Cargo boom fault diagnosis method, device, equipment and medium

Abstract

The invention relates to the technical field of cranes, and discloses a crane boom fault diagnosis method, device and equipment and a medium. The method comprises the following steps: acquiring to-be-detected working data of a plurality of working periods of a target cargo boom; the to-be-detected working data of all the working periods are recognized based on a fault recognition reference model, a plurality of fault detection results are obtained, and the fault recognition reference model is established according to historical working data of the fault cargo boom; and determining a fault diagnosis result of the crane big arm based on the change trend of the plurality of fault detection results. According to the method, a plurality of fault detection results are combined together to analyze the change trend of the fault detection result data, and the fault deterioration condition of the cargo boom is determined based on the data change trend, so that a data-driven long-term fault diagnosis scheme of the cargo boom is realized, and the defect that the cargo boom lacks a long-term fault diagnosis method is overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of cranes, and in particular to a crane arm fault diagnosis method, device, equipment and medium. Background Art

[0002] Cranes are one of the common lifting equipment, usually using a telescopic and lifting boom for cargo lifting operations. Since the boom is usually composed of multiple telescopic sections, it is easy to produce abnormalities during long-term use. At present, some commonly used engineering machinery diagnosis methods mainly evaluate abnormalities based on the real-time operating status of the equipment, but cannot effectively diagnose the long-term failure trend of equipment failures. For crane arms, long-term fault diagnosis is a more important requirement. Summary of the invention

[0003] In view of this, the present invention provides a boom fault diagnosis method, device, equipment, medium and product to solve the problem of lack of long-term fault diagnosis method.

[0004] In a first aspect, the present invention provides a method for diagnosing boom faults, the method comprising: collecting working data to be detected of a target boom for multiple working cycles; identifying the working data to be detected of each of the working cycles based on a fault identification benchmark model to obtain multiple fault detection results, wherein the fault identification benchmark model is established based on historical working data of the faulty boom; and determining the boom fault diagnosis result based on the changing trend of the multiple fault detection results.

[0005] In some optional implementations, the historical working data includes: one or more of: equipment identification of the crane, main boom length signal, second-section boom length signal, main boom angle signal and signal duration.

[0006] In some optional embodiments, the step of creating a fault identification benchmark model based on the historical working data includes: screening normal signals and fault signals in the main arm length signal, the second-section arm length signal and the main arm angle signal; calculating normal speed distribution characteristics using the normal signal and the signal duration; calculating fault speed distribution characteristics using the fault signal and the signal duration; and training a fault identification benchmark model for classifying fault signals through the fault speed distribution characteristics and the normal speed distribution characteristics.

[0007] In some optional embodiments, the fault detection result is the signal similarity between the working data to be detected output by the fault identification benchmark model and the fault signal; the fault diagnosis result of the crane arm is determined based on the change trend of the multiple fault detection results, including: using the signal similarity detected in each working cycle to calculate multiple similarity features of preset time lengths, the similarity features at least include one of the similarity mean, similarity variance, and similarity median; if there is a target similarity feature higher than a preset similarity threshold in the similarity features of the preset time length, then calculating a health score based on the offset trend of the target similarity feature; determining the fault diagnosis result of the crane arm through the health score.

[0008] In some optional embodiments, the health score is calculated based on the deviation trend of the similarity feature, including: obtaining a health baseline score; comparing the target similarity feature of each subsequent preset time length with the target similarity feature of the previous preset time length; if the comparison result indicates that the target similarity feature of the subsequent preset time length has an increase, then deducting the preset score from the health baseline score to obtain the health score.

[0009] In some optional embodiments, the fault identification benchmark model for classifying fault signals is trained by the fault speed distribution characteristics and the normal speed distribution characteristics, comprising: creating a basic model based on the lightGBM classification algorithm; substituting the fault speed distribution characteristics and the normal speed distribution characteristics into the basic model respectively, fitting the model parameters of the basic model, and obtaining the fault identification benchmark model.

[0010] In some optional embodiments, the method further includes: receiving a fault identification instruction; collecting current working data of the crane arm based on the fault identification instruction; correcting the fault identification benchmark model based on the current working data, and re-performing fault detection based on the corrected fault identification benchmark model.

