A crane monitoring system and a prediction-based crane maintenance method

Abstract

This invention proposes a crane monitoring system and a prediction-based crane maintenance method. By acquiring first structural data and first vibration data of a target structural component, the system determines whether the first structural data exhibits damage characteristics. If no damage characteristics are found, it predicts second structural data and second vibration data for a future period. A correspondence analysis is performed on the second structural data and second vibration data to determine if they satisfy a preset correspondence. If the second structural data and second vibration data satisfy the preset correspondence, the system determines that the target structural component has a damage risk. The damage risk level of the target structural component is calculated based on the second vibration data, and a maintenance strategy for the target structural component is generated based on the damage risk level. This system can accurately predict physical damage to the crane and provide effective maintenance strategies.

Description

Technical Field

[0001] This invention relates to the field of crane maintenance technology, and in particular to a crane monitoring system and a prediction-based crane maintenance method. Background Technology

[0002] A crane, also known as a hoist, is a transport auxiliary device capable of moving heavy objects horizontally or vertically from one location to another within a certain range. It is widely used in ports, factory workshops, construction sites, and other situations requiring frequent handling of heavy objects. As a large piece of machinery, cranes require regular maintenance to ensure their safety during operation. To achieve heavy-duty handling capabilities, cranes are typically designed as large, complex machines composed of large mechanical structures and corresponding electrical control systems. Because cranes often operate in harsh environments, such as high-temperature and humid indoor and outdoor environments, coupled with frequent changes in load weight, various mechanical and electrical components of the crane can be damaged due to various internal and external factors. Without timely maintenance and repair, accidents can easily occur when transporting heavy objects, leading to serious consequences such as personal injury and property damage.

[0003] Among the various components of a crane, the electrical system's operational and health status can usually be identified by real-time acquisition of electrical parameters from each component. Therefore, fault identification and routine maintenance plans for the crane's electrical system can be more clearly defined, making the maintenance plans more targeted and purposeful. However, maintaining the crane's mechanical system is far more difficult than maintaining its electrical system. This is because cranes have numerous mechanical components, each a crucial part of the crane's lifting function. Over long-term operation, these mechanical components may experience wear, aging, deformation, and breakage due to various internal and external factors. Some of this damage is subtle and difficult for maintenance personnel to observe during routine maintenance. However, the accumulation of damage over time can lead to physical or functional failure of the crane, and in severe cases, even major safety accidents. Summary of the Invention

[0004] Based on the above-mentioned problems, this invention proposes a crane monitoring system and a prediction-based crane maintenance method, which can accurately predict the physical damage of the crane and provide effective maintenance strategies.

[0005] In view of this, a first aspect of the present invention provides a crane monitoring system, comprising a structural monitoring sensor for monitoring structural changes of a target structural component of the crane, a vibration monitoring sensor for monitoring the structural stability of the target structural component of the crane, and a processing unit communicatively connected to the structural monitoring sensor and the vibration monitoring sensor, wherein the processing unit is configured to:

[0006] Acquire first structural data of the target structural component and first vibration data corresponding to the first structural data, wherein the first structural data includes internal first structural data, shape monitoring data and / or surface monitoring data of the target structural component;

[0007] Damage feature identification is performed on the first structural data to determine whether damage features exist in the first structural data;

[0008] When the first structural data does not have damage characteristics, predict the second structural data and the second vibration data in the future period;

[0009] A correspondence analysis is performed on the second structural data and the second vibration data to determine whether the second structural data and the second vibration data satisfy a preset correspondence relationship;

[0010] When the second structural data and the second vibration data satisfy a preset correspondence, it is determined that the target structural component is at risk of damage.

[0011] The damage risk level of the target structural component is calculated based on the second vibration data;

[0012] A maintenance strategy for the target structural component is generated based on the damage risk level.

[0013] A second aspect of the present invention provides a prediction-based crane maintenance method, comprising:

[0014] Acquire first structural data of the target structural component and first vibration data corresponding to the first structural data, wherein the first structural data includes internal first structural data, shape monitoring data and / or surface monitoring data of the target structural component;

[0015] Damage feature identification is performed on the first structural data to determine whether damage features exist in the first structural data;

[0016] When the first structural data does not have damage characteristics, predict the second structural data and the second vibration data in the future period;

[0017] A correspondence analysis is performed on the second structural data and the second vibration data to determine whether the second structural data and the second vibration data satisfy a preset correspondence relationship;

[0018] When the second structural data and the second vibration data satisfy a preset correspondence, it is determined that the target structural component is at risk of damage.

[0019] The damage risk level of the target structural component is calculated based on the second vibration data;

[0020] A maintenance strategy for the target structural component is generated based on the damage risk level.

[0021] Furthermore, the first structural data is a structural image of the target structural component. The step of performing damage feature identification on the first structural data to determine whether damage features exist in the first structural data specifically includes:

[0022] Determine whether there are structural anomalies in the structural image;

[0023] When structural anomalies exist in the structural image, a local structural image of the target structural component at the structural anomaly point is generated.

[0024] The local structural image is input into a pre-trained damage recognition model to identify damage features, in order to determine whether the first structural data has damage features.

[0025] Furthermore, the step of determining whether there are structural anomalies in the structural image specifically includes:

[0026] Configure a standard image corresponding to the target structural component and the structural image, wherein the standard image is a structural image generated of the target structural component in a damage-free state;

[0027] Obtain the first structural image sequence of the target structural component over a past period of time;

[0028] A first difference feature sequence is generated based on the difference features between each structural image in the first structural image sequence and the standard image;

[0029] Generate a first differential feature size parameter sequence corresponding to the first differential feature sequence. The first differential feature size parameter sequence is composed of one or more size parameters in different directions for each differential feature in the first differential feature sequence.

[0030] A trend analysis is performed on the first differential feature size parameter sequence to determine whether the corresponding differential feature is a structural anomaly based on the trend analysis results.

[0031] Furthermore, the step of performing trend analysis on the first differential feature size parameter sequence to determine whether the corresponding differential feature is a structural anomaly based on the trend analysis results specifically includes:

[0032] Obtain each size parameter p1 in the first differential feature size parameter sequence. i (t), where i∈[1,n] p ], where i is a positive integer and n p The number of size parameter types of the differential features, wherein the first differential feature size parameter sequence is a discrete data sequence constructed in order of the generation time t of the first structural data structure image;

[0033] For the i-th size parameter, fit the size parameter p1 with time t as the independent variable. i The trend line function f(t) i (t)=k i ×t+b i ;

[0034] Determine the trend line function f i The slope k in (t) i Is it greater than 0?

[0035] In [1, n] p Within the range, when there exists any value of i that satisfies k i When the value is greater than 0, it is determined that there are structural anomalies in the structural image.

[0036] Furthermore, the step of performing a correspondence analysis on the second structural data and the second vibration data to determine whether the second structural data and the second vibration data satisfy a preset correspondence relationship specifically includes:

[0037] Generate a first trend feature of the second structural data and a second trend feature of the second vibration data;

[0038] Calculate the matching degree between the first trend feature and the second trend feature;

[0039] Determine whether the matching degree between the first trend feature and the second trend feature is greater than a preset matching degree threshold;

[0040] When the matching degree between the first trend feature and the second trend feature is greater than a preset matching degree threshold, it is determined that the second structural data and the second vibration data satisfy a preset correspondence.

