010] As a preferred solution of the hydraulic gate hoist fault diagnosis method based on multimodal data fusion described in the present invention, the data classification includes extracting key features from each type of collected data, constructing a feature matrix for data amplification, merging the enhanced data into the original data set to form a feature set, extracting the feature set of sound, identifying the data in real time in turn, modeling the features using a multi-layer feedforward network, determining data classification, and inputting suspected fault data into the hydraulic system.

[0011] As a preferred solution of the hydraulic gate hoist fault diagnosis method based on multimodal data fusion described in the present invention, wherein: the system failure mode judgment includes building an association relationship through historical state and current state, performing feature matching on the current and historical states, building an association matrix, and judging the system failure mode includes leakage failure, jamming failure, system overload, and control failure.

[0012] As a preferred solution of the hydraulic hoist fault diagnosis method based on multimodal data fusion described in the present invention, wherein: the fault propagation path includes: representing each component in the hydraulic system as a node V i , make correlation judgments between nodes and build association relationships between hydraulic system components;

[0013] Extract the fault propagation path P through the depth-first search of the graph Path :

[0014] P Path = {V i ,V j ,V n},Weighted ij >th path ·C path

[0015] Among them, P Path is a qualified fault propagation path, V i 、V j 、V n is the node passed during the fault propagation process, th path The threshold used to judge the importance of the transmission path, C path To adjust the propagation restriction factor affecting the standard, Weighted ij For component V i and V j The weighted edges between .

[0016] As a preferred solution of the hydraulic gate hoist fault diagnosis method based on multimodal data fusion described in the present invention, wherein: the fault source positioning includes selecting the node with the maximum score Q according to the weight and influence degree of each node in the propagation path. i The node is the fault source Q source = argmax i (Q i ), combining historical records and expert database to set key components, where Q i For node V i Fault source score, Q source To determine the fault source location, argmax i (Q i ) is to maximize Q i The node index of , indicating the fault source.

[0017] As a preferred solution of the hydraulic gate hoist fault diagnosis method based on multimodal data fusion described in the present invention, the full life cycle management includes: conducting life cycle analysis on each key component of the hydraulic gate hoist, identifying risks and maintenance strategies under different failure modes, evaluating the health status of key components based on data collected during the operation cycle, quantifying operation in combination with historical data, and formulating and optimizing maintenance decisions based on fault prediction results and life cycle cost analysis.

[0018] Another object of the present invention is to provide a hydraulic gate hoist fault diagnosis system based on multimodal data fusion. The present invention uses a variety of sensors to collect and preprocess data, improves the accuracy and reliability of fault diagnosis, solves the complexity of fault mode recognition, overcomes the difficulty of fault source location, and optimizes maintenance decisions. The system is designed to monitor and respond to state changes of the hydraulic system in real time, track historical states, and effectively identify and locate faults, reduce maintenance costs, improve equipment operating efficiency, ensure the stable operation of the hydraulic gate hoist, and extend its service life.

[0019] As a preferred solution of the hydraulic gate hoist fault diagnosis system based on multimodal data fusion described in the present invention, it is characterized by including a data acquisition and preprocessing module, a fault mode recognition module, a fault location module, and a full life cycle management module.

[0020] The data acquisition and preprocessing module is responsible for installing and configuring various sensors to collect multimodal data of the hydraulic hoist system. The collected data is amplified and processed, and abnormal sounds are identified in real time to perform preliminary data classification.

[0021] The fault mode recognition module amplifies and extracts features by constructing a feature matrix, models features using a multi-layer feedforward network, and calculates the total error to update the threshold. The module constructs an association matrix through feature matching and historical state association analysis, and calculates the optimal matching path to determine the system fault mode.

[0022] The fault location module represents hydraulic system components as nodes, constructs association relationships and weighted graphs between nodes, determines how the fault propagates throughout the system by calculating the fault propagation path, locates the fault source through a scoring system, and updates the key component set table.