[0011] In a second aspect, the present invention provides a boom fault diagnosis device, the device comprising: a data acquisition module for detection, used to collect the working data to be detected of multiple working cycles of the target boom; a fault detection module, used to identify the working data to be detected of each of the working cycles based on a fault identification benchmark model, and obtain multiple fault detection results, wherein the fault identification benchmark model is established based on the historical working data of the faulty boom; a fault deterioration detection module, used to determine the fault diagnosis result of the boom based on the changing trend of the multiple fault detection results.

[0012] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0013] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to cause a computer to execute the method of the first aspect or any corresponding embodiment thereof.

[0014] The technical solution provided by the present invention has the following advantages:

[0015] By collecting the historical working data of the crane arm and training and generating a fault identification benchmark model based on the historical working data, the working data to be detected of the crane arm in multiple working cycles are collected and respectively input into the fault identification benchmark model for fault identification, and the fault detection results corresponding to each working cycle are obtained. Finally, the multiple fault detection results are combined together to analyze the change trend of the fault detection result data, and the fault deterioration of the crane arm is determined based on the data change trend, thus realizing a data-driven long-term fault diagnosis scheme for the crane arm, which makes up for the lack of long-term fault diagnosis methods for the crane arm. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 is a schematic diagram of a process flow of boom fault diagnosis according to an embodiment of the present invention;

[0018] Figure 2 is a schematic structural diagram of a boom fault diagnosis device according to an embodiment of the present invention;

[0019] Figure 3 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0021] According to an embodiment of the present invention, an embodiment of a crane arm fault diagnosis method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0022] In this embodiment, a method for diagnosing a boom fault is provided. Figure 1 is a flow chart of a boom fault diagnosis method according to an embodiment of the present invention, the flow chart comprising the following steps:

[0023] Step S101, collecting the working data to be detected of multiple working cycles of the target crane arm;

[0024] Step S102, identifying the working data to be detected in each working cycle based on a fault identification benchmark model to obtain multiple fault detection results, wherein the fault identification benchmark model is established based on the historical working data of the faulty boom;

[0025] Step S103, determining a boom fault diagnosis result based on a change trend of a plurality of fault detection results.

[0026] Specifically, the embodiment of the present invention first collects historical working data of the crane arm, and generates a fault identification benchmark model based on the training of the historical working data. The historical working data of the crane arm is the boom data collected by controlling the movement of the boom when the crane is working in the historical time, including but not limited to the telescopic data of the boom, the rotation angle data of the boom, the pressure signal, the processor control signal, etc. The historical working data can be normal working data or faulty working data, and the fault identification benchmark model for fault detection can be trained and generated through the normal working data or the faulty working data.

[0027] In some embodiments, historical work data may be data collected during the boom maintenance phase, mainly repairing faulty vehicles with slow boom extension and retraction failures, as well as specific time points before and after the repairs, and this part of the data is stored in a big data warehouse.

[0028] The collected data specifically includes data collected before maintenance and data collected after maintenance. The data collected before maintenance can be determined as fault data. After maintenance, the crane arm has been repaired, so the collected data can be determined as normal data. The fault identification benchmark model that classifies fault categories by jointly training the fault data and normal data can improve the accuracy of fault diagnosis.

[0029] After creating the fault identification benchmark model, the embodiment of the present invention collects the working data to be detected of multiple working cycles of the crane arm. It should be noted that the fault identification benchmark model created by the embodiment of the present invention corresponds one-to-one to the crane, that is, a fault identification benchmark model needs to be created for each crane. When subsequently identifying the boom fault, it is also necessary to collect the working data of the crane corresponding to the current fault identification benchmark model.

[0030] In this embodiment, the collected work data to be detected is the same type as the historical work data used for the training model. Multiple work cycles refer to different working time periods of the crane. For example, the crane can be set to work for one day as a cycle. This embodiment is only an example and is not limited to this.

[0031] Afterwards, the working data to be detected in each working cycle are respectively input into the fault identification benchmark model for fault identification, and the fault detection results corresponding to each working cycle are obtained. Then, multiple fault detection results are combined together to analyze the change trend of the fault detection result data. For example, the fault detection result data can show an increase, decrease, etc. trend. Finally, the fault deterioration of the boom is determined based on the data change trend to obtain the final fault diagnosis result. Through the technical solution provided by the embodiment of the present invention, a data-driven long-term fault diagnosis solution for the boom is realized, which can determine whether the boom has a fault deterioration in a long period of time, making up for the lack of a long-term fault diagnosis method for the boom.