[0041] Furthermore, the step of generating the first trend feature of the second structural data specifically includes:

[0042] Generate a second differential feature size parameter sequence for structural anomalies in the second structural data;

[0043] Obtain each size parameter p2 from the second differential feature size parameter sequence. i (t), where i∈[1,n] p ], where i is a positive integer and n p The number of size parameter types of the differential features, wherein the second differential feature size parameter sequence is a discrete data sequence constructed in order of the generation time t of the structure image of the second structure data;

[0044] For the i-th size parameter, fit the size parameter p2 with time t as the independent variable. i The first change curve function C1 of (t) i (t);

[0045] Extracting the first change curve function C1 i The set of time ranges corresponding to the rising interval of (t) (ts) ij ,te ij ), where j∈[1, np i ], np i The first changing curve function C1 i The number of rising intervals in (t), ts ij The first changing curve function C1 i The start time of the j-th ascending interval in (t), te ij The first changing curve function C1 i The end time of the j-th ascending interval in (t);

[0046] The time range set (ts) ij ,te ij This is identified as the first trend feature of the second structural data.

[0047] Furthermore, the step of generating the second trend feature of the second vibration data specifically includes:

[0048] Obtain the vibration intensity v(t) from the second vibration data;

[0049] The second variation curve function C2(t) of the vibration intensity v(t) is fitted with time t as the independent variable;

[0050] Extract the set of rising inflection points of the second change curve function C2(t) tip k , where k∈[1, n v ], n v The number of rising inflection points in the second change curve function C2(t), wherein the rising inflection point is the inflection point in the second change curve function C2(t) where the slope increases;

[0051] The set of rising inflection points tipk The second trend feature of the second vibration data was determined.

[0052] Furthermore, the step of calculating the matching degree between the first trend feature and the second trend feature specifically includes:

[0053] Determine the number of structural anomalies on the target structural component;

[0054] When the number of structural anomalies on the target structural component is 1, obtain the np corresponding to the structural anomaly in the second structural data. i A time range set (ts) ij ,te ij );

[0055] For each time range set (ts) ij ,te ij Based on the set of rising inflection points of the second vibration data, tip k The number of items falling within its time range is used to calculate its matching score s. i ;

[0056] np i The maximum of the matching scores The target matching degree P between the first trend feature and the second trend feature is determined.

[0057] Furthermore, each structural anomaly point has a corresponding first trend characteristic. Following the step of determining the number of structural anomaly points on the target structural component, the method further includes:

[0058] When the number of structural anomalies on the target structural component is greater than one, the distance d between each structural anomaly and the vibration monitoring point is determined. l , where l∈[1, n un ], n un The number of structural anomalies on the target structural component;

[0059] Based on the distance d between each structural anomaly point and the vibration monitoring point l Configure the influence weight σ of each structural anomaly on the matching degree between the first trend feature and the second trend feature. l ;

[0060] Calculate the matching degree p between the first trend feature and the second trend feature for each structural anomaly. l ;

[0061] Based on the influence weight σ of each structural anomaly point l The degree of matching p between its first trend feature and its second trend feature lCalculate the target matching degree used to determine whether the second structural data and the second vibration data satisfy a preset correspondence:

[0062]

[0063] This invention proposes a crane monitoring system and a prediction-based crane maintenance method. By acquiring first structural data and first vibration data of a target structural component, the system determines whether the first structural data exhibits damage characteristics. If no damage characteristics are found, it predicts second structural data and second vibration data for a future period. A correspondence analysis is performed on the second structural data and second vibration data to determine if they satisfy a preset correspondence. If the second structural data and second vibration data satisfy the preset correspondence, the system determines that the target structural component has a damage risk. The damage risk level of the target structural component is calculated based on the second vibration data, and a maintenance strategy for the target structural component is generated based on the damage risk level. This system can accurately predict physical damage to the crane and provide effective maintenance strategies. Attached Figure Description

[0064] Figure 1 This is a schematic diagram of a crane monitoring system provided in one embodiment of the present invention;

[0065] Figure 2 This is a flowchart of a prediction-based crane maintenance method provided in one embodiment of the present invention. Detailed Implementation

[0066] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0067] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0068] In the description of this invention, the term "multiple" refers to two or more. Unless otherwise explicitly defined, the terms "upper," "lower," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. The terms "connect," "install," "fix," etc., should be interpreted broadly. For example, "connect" can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "multiple" means two or more.

[0069] In the description of this specification, the terms "one embodiment," "some implementations," "specific embodiment," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0070] A crane monitoring system and a prediction-based crane maintenance method according to some embodiments of the present invention will now be described with reference to the accompanying drawings.

[0071] like Figure 1 As shown, it includes a structural monitoring sensor for monitoring structural changes of the target structural component of the crane, a vibration monitoring sensor for monitoring the structural stability of the target structural component of the crane, and a processing unit communicatively connected to the structural monitoring sensor and the vibration monitoring sensor.

[0072] Specifically, the mechanical system of a crane consists of several parts, such as the boom, chassis, outriggers, and operator's cab, each of which is composed of several structural components. In the technical solution of this invention, easily damaged structural components on the crane, such as the main beam, coupling, hook, and wire rope, are identified as target structural components. Structural monitoring sensors and vibration monitoring sensors are installed on these target structural components to monitor their structure and vibration, and maintenance strategies for the crane are formulated based on the monitoring data.

[0073] In some embodiments of the present invention, the structural monitoring sensor may include an ultrasonic sensor for internal flaw detection of the target structural component, and an image sensor for external monitoring of the target structural component (including surface crack inspection, overall or local deformation inspection, etc.).

[0074] The crane monitoring system also includes a communication unit, which can be a wired communication unit or a wireless communication unit. The processor establishes a communication connection with the structural monitoring sensor and the vibration monitoring sensor through the communication unit.

[0075] In the technical solution of the present invention, the processing unit is configured as follows:

[0076] Acquire first structural data of the target structural component and first vibration data corresponding to the first structural data, wherein the first structural data includes internal first structural data, shape monitoring data and / or surface monitoring data of the target structural component;

[0077] Damage feature identification is performed on the first structural data to determine whether damage features exist in the first structural data;

[0078] When the first structural data does not have damage characteristics, predict the second structural data and the second vibration data in the future period;

[0079] A correspondence analysis is performed on the second structural data and the second vibration data to determine whether the second structural data and the second vibration data satisfy a preset correspondence relationship;

[0080] When the second structural data and the second vibration data satisfy a preset correspondence, it is determined that the target structural component is at risk of damage.

[0081] The damage risk level of the target structural component is calculated based on the second vibration data;

[0082] A maintenance strategy for the target structural component is generated based on the damage risk level.