[0023] The full life cycle management module is responsible for performing life cycle analysis on the hydraulic gate hoist and its key components, evaluating the health status and life cycle cost of the components, and formulating and optimizing maintenance decisions based on fault prediction results and cost analysis.

[0024] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the hydraulic hoist fault diagnosis method based on multimodal data fusion are implemented.

[0025] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the hydraulic hoist fault diagnosis method based on multimodal data fusion are implemented.

[0026] Beneficial effects of the present invention: The present invention realizes comprehensive monitoring of the operating status of the hydraulic gate hoist system and improves the sensitivity of fault detection by installing and utilizing multiple sensors for data collection, and amplifying the collected data for real-time identification and classification. Furthermore, by constructing the correlation between the historical state and the current state of the hydraulic system, the present invention accurately determines the system fault mode and improves the accuracy and efficiency of fault diagnosis. By analyzing the correlation between the components of the hydraulic system and the fault propagation path, the present invention accurately locates the source of the fault, reduces the fault handling time, improves the maintenance efficiency, and reduces the maintenance cost. By combining historical records and expert databases to set key components, the present invention provides data support for the maintenance of the system, ensures the reliability of key components, and guarantees the stable operation of the entire system. Finally, through the full life cycle management of the hydraulic gate hoist and key components, the present invention prolongs the service life of the equipment, reduces maintenance costs, and reduces the risk of unexpected downtime through preventive maintenance. Overall, the method of the present invention realizes the intelligent and precise fault diagnosis through fine technical details, and provides a strong technical guarantee for the reliable operation and maintenance of the hydraulic gate hoist. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work, among which:

[0028] Figure 1 An overall flow chart of a hydraulic hoist fault diagnosis method based on multimodal data fusion provided by one embodiment of the present invention.

[0029] Figure 2 A system solution flow chart of a hydraulic hoist fault diagnosis system based on multimodal data fusion provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0031] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0032] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it an embodiment that is mutually exclusive with other embodiments, either individually or selectively.

[0033] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0034] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0035] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0036] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and provides a hydraulic hoist fault diagnosis method based on multi-modal data fusion, comprising:

[0037] S1: Install sensors to collect data, amplify the collected data, and identify abnormal sounds in real time for data classification.

[0038] Furthermore, Figure 1 ,Data acquisition includes ,the installed sensors include sound sensors, pressure sensors, temperature sensors and condition monitoring sensors.

[0039] The sound sensor collects abnormal sounds in the hydraulic system including wear and vibration sounds, and operating sound characteristics include the sounds when the equipment status changes.

[0040] The pressure sensor collects the working pressure in the hydraulic cylinder, the output pressure of the plunger pump and proportional pump, and the flow entering and leaving the hydraulic cylinder.

[0041] The temperature sensor collects the working temperature of the hydraulic system.

[0042] The condition monitoring sensor collects the telescopic displacement of the hydraulic cylinder, the specific working position of the gate hoist, the fault alarm of the control cabinet and the system status.

[0043] The scope of data collection includes the full hydraulic hoist system, hydraulic cylinder, plunger pump, proportional pump, proportional valve, reversing valve, overflow valve, solenoid valve, filter, oil product, oil temperature, and hoist on-site control cabinet.

[0044] It should be noted that the key features of each type of data collected are extracted, the feature matrix F0 is constructed, and the data is amplified F aug =F0+F rn , where F aug is the key data feature after amplification, F rn is a small noise of normal distribution. The enhanced data is merged into the original data set to form the feature set F exp =F0∪F aug , extract the feature set of the sound, identify the data in real time, and use a multi-layer feedforward network to model the features:

[0045] Z 1 =ReLU(W 1 F exp,i +b 1 )

[0046]

[0047] Among them, Z 1 is the feature representation of the first layer output, W1 is the weight matrix of the first layer, F exp,i is the i-th data in the feature set, i is the variable index, b1 is the bias vector of the first layer, ReLU is the activation function, The reconstructed output features are used, and the Decoder matches the final output with the input data.