[0032] In some optional implementations, the above historical working data includes one or more of the equipment identification of the crane, the main boom length signal, the second-section boom length signal, the main boom angle signal and the signal duration.

[0033] Specifically, an embodiment of the present invention provides a solution for creating a fault identification model based on the boom extension speed. In order to determine the boom extension speed, at least the main boom length signal, the second boom length signal, the main boom angle signal and the signal duration are collected. The main boom extension speed, the second boom extension speed and the main boom angular velocity of the crane can be determined by the ratio of the main boom length signal, the second boom length signal and the main boom angle signal to the signal duration. The device identifier is used to match the training model data with the crane one-to-one.

[0034] Based on this, the steps to create a fault identification benchmark model based on historical working data include:

[0035] Step b1, screening the normal signal and the fault signal among the main boom length signal, the second-section boom length signal and the main boom angle signal;

[0036] Step b2, calculating normal speed distribution characteristics using normal signals and signal duration;

[0037] Step b3, calculating the fault velocity distribution characteristics using the fault signal and the signal duration;

[0038] Step b4, training a fault identification benchmark model for classifying fault signals by using the fault velocity distribution characteristics and the normal velocity distribution characteristics.

[0039] Specifically, the embodiment of the present invention first selects normal signals and fault signals from the main boom length signal, the second boom length signal and the main boom angle signal, wherein the normal signal and the fault signal can be determined by the above-mentioned maintenance time. Afterwards, for the fault signal and the normal signal, the arm length telescopic speed signal is calculated, and the main boom telescopic speed, the second boom telescopic speed and the main boom angular velocity of the crane can be determined by the ratio of the main boom length signal, the second boom length signal and the main boom angle signal to the signal duration. The above-mentioned speed signal corresponding to the fault signal is the fault arm length telescopic speed signal, and the speed signal corresponding to the normal signal is the normal arm length telescopic speed signal. Afterwards, the speed distribution characteristics are calculated for the fault arm length telescopic speed signal and the normal arm length telescopic speed signal, for example: the proportion of [-1,-0.9]m / s, the proportion of [-0.9,-0.8]m / s, and so on. Finally, a fault label is added to the fault speed distribution feature, and a normal label is added to the normal speed distribution feature. The two types of data are respectively input into a specific machine learning model, and a fault identification benchmark model for classifying fault signals can be trained and generated.

[0040] The embodiment of the present invention calculates the arm length extension and retraction speed from the arm length extension and retraction data, and then uses the distribution range of the arm length extension and retraction speed to form sample features to train a fault recognition benchmark model. It combines the characteristic parameters of the crane arm extension and retraction operation to make the training data more recognizable and in-depth, thereby improving the accuracy of subsequent fault recognition of the fault recognition benchmark model.

[0041] In addition, in the embodiment of the present invention, data such as hoisting signals and accelerator pedal sampling values ​​may be further collected to assist in training the fault identification benchmark model, so as to improve the accuracy of the fault identification benchmark model.

[0042] In addition, in the embodiment of the present invention, the operation data collected by the crane can be wirelessly transmitted in real time to the cloud data center for storage, processing, and analysis. The fault identification benchmark model can be calculated and trained by the cloud server. Based on the powerful computing power of the cloud server, the model training efficiency and the subsequent fault identification efficiency can be significantly improved.

[0043] In some optional embodiments, the present invention can train a fault identification benchmark model to create a basic model based on the lightGBM classification algorithm; then substitute the fault speed distribution characteristics and the normal speed distribution characteristics into the basic model respectively, fit the model parameters of the basic model, and obtain the fault identification benchmark model. Among them, the model training set is 80% of the historical fault data, and the test set is the remaining 20% ​​of the historical fault data. The current model performance is evaluated by the test set. The input of the model is the arm retraction feature data within the cycle, and the output is whether there is an arm retraction abnormality within the cycle, 0 represents an abnormality, and 1 represents no abnormality.