[0083] In the technical solution of this invention, both the first structural data and the first vibration data are monitoring data collected by structural monitoring sensors and vibration monitoring sensors installed on the target structural component to monitor the target structural component. The first structural data and the first vibration data have a temporal correspondence, that is, the monitoring data collected at each data sampling time point includes at least one set of first structural data and one set of first vibration data for the target structural component. Therefore, the first vibration data corresponding to the first structural data refers to the first vibration data that corresponds to the first structural data in terms of the sampling time for the same target structural component.

[0084] In the first structural data, the internal first structural data is the internal structural data of the target structural component obtained by detection using ultrasound, X-rays, gamma rays, etc. It can be a single internal structural image corresponding to a specific view of the target structural component, or several internal structural images corresponding to different viewpoints of the target structural component. The shape monitoring data and the surface monitoring data are the appearance data of the target structural component obtained by image sensors such as cameras. They include one or more appearance images reflecting the overall shape and size of the target structural component, and surface images of one or more specific surface areas. The specific surface areas can be several pre-configured vulnerable areas.

[0085] On the structural components of the crane, various types of damage, such as wear, aging, deformation, and fracture, manifest as different damage characteristics in the first structural data. These damage characteristics are the quantitative features of various damage types in the first structural data.

[0086] The second structural data is the structural data of the target structural component over a future period, predicted using traditional prediction algorithms or artificial intelligence prediction models based on the first structural data. Similarly, the second vibration data is the vibration data of the target structural component over a future period, predicted using traditional prediction algorithms or artificial intelligence prediction models based on the first vibration data. Preferably, both the second structural data and the second vibration data are data predicted using a pre-trained artificial intelligence prediction model. The artificial intelligence prediction model is trained using machine learning techniques, with historical structural data and historical vibration data of the target structural component collected by the structural monitoring sensor and the vibration monitoring sensor as sample data.

[0087] Furthermore, in the step of calculating the damage risk level of the target structural member based on the second vibration data, the processing unit is configured to:

[0088] The standard vibration amplitude of the target structural component is obtained, which is the maximum vibration amplitude measured in the working state of the target structural component when it has no physical damage.

[0089] The damage vibration amplitude of the target structural component in the second vibration data is obtained, wherein the damage vibration amplitude is the maximum vibration amplitude of the target structural component in the second vibration data;

[0090] Calculate the difference between the damaged vibration amplitude and the standard vibration amplitude;

[0091] The damage risk level of the target structural component is determined based on the difference between the damaged vibration amplitude and the standard vibration amplitude.

[0092] In the above-described embodiment, a mapping relationship between the difference between the damaged vibration amplitude and the standard vibration amplitude and the damage risk level is pre-configured in the database. After determining the difference between the damaged vibration amplitude and the standard vibration amplitude, the damage risk level of the target structural component is queried from the database.

[0093] In some embodiments of the present invention, maintenance strategies for each target structural component under different damage risk levels are pre-configured in a database. In the step of generating a maintenance strategy for the target structural component based on the damage risk level, the corresponding maintenance strategy is queried from the database according to the type of the target structural component and its damage risk level. Furthermore, in the step of predicting second structural data and second vibration data over a future period, the processing unit is configured to:

[0094] Obtain a pre-configured first time length and a second time length, wherein the first time length is the time length corresponding to the input data of the artificial intelligence prediction model, and the second time length is the time length corresponding to the second structural data and the second vibration data to be predicted, that is, the time length corresponding to the output data of the artificial intelligence prediction model;

[0095] The time when the first structural data of the target structural component and the first vibration data corresponding to the first structural data are acquired is determined as the current time.

[0096] Determine a first time range with the first time length ending at the current time, and a second time range with the second time length starting at the current time;

[0097] The first structural data collected within the first time range is organized into the first input data sequence of the structural prediction model, and the first vibration data collected within the first time range is organized into the second input data sequence of the vibration prediction model.

[0098] The first input data sequence is input into the structure prediction model to predict the structure data within the second time range, which is then used as the second structure data.

[0099] The second input data sequence is input into the vibration prediction model to predict the vibration data within the second time range, which is then used as the second vibration data.

[0100] In the technical solution of the above embodiments, the so-called future period of time is a second time range with the second time length starting from the current time.

[0101] The structural prediction model is an artificial intelligence prediction model trained using historical structural data of the target structural component as sample data, and the vibration prediction model is an artificial intelligence prediction model trained using historical vibration data of the target structural component as sample data.

[0102] In some embodiments of the present invention, the first time length and the second time length can be the same.

[0103] Furthermore, the first structural data is a structural image of the target structural component. In the step of performing damage feature identification on the first structural data to determine whether damage features exist in the first structural data, the processing unit is configured to:

[0104] Determine whether there are structural anomalies in the structural image;

[0105] When structural anomalies exist in the structural image, a local structural image of the target structural component at the structural anomaly point is generated.

[0106] The local structural image is input into a pre-trained damage recognition model to identify damage features, in order to determine whether the first structural data has damage features.

[0107] Based on image content, the structural image can be one or more of the following: internal structure image, external appearance image, and surface image of the target structural component. Based on image generation method, the structural image can be one or more of the following: ultrasonic image, X-ray image, gamma-ray image, infrared image, and visible light image.

[0108] The local structural image is a partial image of the structural image at the structural anomaly point. Further, the step of generating the local structural image of the target structural component at the structural anomaly point specifically includes:

[0109] Obtain the standard size of the input image for the pre-configured damage recognition model;

[0110] Determine the center point and maximum width of the structural anomaly point in the structural image, where the center point may be the geometric center of the structural anomaly point;

[0111] A square partial image centered at the center point is extracted from the structural image, using the maximum width as the side length.

[0112] The square-open local image is scaled to the standard size to obtain a local structural image of the target structural component at the structural anomaly point.

[0113] The damage identification model is a pre-trained deep learning model; more specifically, it is a damage type classification model trained using deep learning techniques. By collecting and organizing local structural images of the target structural component under various damage conditions with standard dimensions, and labeling each local structural image with a damage type to construct training sample data, deep learning is performed using these training sample data to obtain the damage identification model. In specific implementation scenarios, independent damage identification models can be trained separately for structural components of different materials on the crane, or for different damage types, to improve the accuracy of damage identification.

[0114] Furthermore, in the step of determining whether there are structural anomalies in the structural image, the processing unit is configured to:

[0115] Configure a standard image corresponding to the target structural component and the structural image, wherein the standard image is a structural image generated of the target structural component in a damage-free state;

[0116] Obtain the first structural image sequence of the target structural component over a past period of time;

[0117] A first difference feature sequence is generated based on the difference features between each structural image in the first structural image sequence and the standard image;

[0118] Generate a first differential feature size parameter sequence corresponding to the first differential feature sequence. The first differential feature size parameter sequence is composed of one or more size parameters in different directions for each differential feature in the first differential feature sequence.

[0119] A trend analysis is performed on the first differential feature size parameter sequence to determine whether the corresponding differential feature is a structural anomaly based on the trend analysis results.

[0120] The standard image and the structured image have the same image generation method, shooting angle, shooting distance and other shooting parameters.

[0121] The term "past period" refers to a connected time range with the current time as the end point and possessing the third time length. The third time length is a pre-configured specific time length used for performing structural anomaly identification.