[0048] Calculate the total error:

[0049] E total =E reconstruction +E temporal +ω·variance(E)

[0050] Among them, E total is the total error.

[0051] ω is the adjustment constant of the control error fluctuation to the total error, variance(E) is the variance of the error, E reconstruction is the reconstruction error:

[0052]

[0053] E temporal is the time-dependent error:

[0054]

[0055] Among them, N is the number of historical time points, F t-i is the input feature at the ti-th moment, The reconstructed features at the ti-th moment, λ temporal is the time-dependent weight.

[0056] Update the threshold Z using the distribution information of the current sample th =α·mean(E total ), where α is the adjustment factor, mean(E total ) is the average value of the total error, and the anomaly detection result is combined with the amplified data F' exp , enter Z 2 =ReLU(W 2 F' exp +b 2 ), where Z 2 is the feature representation of the second layer output, W 2 is the weight matrix of the second layer, b 2 is the bias vector of the second layer, when Z 2 <Z th When Z 2 ≥Z th When the data is detected, it is judged as suspected fault data and the suspected fault data is input into the hydraulic system.

[0057] S2: Build a correlation between the historical state and current state of the hydraulic system based on the classified data to determine the system failure mode.

[0058] Furthermore, an association relationship is established between the historical state and the current state.

[0059] Perform feature matching on current and historical states:

[0060]

[0061] Among them, M(i,j) is the matching measure between the current state and the historical state, F current (i) is the i-th eigenvalue of the current state when the fault data is suspected, F historical (j) is the jth eigenvalue of the historical state of the suspected fault data, i and j are both variable indexes, β is the weight coefficient for adjusting similarity, similarity(F i ,F j ) Current feature F i and historical characteristics F j The similarity measure between them, γ is the weight coefficient for adjusting noise, noise(F current,i ) is a noise measure associated with the current state.

[0062] Construct the incidence matrix:

[0063] R associative ={9S current ,S historical )|S current -S historical <∈+ζ·trend(T)}

[0064] Among them, R associative is the constructed association relationship set, S current is the similarity value of the current state, S historical is the similarity value of the historical state, ∈ is the threshold for judging the similarity difference, ζ is the weight coefficient for adjusting the trend influence, and trend(T) is the trend measurement of the current state.

[0065] By calculating the minimum path and obtaining the optimal matching path between the current state and the historical state:

[0066] S current,historical =min(M(i,j)+δ·R associative )

[0067] Among them, S current,historical is the similarity value between the current state and the historical state, and δ adjusts the weight coefficient of the variance impact.

[0068] Failure Mode F mod Including leakage fault F leak , stuck fault F sticking 、The system is overloaded F overload, control fault F Control :When S current,historical The value is greater than Indicates that the current status is similar to the characteristics of a leakage fault.

[0069] When S current,historical The value is greater than Indicates that the current status is a stuck fault fault mode.

[0070] When S current,historical The value is greater than Indicates that the current state is that the system is in an overload failure mode.

[0071] When S current,historical The value is greater than Indicates that the current state is a control fault fault mode.

[0072] in, The threshold used to confirm a leak fault, Threshold value for confirming stuck fault, Thresholds used to identify system overloads, Threshold used to acknowledge a control fault.

[0073] S3: Analyze the fault propagation path and locate the fault source through the correlation between hydraulic system components, and set key components based on historical records and expert database.

[0074] Furthermore, each component in the hydraulic system is represented as a node V i , to judge the correlation between nodes, to build the association relationship between hydraulic system components, and to extract the fault propagation path P through the depth-first search of the graph Path :

[0075] P Path = {V i ,V j ,V n},Weighted ij >th path ·C path

[0076] Among them, P Path is a qualified fault propagation path, V i 、V j 、V n is the node passed during the fault propagation process, th path The threshold used to judge the importance of the transmission path, C path To adjust the propagation restriction factor affecting the standard, Weighted ij For component V i and V j The weighted edges between .