[0044] lightGBM (Light Gradient Boosting Machine) is a machine learning algorithm based on the gradient boosting framework, specifically designed to solve problems such as classification and regression. Compared with traditional gradient boosting methods, lightGBM has advantages in many aspects. Efficiency: lightGBM uses a technique called "histogram-based learning", which can process data efficiently, reduce memory consumption and increase training speed. This is achieved by dividing the data into histograms, thereby reducing complexity. Fast: Due to the use of histogram-based learning and some other optimization techniques, lightGBM is very fast to train on large-scale data sets, so it is particularly useful when processing large amounts of data. Accuracy: lightGBM uses a technique called "GOSS" (Gradient-basedOne-Side Sampling) in the process of growing decision trees to more efficiently select leaf nodes that need to be split. This helps to improve the accuracy of the model. Support for parallelization: LightGBM supports multi-threaded parallelization, which can make better use of computing resources on multi-core processors and further improve training speed. Scalability: LightGBM supports training and prediction of large-scale datasets and can handle a large number of features and samples with limited memory.

[0045] The embodiment of the present invention creates a fault identification benchmark model based on the lightGBM algorithm, which can significantly improve the model performance of the fault identification benchmark model.

[0046] In some optional implementations, the fault detection result is the signal similarity between the working data to be detected output by the fault identification benchmark model and the fault signal. The above step S103 includes:

[0047] Step c1, using the signal similarity detected in each working cycle to calculate the similarity features of multiple preset time lengths, the similarity features at least including one of the similarity mean, similarity variance, and similarity median;

[0048] Step c2: if there is a target similarity feature higher than a preset similarity threshold in the similarity features of the preset time length, a health score is calculated based on the deviation trend of the target similarity feature;

[0049] Step c3, determining the fault diagnosis result of the boom through the health score.

[0050] Specifically, since the fault identification benchmark model can often output the classification result in the form of probability, its output fault diagnosis result is the similarity between the current signal and the fault signal. For example, if the similarity between the current signal and the fault signal is 60%, it means that there is a 60% probability of classifying the current signal as a fault. If the similarity between the current signal and the fault signal is 90%, it means that there is a 90% probability of classifying the current signal as a fault. Thus, the embodiment of the present invention can use the signal similarity detected in each working cycle to calculate the similarity characteristics of multiple preset time lengths. For example, if the working cycle is 1 day, the preset time length can be 3 days, 7 days and 30 days. This embodiment is only used as an example and is not limited to this. Afterwards, the average value of the signal similarity in the past 3 / 7 / 30 days, the deviation ratio of the signal similarity in the past 3 / 7 / 30 days, the variance of the signal similarity in the past 3 / 7 / 30 days, the median of the signal similarity in the past 3 / 7 / 30 days, etc. can be calculated. The above-mentioned indicators calculated are called similarity characteristics. By calculating the above similarity features, it is necessary to compare the calculated index with the preset similarity threshold, wherein the preset similarity threshold needs to be defined separately for the similarity mean, similarity variance, similarity median, etc. The specific definition value is determined according to user needs, and this embodiment does not make any special restrictions. If the above similarity feature is higher than the preset similarity threshold, it means that the similarity with the fault signal is high, so the boom should have a fault, and the corresponding similarity feature that can determine the fault is called the target similarity feature. However, the purpose of the embodiment of the present invention is to detect the fault deterioration of the boom, which is a long-term fault detection method, so it is necessary to further use the offset trend of the target similarity feature that can determine the fault to calculate the health score, and finally, determine the fault deterioration of the boom through the health score.

[0051] Based on the technical solution provided by the embodiment of the present invention, the similarity features of multiple different time periods are calculated using the result data of the fault diagnosis, and the comprehensive performance of the boom in each working cycle is further integrated, which can weaken the impact of the data deviation of a single working cycle on the overall fault deterioration detection. The health score is calculated using the deviation trend of the target similarity feature of the determined fault, and the effect of long-term detection of the boom fault is also achieved at the same time.

[0052] In some optional implementations, the above step c4 includes:

[0053] Step d1, obtaining a health benchmark score;

[0054] Step d2, comparing the target similarity feature of each subsequent preset time length with the target similarity feature of the previous preset time length;

[0055] Step d3: If the comparison result indicates that the target similarity feature of the next preset time length has increased, the preset score is deducted from the health benchmark score to obtain a health score.