[0122] Furthermore, in the step of generating a first difference feature sequence based on the difference features between each structural image in the first structural image sequence and the standard image, the processing unit is configured to:

[0123] The location where there is a significant data difference between the structured image and the standard image is determined as the difference point. The significant data difference can be the case where the difference between the two is greater than a pre-configured threshold.

[0124] Analyze and identify the presence of each difference point on the structural image in the first structural image sequence;

[0125] When a difference point persists in several consecutive structural images in the first structural image sequence, the difference region where the difference point is located is determined as a difference feature, and the difference region is a local image region formed by consecutive difference points in the structural image.

[0126] The difference features at the same location in the first structural image sequence are constructed into the first difference feature sequence in chronological order.

[0127] The size parameters of the differential feature can be one or more of the shape parameters such as width, height, area (pixel area), perimeter, and circumscribed circle radius of the differential feature.

[0128] Furthermore, in the step of performing trend analysis on the first differential feature size parameter sequence to determine whether the corresponding differential feature is a structural anomaly point based on the trend analysis result, the processing unit is configured as follows:

[0129] Obtain each size parameter p1 in the first differential feature size parameter sequence. i (t), where i∈[1,n] p ], where i is a positive integer and n p The number of size parameter types of the differential features, wherein the first differential feature size parameter sequence is a discrete data sequence constructed in order of the generation time t of the first structural data structure image;

[0130] For the i-th size parameter, fit the size parameter p1 with time t as the independent variable. i The trend line function f(t) i (t)=k i ×t+b i;

[0131] Determine the trend line function f i The slope k in (t) i Is it greater than 0?

[0132] In [1, n] p Within the range, when there exists any value of i that satisfies k i When the value is greater than 0, it is determined that there are structural anomalies in the structural image.

[0133] Specifically, when n p When the value is greater than 1, the first differential feature size parameter sequence contains n p There are several parallel subsequences, each corresponding to a time series with a specific dimensional parameter such as width, height, and area. Therefore, p1 i (t) represents the size parameter of the differential feature in the structural image generated at time t in the i-th subsequence of the first differential feature size parameter sequence, or p1 i (t) represents the size of the i-th size parameter of the differential feature in the structural image generated at time t.

[0134] In the technical solution of the above embodiment, f i (t) represents the size parameter p1 obtained by fitting using the least squares linear fitting algorithm. i The trend line function f(t) is a linear function representing the trend line function f. i (t) can represent the size parameter p1 i (t) The trend of change over a period of time.

[0135] Furthermore, in the step of performing a correspondence analysis on the second structural data and the second vibration data to determine whether the second structural data and the second vibration data satisfy a preset correspondence relationship, the processing unit is configured as follows:

[0136] Generate a first trend feature of the second structural data and a second trend feature of the second vibration data;

[0137] Calculate the matching degree between the first trend feature and the second trend feature;

[0138] Determine whether the matching degree between the first trend feature and the second trend feature is greater than a preset matching degree threshold;

[0139] When the matching degree between the first trend feature and the second trend feature is greater than a preset matching degree threshold, it is determined that the second structural data and the second vibration data satisfy a preset correspondence.

[0140] In the technical solution of the above embodiments, the first trend feature is a feature reflecting the changing trend of the dimensional parameters of one or more structural anomalies in the second structural data. Similarly, the second trend feature is a feature reflecting the changing trend of the second vibration data.

[0141] Furthermore, prior to the step of generating the first trend feature of the second structural data and the second trend feature of the second vibration data, the processing unit is configured to:

[0142] Determine the number of structural outliers in the second structural data;

[0143] The matching degree threshold is configured based on the number of structural anomalies determined in the second structural data.

[0144] In the technical solution of the above embodiments, the matching degree threshold is not a single fixed threshold. The size of the matching degree threshold is related to the number of structural anomalies in the second structural data. For second structural data with different numbers of structural anomalies, the matching degree threshold for matching with the second vibration data is different.

[0145] Furthermore, in the step of generating the first trend feature of the second structural data, the processing unit is configured to:

[0146] Generate a second differential feature size parameter sequence for structural anomalies in the second structural data;

[0147] Obtain each size parameter p2 from the second differential feature size parameter sequence. i (t), where i∈[1,n] p ], where i is a positive integer and n p The number of size parameter types of the differential features, wherein the second differential feature size parameter sequence is a discrete data sequence constructed in order of the generation time t of the structure image of the second structure data;

[0148] For the i-th size parameter, fit the size parameter p2 with time t as the independent variable. i The first change curve function C1 of (t) i (t);

[0149] Extracting the first change curve function C1 i The set of time ranges corresponding to the rising interval of (t) (ts) ij ,te ij ), where j∈[1, np i ], np i The first changing curve function C1 iThe number of rising intervals in (t), ts ij The first changing curve function C1 i The start time of the j-th ascending interval in (t), te ij The first changing curve function C1 i The end time of the j-th ascending interval in (t);

[0150] The time range set (ts) ij ,te ij This is identified as the first trend feature of the second structural data.

[0151] In some embodiments of the present invention, the first change curve function C1 i (t) represents the size parameter p2 obtained by fitting using a polynomial fitting and curve fitting algorithm. i The first curve function C1 is a curve function showing how the magnitude of (t) changes over time. i (t) is used to represent the size parameter p2 i The magnitude of (t) changes over time in the future.

[0152] The first change curve function C1 i The rising interval of (t) refers to the first changing curve function C1 i The interval with a slope greater than 0 in (t), i.e., ts ij and te ij All are the first change curve function C1 i The moment when the slope of (t) is 0.

[0153] Furthermore, in the step of generating a second differential feature size parameter sequence of structural anomalies in the second structural data when the second structural data is a structural image of the target structural component, the processing unit is configured to:

[0154] Based on the first structural data, determine the target location of the structural anomaly point in the target structural component;

[0155] Generate a sequence of second structural images of the target structural component over a future period based on the second structural data;

[0156] Generate a second differential feature size parameter sequence corresponding to the target location in the second structural image sequence.

[0157] Furthermore, in the step of generating the second trend feature of the second vibration data, the processing unit is configured to:

[0158] Obtain the vibration intensity v(t) from the second vibration data;

[0159] The second variation curve function C2(t) of the vibration intensity v(t) is fitted with time t as the independent variable;

[0160] Extract the set of rising inflection points of the second change curve function C2(t) tip k , where k∈[1, n v ], n v The number of rising inflection points in the second change curve function C2(t), wherein the rising inflection point is the inflection point in the second change curve function C2(t) where the slope increases;

[0161] The set of rising inflection points tip k The second trend feature of the second vibration data was determined.

[0162] The vibration intensity v(t) is the vibration intensity of the target structure detected by the vibration monitoring sensor at time t. The second variation curve function C2(t) is a curve function of the magnitude of the vibration intensity v(t) changing with time, obtained by using curve fitting algorithms such as polynomial fitting. The second variation curve function C2(t) is used to represent the change of the magnitude of the vibration intensity v(t) with time over a future period.