[0077] It should be noted that according to the weight and influence of each node in the propagation path, the node with the maximum score Q i The node is the source of the failure:

[0078]

[0079] Combine historical records and expert database to set key components, among which Q i For node V i The fault source score of Neighbor(i) is the fault source score of node V i The set of neighbor nodes, d j is the neighbor node V j The relevant importance rating, For node V i With neighboring node V j The weighted value of is a fixed parameter for weight adjustment, count(P Path,i ) is the node V i The number of fault propagation paths.

[0080] S4: Establish full life cycle management for hydraulic gate hoists and key components.

[0081] Furthermore, a life cycle analysis is conducted on each key component of the hydraulic gate hoist to identify risks and maintenance strategies under different failure modes. The health status of key components is evaluated based on data collected during the operation cycle, and the operation is quantified based on historical data. Maintenance decisions are made and optimized based on failure prediction results and life cycle cost analysis.

[0082] 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 the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0083] Example 2, reference Figure 1 , which is the first embodiment of the present invention, and provides a hydraulic hoist fault diagnosis method based on multi-modal data fusion, comprising:

[0084] In hydraulic systems, failures are often the result of multiple factors. This method helps technicians find the root cause of the failure by analyzing the fault propagation path, so as to make fundamental repairs; continuously monitor the key components of the hydraulic gate hoist, evaluate their health status, and ensure stable operation of the system; optimize maintenance strategies, reduce maintenance costs, and extend equipment service life through full life cycle management of the hydraulic gate hoist and its key components.

[0085] S1: Install sensors to collect data, amplify the collected data, and identify abnormal sounds in real time for data classification.

[0086] Furthermore, Figure 1 ,Data acquisition includes installed sensors including sound sensors, pressure sensors, temperature sensors and condition monitoring sensors.

[0087] The scope of data collection includes the full hydraulic hoist system, hydraulic cylinder, plunger pump, proportional pump, proportional valve, reversing valve, overflow valve, solenoid valve, filter, oil product, oil temperature, and hoist on-site control cabinet.

[0088] It should be noted that key features are extracted from each type of data collected, a feature matrix is ​​constructed for data amplification, the enhanced data is merged into the original data set to form a feature set, the feature set of sound is extracted, the data is sequentially identified in real time, a multi-layer feedforward network is used to model the features, the data classification is determined, and the suspected fault data is input into the hydraulic system.

[0089] S2: Build a correlation between the historical state and current state of the hydraulic system based on the classified data to determine the system failure mode.

[0090] Furthermore, by building an association relationship between the historical state and the current state, feature matching is performed on the current and historical states, and an association matrix is ​​constructed to determine the system failure mode, including leakage failure, jam failure, system overload, and control failure.

[0091] S3: Analyze the fault propagation path and locate the fault source through the correlation between hydraulic system components, and set key components based on historical records and expert database.

[0092] Furthermore, each component in the hydraulic system is represented as a node V i , determine the correlation between nodes, build the association between hydraulic system components, determine the influence relationship between related component nodes according to the identified fault mode, and build a weighted graph ij And calculate the fault propagation path:

[0093]

[0094] Among them, P ij From Node V i To Node V j The minimum propagation path weight of Path(i,j) from node V i To Node V j The path, For node V i and the next node V in the path n , n is the variable index, and Hyp is the hyperparameter for enhancing the priority of high-impact paths.

[0095] Extract the fault propagation path P through the depth-first search of the graph Path :

[0096] P Path = {V i ,V j ,V n},Weighted ij >th path ·C path

[0097] Among them, P Path is the fault propagation path P obtained in ij The fault propagation paths that meet the conditions are screened out, V i 、V j 、V n is the node passed during the fault propagation process, th path The threshold used to judge the importance of the transmission path, C path To adjust the propagation limitation factor affecting the standard.