[0056] Specifically, the embodiment of the present invention sets a health benchmark score, such as 100 points, which is only used as an example and is not limited to this. Afterwards, the target similarity feature of each subsequent preset time length is compared with the target similarity feature of the previous preset time length, for example, the target similarity feature of 7 days is compared with the target similarity feature of 3 days. If there is an increase in the target similarity feature, it means that the similarity of the fault increases, and the abnormality continues to occur, so that the score will continue to decrease. In each time period, the score of the arm retraction abnormality decreases by a fixed score, for example, 10 points are deducted from the preset score each time. When the score is lower than the preset score threshold (for example, 60 points), it can be determined that the fault has deteriorated, which significantly improves the accuracy of long-term fault detection.

[0057] In some optional implementations, the boom fault diagnosis method provided by the present invention further includes:

[0058] Step e1, receiving a fault identification instruction;

[0059] Step e2, collecting current working data of the crane arm based on the fault identification instruction;

[0060] Step e3, correcting the fault identification benchmark model based on the current working data, and re-performing fault detection based on the corrected fault identification benchmark model.

[0061] Specifically, when a fault is discovered manually but the fault identification benchmark model does not identify the fault, the embodiment of the present invention can also upload a fault identification instruction to the server through the fault reporting terminal deployed on the crane. The server receives the fault identification instruction input by the user, and collects the current working data of the crane arm based on the fault identification instruction, and corrects the fault identification benchmark model through the current working data, thereby further improving the accuracy of the fault identification benchmark model. Afterwards, the fault deterioration condition is re-detected using the corrected fault identification benchmark model to verify the accuracy of the corrected fault identification benchmark model. If the correction effect is not good, the correction can continue to be made, thereby further improving the reliability of the fault identification benchmark model.

[0062] In this embodiment, a boom fault diagnosis device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0063] This embodiment provides a boom fault diagnosis device, such as Figure 2 As shown, including:

[0064] The detection data collection module 201 is used to collect the detection working data of the target crane arm in multiple working cycles;

[0065] A fault detection module 202 is used to identify the working data to be detected in each working cycle based on a fault identification benchmark model to obtain multiple fault detection results, wherein the fault identification benchmark model is established based on the historical working data of the faulty crane arm;

[0066] The fault deterioration detection module 203 is used to determine the fault diagnosis result of the boom based on the change trend of multiple fault detection results.

[0067] In some optional implementations, the device further includes a historical data collection module, specifically including:

[0068] The data acquisition unit collects the equipment identification of the crane, the main boom length signal, the second-section boom length signal, the main boom angle signal and the signal duration.

[0069] In some optional implementations, the apparatus further includes a reference model creation module, specifically including:

[0070] A signal screening unit, used for screening out normal signals and fault signals from the main boom length signal, the second-section boom length signal and the main boom angle signal;

[0071] A first speed feature calculation unit, used for calculating a normal speed distribution feature using a normal signal and a signal duration;

[0072] A second speed characteristic calculation unit is used to calculate the fault speed distribution characteristic by using the fault signal and the signal duration;

[0073] The training unit is used to train a fault identification benchmark model for classifying fault signals through fault velocity distribution features and normal velocity distribution features.

[0074] In some optional implementations, the fault detection result is the signal similarity between the working data to be detected output by the fault identification benchmark model and the fault signal. The fault deterioration detection module 203 includes:

[0075] A similarity feature calculation unit, used to calculate similarity features of multiple preset time lengths using the signal similarities detected in each working cycle, wherein the similarity feature includes at least one of a similarity mean, a similarity variance, and a similarity median;

[0076] A target feature detection unit, used to determine whether there is a target similarity feature higher than a preset similarity threshold among the similarity features of each preset time length;

[0077] A boom scoring unit, configured to calculate a health score based on a deviation trend of the target similarity feature if there is a target similarity feature higher than a preset similarity threshold in the similarity features of the preset time length;

[0078] The deterioration evaluation unit is used to determine the fault diagnosis result of the crane arm through the health score.