[0163] Furthermore, the set of rising inflection points of the second change curve function C2(t) is extracted. k In the steps described above, the processing unit is configured as follows:

[0164] Identify each inflection point in the second change curve function C2(t);

[0165] Calculate the first slope at Δt before each inflection point and the second slope at Δt after each inflection point, where Δt is a pre-configured slope calculation interval;

[0166] The inflection point where the second slope is greater than the first slope is determined as the rising inflection point.

[0167] Furthermore, in the step of calculating the matching degree between the first trend feature and the second trend feature, the processing unit is configured to:

[0168] Determine the number of structural anomalies on the target structural component;

[0169] When the number of structural anomalies on the target structural component is 1, obtain the np corresponding to the structural anomaly in the second structural data. i A time range set (ts) ij ,te ij );

[0170] For each time range set (ts) ij ,te ij Based on the set of rising inflection points of the second vibration data, tip k The number of items falling within its time range is used to calculate its matching score s. i ;

[0171] np i The maximum of the matching scores The target matching degree P between the first trend feature and the second trend feature is determined.

[0172] In the technical solution of the above embodiments, the target matching degree P is a matching degree used to determine whether the second structural data and the second vibration data satisfy a preset correspondence. That is, in the step of determining whether the matching degree between the first trend feature and the second trend feature is greater than a preset matching degree threshold, specifically, it is determined whether the target matching degree P is greater than the preset matching degree threshold.

[0173] Furthermore, for each time range set (ts) ij ,te ij Based on the set of rising inflection points of the second vibration data, tip k The number of items falling within its time range is used to calculate its matching score s. i In the steps described above, the processing unit is configured as follows:

[0174] Statistical analysis of the set of rising inflection points of the second vibration data tip k The number of rising inflection points falling into the i-th time range set, np i ;

[0175] The set of rising inflection points of the second vibration data tip k The number of rising inflection points that fall into the i-th time range set, np i The ratio of the total number of rising inflection points (nv) relative to the set of rising inflection points in the second vibration data is used to determine its matching score:

[0176]

[0177] Furthermore, each structural anomaly point has a corresponding first trend characteristic. After determining the number of structural anomalies on the target structural component, the processing unit is configured to:

[0178] When the number of structural anomalies on the target structural component is greater than one, the distance d between each structural anomaly and the vibration monitoring point is determined. l , where l∈[1, n un ], n unThe number of structural anomalies on the target structural component;

[0179] Based on the distance d between each structural anomaly point and the vibration monitoring point l Configure the influence weight σ of each structural anomaly on the matching degree between the first trend feature and the second trend feature. l ;

[0180] Calculate the matching degree p between the first trend feature and the second trend feature for each structural anomaly. l ;

[0181] Based on the influence weight σ of each structural anomaly point l The degree of matching p between its first trend feature and its second trend feature l Calculate the target matching degree used to determine whether the second structural data and the second vibration data satisfy a preset correspondence:

[0182]

[0183] Specifically, the vibration monitoring point is the location on the target structural component where the vibration monitoring sensor is installed. The greater the distance from the vibration monitoring point, the smaller the weight of the influence of the structural anomaly on the matching degree of the first trend feature and the second trend feature.

[0184] like Figure 2 As shown, a second aspect of the present invention proposes a prediction-based crane maintenance method, comprising:

[0185] Acquire first structural data of the target structural component and first vibration data corresponding to the first structural data, wherein the first structural data includes internal first structural data, shape monitoring data and / or surface monitoring data of the target structural component;

[0186] Damage feature identification is performed on the first structural data to determine whether damage features exist in the first structural data;

[0187] When the first structural data does not have damage characteristics, predict the second structural data and the second vibration data in the future period;

[0188] A correspondence analysis is performed on the second structural data and the second vibration data to determine whether the second structural data and the second vibration data satisfy a preset correspondence relationship;

[0189] When the second structural data and the second vibration data satisfy a preset correspondence, it is determined that the target structural component is at risk of damage.

[0190] The damage risk level of the target structural component is calculated based on the second vibration data;

[0191] A maintenance strategy for the target structural component is generated based on the damage risk level.

[0192] In the technical solution of this invention, both the first structural data and the first vibration data are monitoring data collected by structural monitoring sensors and vibration monitoring sensors installed on the target structural component to monitor the target structural component. The first structural data and the first vibration data have a temporal correspondence, that is, the monitoring data collected at each data sampling time point includes at least one set of first structural data and one set of first vibration data for the target structural component. Therefore, the first vibration data corresponding to the first structural data refers to the first vibration data that corresponds to the first structural data in terms of the sampling time for the same target structural component.

[0193] In the first structural data, the internal first structural data is the internal structural data of the target structural component obtained by detection using ultrasound, X-rays, gamma rays, etc. It can be a single internal structural image corresponding to a specific view of the target structural component, or several internal structural images corresponding to different viewpoints of the target structural component. The shape monitoring data and the surface monitoring data are the appearance data of the target structural component obtained by image sensors such as cameras. They include one or more appearance images reflecting the overall shape and size of the target structural component, and surface images of one or more specific surface areas. The specific surface areas can be several pre-configured vulnerable areas.

[0194] On the structural components of the crane, various types of damage, such as wear, aging, deformation, and fracture, manifest as different damage characteristics in the first structural data. These damage characteristics are the quantitative features of various damage types in the first structural data.

[0195] The second structural data is the structural data of the target structural component over a future period, predicted using traditional prediction algorithms or artificial intelligence prediction models based on the first structural data. Similarly, the second vibration data is the vibration data of the target structural component over a future period, predicted using traditional prediction algorithms or artificial intelligence prediction models based on the first vibration data. Preferably, both the second structural data and the second vibration data are data predicted using a pre-trained artificial intelligence prediction model. The artificial intelligence prediction model is trained using machine learning techniques, with historical structural data and historical vibration data of the target structural component collected by the structural monitoring sensor and the vibration monitoring sensor as sample data.

[0196] Furthermore, the step of calculating the damage risk level of the target structural component based on the second vibration data specifically includes:

[0197] The standard vibration amplitude of the target structural component is obtained, which is the maximum vibration amplitude measured in the working state of the target structural component when it has no physical damage.

[0198] The damage vibration amplitude of the target structural component in the second vibration data is obtained, wherein the damage vibration amplitude is the maximum vibration amplitude of the target structural component in the second vibration data;

[0199] Calculate the difference between the damaged vibration amplitude and the standard vibration amplitude;

[0200] The damage risk level of the target structural component is determined based on the difference between the damaged vibration amplitude and the standard vibration amplitude.

[0201] In the above-described embodiment, a mapping relationship between the difference between the damaged vibration amplitude and the standard vibration amplitude and the damage risk level is pre-configured in the database. After determining the difference between the damaged vibration amplitude and the standard vibration amplitude, the damage risk level of the target structural component is queried from the database.

[0202] In some embodiments of the present invention, maintenance strategies for each target structural component under different damage risk levels are pre-configured in the database. In the step of generating the maintenance strategy for the target structural component according to the damage risk level, the corresponding maintenance strategy is queried from the database according to the type of the target structural component and its damage risk level.