[0098] It should be noted that according to the weight and influence of each node in the propagation path, the node with the maximum score Q i The node is the source of the failure:

[0099]

[0100] Among them, Q i For node V i The fault source score of Neighbor(i) is the fault source score of node V i The set of neighbor nodes, d j is the neighbor node V j The relevant importance rating, For node V i With neighboring node V j The weighted value of is a fixed parameter for weight adjustment, count(P Path,i ) is the node V i The number of fault propagation paths.

[0101] A key component set table is constructed in the system. Each time the fault source is located, the failure count of the faulty component is updated in the system components of the fault source. The future failure probability is predicted based on the failure count. When the probability of failure of this component is greater than 75%, it is marked as a key fault component. Key protection components are set based on expert opinions, and the key fault components and key protection components are counted in the key component set table for key monitoring.

[0102] S4: Establish full life cycle management for hydraulic gate hoists and key components.

[0103] Furthermore, a life cycle analysis is conducted on each key component of the hydraulic gate hoist to identify the risks and maintenance strategies under different failure modes. The data collected during the operation cycle evaluates the health status of key components. Combined with historical data to quantify the operation, the life cycle cost evaluation model is:

[0104]

[0105] Among them, B lifecycle For the whole life cycle cost assessment, c maintenance (t) is the maintenance cost, including the cost of planned maintenance and unplanned breakdown repairs, future (t) is the future failure probability at time t, t' is the end point of the time range considered, and r1 and r2 are cost factors related to the future failure probability.

[0106] Make and optimize maintenance decisions based on failure prediction results and life cycle cost analysis.

[0107] Embodiment 3, the third embodiment of the present invention, is different from the first two embodiments in that:

[0108] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0109] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0110] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0111] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0112] Example 3, reference Figure 2 , which is the fourth embodiment of the present invention, provides a hydraulic gate hoist fault diagnosis system based on multimodal data fusion, including a data acquisition and preprocessing module 10, a fault mode recognition module 20, a fault location module 30, and a full life cycle management module 40.

[0113] The data acquisition and preprocessing module 10 is responsible for installing and configuring various sensors to collect multimodal data of the hydraulic hoist system. The collected data is amplified and processed, and abnormal sounds are identified in real time to perform preliminary data classification.

[0114] The fault mode recognition module 20 constructs a feature matrix, amplifies the data and extracts features, uses a multi-layer feedforward network to model the features, and calculates the total error to update the threshold. The module constructs an association matrix through feature matching and historical state association analysis, and calculates the optimal matching path to determine the system fault mode.

[0115] The fault location module 30 represents the hydraulic system components as nodes, constructs the association relationship and weighted graph between the nodes, determines how the fault propagates throughout the system by calculating the fault propagation path, locates the fault source through the scoring system, and updates the key component set table.

[0116] The life cycle management module 40 is responsible for performing life cycle analysis on the hydraulic gate hoist and its key components, evaluating the health status and life cycle cost of the components, and formulating and optimizing maintenance decisions based on fault prediction results and cost analysis.

[0117] The data acquisition and preprocessing module 10 collects real-time data and passes it to the fault mode recognition module 20 for data processing, identifies the fault mode, and passes the result to the fault location module 30. The fault location module 30 determines the fault propagation path and fault source, updates the key component information, and passes it to the full life cycle management module 40. The full life cycle management module 40 formulates a maintenance strategy based on all the collected information and feeds back the maintenance decision to the entire system to optimize the operation and maintenance process of the hydraulic gate hoist.

[0118] 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 the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A hydraulic gate hoist fault diagnosis method based on multimodal data fusion, characterized in that: include, Install sensors to collect data, amplify the collected data and identify abnormal sounds in real time for data classification; According to the classified data, the historical status and current status of the hydraulic system are correlated to determine the system failure mode; Through the correlation between hydraulic system components, analyze the fault propagation path, locate the fault source, and set the key components based on historical records and expert database; Establish full life cycle management for hydraulic gate hoists and key components.