[0079] In some optional implementations, the scoring unit includes:

[0080] A benchmark score acquisition unit, used to acquire a health benchmark score;

[0081] A comparison unit, used to compare the target similarity feature of each subsequent preset time length with the target similarity feature of the previous preset time length;

[0082] The score updating unit is used to deduct a preset score from the health benchmark score to obtain a health score if the comparison result indicates that the target similarity feature of the next preset time length has increased.

[0083] In some optional embodiments, the device further comprises:

[0084] A manual instruction receiving module, used for receiving fault identification instructions;

[0085] The data upload module is used to collect the current working data of the crane arm based on the fault identification instruction;

[0086] The model correction module is used to correct the fault identification benchmark model based on the current working data, and re-perform fault detection based on the corrected fault identification benchmark model.

[0087] The further functional description of each of the above modules and units is the same as that of the above corresponding method embodiments and will not be repeated here.

[0088] The boom fault diagnosis device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0089] The embodiment of the present invention also provides a computer device having the above Figure 2 The boom fault diagnostic device shown.

[0090] See also Figure 3 , Figure 3 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 3 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 3 A processor 10 is taken as an example.

[0091] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0092] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0093] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0094] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0095] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0096] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0097] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0098] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A crane arm fault diagnosis method, characterized in that: The method comprises: Collect the working data to be tested of multiple working cycles of the target crane arm; Based on the fault identification benchmark model, the working data to be detected of each working cycle is identified to obtain multiple fault detection results, wherein the fault identification benchmark model is established based on the historical working data of the faulty crane arm; A fault diagnosis result of the boom is determined based on the change trends of the multiple fault detection results.

2. The method according to claim 1, characterized in that: The historical working data includes: one or more of: equipment identification of the crane, main boom length signal, second-section boom length signal, main boom angle signal and signal duration.

3. The method according to claim 2, characterized in that The step of creating a fault identification benchmark model based on the historical working data comprises: Screening normal signals and fault signals among the main boom length signal, the second-section boom length signal and the main boom angle signal; calculating a normal speed distribution characteristic using the normal signal and the signal duration; Calculating fault speed distribution characteristics using the fault signal and the signal duration; A fault identification benchmark model for classifying fault signals is trained using the fault speed distribution features and the normal speed distribution features.

4. The method according to claim 3, characterized in that The fault detection result is the signal similarity between the working data to be detected output by the fault identification benchmark model and the fault signal; The determining of the boom fault diagnosis result based on the change trend of the multiple fault detection results includes: Calculate similarity features of multiple preset time lengths using the signal similarities detected in each working cycle, wherein the similarity features include at least one of a similarity mean, a similarity variance, and a similarity median; If there is a target similarity feature higher than a preset similarity threshold in the similarity features of the preset time length, then calculating the health score based on the deviation trend of the target similarity feature; The fault diagnosis result of the boom is determined by the health score.

5. The method according to claim 4, characterized in that The calculating of the health score based on the deviation trend of the similarity feature includes: Get a baseline health score; Comparing the target similarity feature of each subsequent preset time length with the target similarity feature of the previous preset time length; If the comparison result indicates that the target similarity feature of the next preset time length has increased, the preset score is deducted from the health benchmark score to obtain the health score.

6. The method according to claim 3, characterized in that The fault identification benchmark model for classifying fault signals is trained by the fault speed distribution feature and the normal speed distribution feature, comprising: Create a basic model based on the lightGBM classification algorithm; The fault velocity distribution characteristics and the normal velocity distribution characteristics are respectively substituted into the basic model, and the model parameters of the basic model are fitted to obtain the fault identification reference model.

7. The method according to claim 1, characterized in that The method further comprises: Receive fault identification instructions; Collecting current working data of the crane arm based on the fault identification instruction; The fault identification reference model is corrected based on the current working data, and fault detection is re-performed based on the corrected fault identification reference model.

8. A crane arm fault diagnosis device, characterized in that: The device comprises: The data collection module to be tested is used to collect the working data to be tested of the target crane arm in multiple working cycles; A fault detection module, used for identifying the working data to be detected in each working cycle based on a fault identification benchmark model to obtain multiple fault detection results, wherein the fault identification benchmark model is established based on the historical working data of the faulty crane arm; The fault deterioration detection module is used to determine the fault diagnosis result of the crane arm based on the change trend of the multiple fault detection results.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 7.

Images