[0203] Furthermore, the specific steps for predicting the second structural data and the second vibration data over a future period include:

[0204] Obtain a pre-configured first time length and a second time length, wherein the first time length is the time length corresponding to the input data of the artificial intelligence prediction model, and the second time length is the time length corresponding to the second structural data and the second vibration data to be predicted, that is, the time length corresponding to the output data of the artificial intelligence prediction model;

[0205] The time when the first structural data of the target structural component and the first vibration data corresponding to the first structural data are acquired is determined as the current time.

[0206] Determine a first time range with the first time length ending at the current time, and a second time range with the second time length starting at the current time;

[0207] The first structural data collected within the first time range is organized into the first input data sequence of the structural prediction model, and the first vibration data collected within the first time range is organized into the second input data sequence of the vibration prediction model.

[0208] The first input data sequence is input into the structure prediction model to predict the structure data within the second time range, which is then used as the second structure data.

[0209] The second input data sequence is input into the vibration prediction model to predict the vibration data within the second time range, which is then used as the second vibration data.

[0210] In the technical solution of the above embodiments, the so-called future period of time is a second time range with the second time length starting from the current time.

[0211] The structural prediction model is an artificial intelligence prediction model trained using historical structural data of the target structural component as sample data, and the vibration prediction model is an artificial intelligence prediction model trained using historical vibration data of the target structural component as sample data.

[0212] In some embodiments of the present invention, the first time length and the second time length can be the same.

[0213] Furthermore, the first structural data is a structural image of the target structural component. The step of performing damage feature identification on the first structural data to determine whether damage features exist in the first structural data specifically includes:

[0214] Determine whether there are structural anomalies in the structural image;

[0215] When structural anomalies exist in the structural image, a local structural image of the target structural component at the structural anomaly point is generated.

[0216] The local structural image is input into a pre-trained damage recognition model to identify damage features, in order to determine whether the first structural data has damage features.

[0217] Based on image content, the structural image can be one or more of the following: internal structure image, external appearance image, and surface image of the target structural component. Based on image generation method, the structural image can be one or more of the following: ultrasonic image, X-ray image, gamma-ray image, infrared image, and visible light image.

[0218] The local structural image is a partial image of the structural image at the structural anomaly point. Further, the step of generating the local structural image of the target structural component at the structural anomaly point specifically includes:

[0219] Obtain the standard size of the input image for the pre-configured damage recognition model;

[0220] Determine the center point and maximum width of the structural anomaly point in the structural image, where the center point may be the geometric center of the structural anomaly point;

[0221] A square partial image centered at the center point is extracted from the structural image, using the maximum width as the side length.

[0222] The square-open local image is scaled to the standard size to obtain a local structural image of the target structural component at the structural anomaly point.

[0223] The damage identification model is a pre-trained deep learning model; more specifically, it is a damage type classification model trained using deep learning techniques. By collecting and organizing local structural images of the target structural component under various damage conditions with standard dimensions, and labeling each local structural image with a damage type to construct training sample data, deep learning is performed using these training sample data to obtain the damage identification model. In specific implementation scenarios, independent damage identification models can be trained separately for structural components of different materials on the crane, or for different damage types, to improve the accuracy of damage identification.

[0224] Furthermore, the step of determining whether there are structural anomalies in the structural image specifically includes:

[0225] Configure a standard image corresponding to the target structural component and the structural image, wherein the standard image is a structural image generated of the target structural component in a damage-free state;

[0226] Obtain the first structural image sequence of the target structural component over a past period of time;

[0227] A first difference feature sequence is generated based on the difference features between each structural image in the first structural image sequence and the standard image;

[0228] Generate a first differential feature size parameter sequence corresponding to the first differential feature sequence. The first differential feature size parameter sequence is composed of one or more size parameters in different directions for each differential feature in the first differential feature sequence.

[0229] A trend analysis is performed on the first differential feature size parameter sequence to determine whether the corresponding differential feature is a structural anomaly based on the trend analysis results.

[0230] The standard image and the structured image have the same image generation method, shooting angle, shooting distance and other shooting parameters.

[0231] The term "past period" refers to a connected time range with the current time as the end point and possessing the third time length. The third time length is a pre-configured specific time length used for performing structural anomaly identification.

[0232] Furthermore, the step of generating a first difference feature sequence based on the difference features between each structural image in the first structural image sequence and the standard image specifically includes:

[0233] The location where there is a significant data difference between the structured image and the standard image is determined as the difference point. The significant data difference can be the case where the difference between the two is greater than a pre-configured threshold.

[0234] Analyze and identify the presence of each difference point on the structural image in the first structural image sequence;

[0235] When a difference point persists in several consecutive structural images in the first structural image sequence, the difference region where the difference point is located is determined as a difference feature, and the difference region is a local image region formed by consecutive difference points in the structural image.

[0236] The difference features at the same location in the first structural image sequence are constructed into the first difference feature sequence in chronological order.

[0237] The size parameters of the differential feature can be one or more of the shape parameters such as width, height, area (pixel area), perimeter, and circumscribed circle radius of the differential feature.

[0238] Furthermore, the step of performing trend analysis on the first differential feature size parameter sequence to determine whether the corresponding differential feature is a structural anomaly based on the trend analysis results specifically includes:

[0239] Obtain each size parameter p1 in the first differential feature size parameter sequence. i (t), where i∈[1,n] p ], where i is a positive integer and n p The number of size parameter types of the differential features, wherein the first differential feature size parameter sequence is a discrete data sequence constructed in order of the generation time t of the first structural data structure image;

[0240] For the i-th size parameter, fit the size parameter p1 with time t as the independent variable. i The trend line function f(t) i (t)=k i ×t+b i ;

[0241] Determine the trend line function f i The slope k in (t) i Is it greater than 0?

[0242] In [1, n] p Within the range, when there exists any value of i that satisfies k i When the value is greater than 0, it is determined that there are structural anomalies in the structural image.

[0243] Specifically, when n p When the value is greater than 1, the first differential feature size parameter sequence contains n p There are several parallel subsequences, each corresponding to a time series with a specific dimensional parameter such as width, height, and area. Therefore, p1 i (t) represents the size parameter of the differential feature in the structural image generated at time t in the i-th subsequence of the first differential feature size parameter sequence, or p1 i (t) represents the size of the i-th size parameter of the differential feature in the structural image generated at time t.

[0244] In the technical solution of the above embodiment, f i (t) represents the size parameter p1 obtained by fitting using the least squares linear fitting algorithm. i The trend line function f(t) is a linear function representing the trend line function f. i (t) can represent the size parameter p1 i (t) The trend of change over a period of time.

[0245] Furthermore, the step of performing a correspondence analysis on the second structural data and the second vibration data to determine whether the second structural data and the second vibration data satisfy a preset correspondence relationship specifically includes:

[0246] Generate a first trend feature of the second structural data and a second trend feature of the second vibration data;

[0247] Calculate the matching degree between the first trend feature and the second trend feature;

[0248] Determine whether the matching degree between the first trend feature and the second trend feature is greater than a preset matching degree threshold;

[0249] When the matching degree between the first trend feature and the second trend feature is greater than a preset matching degree threshold, it is determined that the second structural data and the second vibration data satisfy a preset correspondence.