2. The hydraulic hoist fault diagnosis method based on multimodal data fusion according to claim 1 is characterized in that: The data acquisition includes, the installed sensors include sound sensors, pressure sensors, temperature sensors and condition monitoring sensors; The scope of data collection is: full hydraulic hoist system, hydraulic cylinder, plunger pump, proportional pump, proportional valve, reversing valve, overflow valve, solenoid valve, filter, oil, oil temperature, and hoist on-site control cabinet.

3. The hydraulic hoist fault diagnosis method based on multimodal data fusion according to claim 2 is characterized in that: The data classification includes extracting key features from each type of collected data, constructing a feature matrix for data amplification, merging the enhanced data into the original data set to form a feature set, extracting the feature set of sound, identifying the data in real time in turn, modeling the features using a multi-layer feedforward network, determining data classification, and inputting suspected fault data into the hydraulic system.

4. The hydraulic hoist fault diagnosis method based on multimodal data fusion according to claim 3 is characterized in that: The system failure mode determination includes building an association relationship between the historical state and the current state, matching the features of the current and historical states, and building an association matrix. The system failure mode determination includes leakage failure, stuck failure, system overload, and control failure.

5. The hydraulic hoist fault diagnosis method based on multimodal data fusion according to claim 4 is characterized in that: The fault propagation path includes representing each component in the hydraulic system as a node V i , make correlation judgments between nodes and build association relationships between hydraulic system components; Extract the fault propagation path P through the depth-first search of the graph Path : P Path ={V i ,V j ,V n },Weighted ij >th path ·C path Among them, P Path is a qualified fault propagation path, V i 、V j 、V n is the node passed during the fault propagation process, th path The threshold used to judge the importance of the transmission path, C path To adjust the propagation restriction factor affecting the standard, Weighted ij For component V i and V j The weighted edges between .

6. The hydraulic hoist fault diagnosis method based on multimodal data fusion according to claim 5 is characterized in that: The fault source location includes selecting the node with the maximum score Q according to the weight and influence degree of each node in the propagation path. i The node is the fault source Q source = argmax i (Q i ), combining historical records and expert database to set key components, where Q i For node V i Fault source score, Q source To determine the fault source location, argmax i (Q i ) is to maximize Q i The node index of , indicating the fault source.

7. The hydraulic hoist fault diagnosis method based on multimodal data fusion according to claim 6 is characterized in that: The full life cycle management includes life cycle analysis of each key component of the hydraulic gate hoist, identification of risks and maintenance strategies under different failure modes, evaluation of the health status of key components with data collected during the operation cycle, quantification of operation based on historical data, and formulation and optimization of maintenance decisions based on failure prediction results and life cycle cost analysis.

8. A system using the hydraulic hoist fault diagnosis method based on multimodal data fusion as claimed in any one of claims 1 to 7, characterized in that: It includes data acquisition and preprocessing module, fault mode identification module, fault location module and full life cycle management module; The data acquisition and preprocessing module is responsible for installing and configuring various sensors to collect multimodal data of the hydraulic hoist system. The collected data is amplified and processed, and abnormal sounds are identified in real time to perform preliminary data classification; The fault mode recognition module amplifies and extracts features by constructing a feature matrix, models features using a multi-layer feedforward network, and calculates the total error to update the threshold. The module constructs a correlation matrix through feature matching and historical state correlation analysis, and calculates the optimal matching path to determine the system fault mode; The fault location module represents hydraulic system components as nodes, constructs association relationships and weighted graphs between nodes, determines how the fault propagates throughout the system by calculating the fault propagation path, locates the fault source through a scoring system, and updates the key component set table; The full life cycle management module is responsible for performing life cycle analysis on the hydraulic gate hoist and its key components, evaluating the health status and life cycle cost of the components, and formulating and optimizing maintenance decisions based on fault prediction results and cost analysis.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the hydraulic gate hoist fault diagnosis method based on multimodal data fusion described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the hydraulic gate hoist fault diagnosis method based on multimodal data fusion described in any one of claims 1 to 7 are implemented.

Images