[0250] In the technical solution of the above embodiments, the first trend feature is a feature reflecting the changing trend of the dimensional parameters of one or more structural anomalies in the second structural data. Similarly, the second trend feature is a feature reflecting the changing trend of the second vibration data.

[0251] Furthermore, prior to the step of generating the first trend feature of the second structural data and the second trend feature of the second vibration data, the method further includes:

[0252] Determine the number of structural outliers in the second structural data;

[0253] The matching degree threshold is configured based on the number of structural anomalies determined in the second structural data.

[0254] In the technical solution of the above embodiments, the matching degree threshold is not a single fixed threshold. The size of the matching degree threshold is related to the number of structural anomalies in the second structural data. For second structural data with different numbers of structural anomalies, the matching degree threshold for matching with the second vibration data is different.

[0255] Furthermore, the step of generating the first trend feature of the second structural data specifically includes:

[0256] Generate a second differential feature size parameter sequence for structural anomalies in the second structural data;

[0257] Obtain each size parameter p2 from the second differential feature size parameter sequence. i (t), where i∈[1,n] p ], where i is a positive integer and n p The number of size parameter types of the differential features, wherein the second differential feature size parameter sequence is a discrete data sequence constructed in order of the generation time t of the structure image of the second structure data;

[0258] For the i-th size parameter, fit the size parameter p2 with time t as the independent variable. i The first change curve function C1 of (t) i (t);

[0259] Extracting the first change curve function C1 i The set of time ranges corresponding to the rising interval of (t) (ts) ij ,teij ), where j∈[1, np i ], np i The first changing curve function C1 i The number of rising intervals in (t), ts ij The first changing curve function C1 i The start time of the j-th ascending interval in (t), te ij The first changing curve function C1 i The end time of the j-th ascending interval in (t);

[0260] The time range set (ts) ij ,te ij This is identified as the first trend feature of the second structural data.

[0261] In some embodiments of the present invention, the first change curve function C1 i (t) represents the size parameter p2 obtained by fitting using a polynomial fitting and curve fitting algorithm. i The first curve function C1 is a curve function showing how the magnitude of (t) changes over time. i (t) is used to represent the size parameter p2 i The magnitude of (t) changes over time in the future.

[0262] The first change curve function C1 i The rising interval of (t) refers to the first changing curve function C1 i The interval with a slope greater than 0 in (t), i.e., ts ij and te ij All are the first change curve function C1 i The moment when the slope of (t) is 0.

[0263] Furthermore, the second structural data is a structural image of the target structural component, and the step of generating a second differential feature size parameter sequence for structural anomalies in the second structural data specifically includes:

[0264] Based on the first structural data, determine the target location of the structural anomaly point in the target structural component;

[0265] Generate a sequence of second structural images of the target structural component over a future period based on the second structural data;

[0266] Generate a second differential feature size parameter sequence corresponding to the target location in the second structural image sequence.

[0267] Furthermore, the step of generating the second trend feature of the second vibration data specifically includes:

[0268] Obtain the vibration intensity v(t) from the second vibration data;

[0269] The second variation curve function C2(t) of the vibration intensity v(t) is fitted with time t as the independent variable;

[0270] Extract the set of rising inflection points of the second change curve function C2(t) tip k , where k∈[1, n v ], n v The number of rising inflection points in the second change curve function C2(t), wherein the rising inflection point is the inflection point in the second change curve function C2(t) where the slope increases;

[0271] The set of rising inflection points tip k The second trend feature of the second vibration data was determined.

[0272] The vibration intensity v(t) is the vibration intensity of the target structure detected by the vibration monitoring sensor at time t. The second variation curve function C2(t) is a curve function of the magnitude of the vibration intensity v(t) changing with time, obtained by using curve fitting algorithms such as polynomial fitting. The second variation curve function C2(t) is used to represent the change of the magnitude of the vibration intensity v(t) with time over a future period.

[0273] Furthermore, extract the set of rising inflection points tip of the second change curve function C2(t). k The specific steps include:

[0274] Identify each inflection point in the second change curve function C2(t);

[0275] Calculate the first slope at Δt before each inflection point and the second slope at Δt after each inflection point, where Δt is a pre-configured slope calculation interval;

[0276] The inflection point where the second slope is greater than the first slope is determined as the rising inflection point.

[0277] Furthermore, the step of calculating the matching degree between the first trend feature and the second trend feature specifically includes:

[0278] Determine the number of structural anomalies on the target structural component;

[0279] When the number of structural anomalies on the target structural component is 1, obtain the np corresponding to the structural anomaly in the second structural data.i A time range set (ts) ij ,te ij );

[0280] For each time range set (ts) ij ,te ij Based on the set of rising inflection points of the second vibration data, tip k The number of items falling within its time range is used to calculate its matching score s. i ;

[0281] np i The maximum of the matching scores The target matching degree P between the first trend feature and the second trend feature is determined.

[0282] In the technical solution of the above embodiments, the target matching degree P is a matching degree used to determine whether the second structural data and the second vibration data satisfy a preset correspondence. That is, in the step of determining whether the matching degree between the first trend feature and the second trend feature is greater than a preset matching degree threshold, specifically, it is determined whether the target matching degree P is greater than the preset matching degree threshold.

[0283] Furthermore, for each time range set (ts) ij ,te ij Based on the set of rising inflection points of the second vibration data, tip k The number of items falling within its time range is used to calculate its matching score s. i The specific steps include:

[0284] Statistical analysis of the set of rising inflection points of the second vibration data tip k The number of rising inflection points falling into the i-th time range set, np i ;

[0285] The set of rising inflection points of the second vibration data tip k The number of rising inflection points that fall into the i-th time range set, np i The ratio of the total number of rising inflection points (nv) relative to the set of rising inflection points in the second vibration data is used to determine its matching score:

[0286]

[0287] Furthermore, each structural anomaly point has a corresponding first trend characteristic. Following the step of determining the number of structural anomaly points on the target structural component, the method further includes:

[0288] When the number of structural anomalies on the target structural component is greater than one, the distance d between each structural anomaly and the vibration monitoring point is determined. l , where l∈[1, n un ], n un The number of structural anomalies on the target structural component;

[0289] Based on the distance d between each structural anomaly point and the vibration monitoring point l Configure the influence weight σ of each structural anomaly on the matching degree between the first trend feature and the second trend feature. l ;

[0290] Calculate the matching degree pl between the first trend feature and the second trend feature of each structural anomaly. ；

[0291] Based on the influence weight σ of each structural anomaly point l The degree of matching p between its first trend feature and its second trend feature l Calculate the target matching degree used to determine whether the second structural data and the second vibration data satisfy a preset correspondence:

[0292]

[0293] Specifically, the vibration monitoring point is the location on the target structural component where the vibration monitoring sensor is installed. The greater the distance from the vibration monitoring point, the smaller the weight of the influence of the structural anomaly on the matching degree of the first trend feature and the second trend feature.

[0294] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0295] As described above, these embodiments of the present invention do not exhaustively cover all details, nor do they limit the invention to the specific embodiments described. Clearly, many modifications and variations can be made based on the above description. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to effectively utilize the invention and its modifications. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A crane monitoring system, characterized in that, It includes a structural monitoring sensor for monitoring structural changes of target structural components of the crane, a vibration monitoring sensor for monitoring the structural stability of target structural components of the crane, and a processing unit communicatively connected to the structural monitoring sensor and the vibration monitoring sensor. The processing unit is configured to: Acquire first structural data of the target structural component and first vibration data corresponding to the first structural data. The first structural data includes internal first structural data, shape monitoring data and / or surface monitoring data of the target structural component. The first structural data is a structural image of the target structural component. Configure a standard image corresponding to the target structural component and the structural image, wherein the standard image is a structural image generated of the target structural component in a damage-free state; Obtain the first structural image sequence of the target structural component over a past period of time; A first difference feature sequence is generated based on the difference features between each structural image in the first structural image sequence and the standard image; Generate a first differential feature size parameter sequence corresponding to the first differential feature sequence. The first differential feature size parameter sequence is composed of one or more size parameters in different directions for each differential feature in the first differential feature sequence. A trend analysis is performed on the first differential feature size parameter sequence to determine whether the corresponding differential feature is a structural anomaly based on the trend analysis results; When structural anomalies exist in the structural image, a local structural image of the target structural component at the structural anomaly point is generated. The local structural image is input into a pre-trained damage recognition model to identify damage features in order to determine whether the first structural data has damage features. When the first structural data does not have damage characteristics, predict the second structural data and the second vibration data in the future period; Generate a first trend feature of the second structural data and a second trend feature of the second vibration data; Calculate the matching degree between the first trend feature and the second trend feature; Determine whether the matching degree between the first trend feature and the second trend feature is greater than a preset matching degree threshold; When the matching degree between the first trend feature and the second trend feature is greater than a preset matching degree threshold, it is determined that the second structural data and the second vibration data satisfy a preset correspondence. When the second structural data and the second vibration data satisfy a preset correspondence, it is determined that the target structural component is at risk of damage. The damage risk level of the target structural component is calculated based on the second vibration data; A maintenance strategy for the target structural component is generated based on the damage risk level.

2. A prediction-based crane maintenance method, characterized in that, include: Acquire first structural data of the target structural component and first vibration data corresponding to the first structural data. The first structural data includes internal first structural data, shape monitoring data and / or surface monitoring data of the target structural component. The first structural data is a structural image of the target structural component. Configure a standard image corresponding to the target structural component and the structural image, wherein the standard image is a structural image generated of the target structural component in a damage-free state; Obtain the first structural image sequence of the target structural component over a past period of time; A first difference feature sequence is generated based on the difference features between each structural image in the first structural image sequence and the standard image; Generate a first differential feature size parameter sequence corresponding to the first differential feature sequence. The first differential feature size parameter sequence is composed of one or more size parameters in different directions for each differential feature in the first differential feature sequence. A trend analysis is performed on the first differential feature size parameter sequence to determine whether the corresponding differential feature is a structural anomaly based on the trend analysis results; When structural anomalies exist in the structural image, a local structural image of the target structural component at the structural anomaly point is generated. The local structural image is input into a pre-trained damage recognition model to identify damage features in order to determine whether the first structural data has damage features. When the first structural data does not have damage characteristics, predict the second structural data and the second vibration data in the future period; Generate a first trend feature of the second structural data and a second trend feature of the second vibration data; Calculate the matching degree between the first trend feature and the second trend feature; Determine whether the matching degree between the first trend feature and the second trend feature is greater than a preset matching degree threshold; When the matching degree between the first trend feature and the second trend feature is greater than a preset matching degree threshold, it is determined that the second structural data and the second vibration data satisfy a preset correspondence. When the second structural data and the second vibration data satisfy a preset correspondence, it is determined that the target structural component is at risk of damage. The damage risk level of the target structural component is calculated based on the second vibration data; A maintenance strategy for the target structural component is generated based on the damage risk level.

3. The prediction-based crane maintenance method according to claim 2, characterized in that, The step of performing trend analysis on the first differential feature size parameter sequence, and determining whether the corresponding differential feature is a structural anomaly based on the trend analysis results, specifically includes: Obtain each size parameter from the first differential feature size parameter sequence. ,in , It is a positive integer. The number of size parameter types for the differential features, wherein the first differential feature size parameter sequence is generated according to the generation time of the image based on the first structural data structure. A discrete data sequence constructed for a given order; For the Various size parameters, in time Fit the size parameters to the independent variable. Trend line function ; Determine the trend line function slope in Is it greater than 0? exist Within the range, when any one The value satisfies At that time, it is determined that there are structural anomalies in the structural image.

4. The prediction-based crane maintenance method according to claim 2, characterized in that, The steps for generating the first trend feature of the second structural data specifically include: Generate a second differential feature size parameter sequence for structural outliers in the second structural data; Obtain each size parameter in the second differential feature size parameter sequence. ,in , It is a positive integer. The second difference feature size parameter sequence is the generation time of the structural image based on the second structural data, representing the number of size parameter types of the difference features. A discrete data sequence constructed for a given order; For the Various size parameters, in time Fit the size parameters to the independent variable. The first change curve function ; Extract the first change curve function The set of time ranges corresponding to the rising interval ,in , The first change curve function The number of rising intervals The first change curve function The Middle The start time of each upward interval. The first change curve function The Middle The end time of each upward interval; The time range set The first trend feature of the second structural data was identified.

5. The prediction-based crane maintenance method according to claim 4, characterized in that, The steps for generating the second trend feature of the second vibration data specifically include: Obtain the vibration intensity from the second vibration data ; In time Fit the vibration intensity to the independent variable The second change curve function ; Extract the second change curve function set of rising inflection points ,in , The second change curve function The number of rising inflection points in the second change curve function, wherein the rising inflection points are the number of rising inflection points in the second change curve function. The inflection point where the slope increases; The set of rising inflection points The second trend feature of the second vibration data was determined.

6. The prediction-based crane maintenance method according to claim 5, characterized in that, The steps for calculating the matching degree between the first trend feature and the second trend feature specifically include: Determine the number of structural anomalies on the target structural component; When the number of structural anomalies on the target structural component is 1, the corresponding structural anomaly point is obtained from the second structural data. A set of time ranges ; For each time range set Based on the set of rising inflection points of the second vibration data The number of items falling within its time range is used to calculate its matching score. ; Will The maximum of the matching scores The target matching degree between the first trend feature and the second trend feature is determined. .

7. The prediction-based crane maintenance method according to claim 6, characterized in that, Each structural anomaly has a corresponding first trend characteristic. After determining the number of structural anomalies on the target structural component, the method further includes: When the number of structural anomalies on the target structural component is greater than one, the distance between each structural anomaly and the vibration monitoring point is determined. ,in , The number of structural anomalies on the target structural component; Based on the distance between each structural anomaly point and the vibration monitoring point Configure the influence weight of each structural anomaly on the matching degree between the first trend feature and the second trend feature. ; Calculate the matching degree between the first trend feature and the second trend feature for each structural anomaly. ; Based on the influence weight of each structural anomaly point The degree of matching between its first trend feature and its second trend feature. Calculate the target matching degree used to determine whether the second structural data and the second vibration data satisfy a preset correspondence: 。

Images