Self-adaptive industrial robot fault diagnosis method and device and industrial robot

Abstract

The invention provides a fault diagnosis method and device for a self-adaptive industrial robot and the industrial robot, and the method comprises the steps: obtaining a vibration spectrum, a current harmonic wave and a temperature gradient of the industrial robot, and obtaining a multi-modal data set; determining an equipment fault index of the industrial robot according to the multi-modal data set; the accumulative operation duration and the fault history of the industrial robot are obtained, the aging degree of the industrial robot is determined according to the accumulative operation duration and the fault history, and the aging correction amount is obtained; and correcting a preset fault threshold value by adopting the aging correction quantity to obtain a dynamic fault threshold value, and generating alarm information under the condition that the equipment fault index is greater than the dynamic fault threshold value. According to the method, the problem of low accuracy of industrial robot fault diagnosis by adopting a static threshold model in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of robot fault diagnosis technology, and more specifically, to an adaptive industrial robot fault diagnosis method, apparatus, computer-readable storage medium, and industrial robot. Background Technology

[0002] Modern production lines place increasing demands on the continuous operational stability of industrial robots. Traditional "post-maintenance" methods result in significant losses, while regular maintenance can lead to premature replacement of some parts.

[0003] To address the aforementioned issues, existing technologies use fault prediction techniques to estimate the condition of components and replace them accordingly. However, because these fault prediction techniques employ static threshold models for fault diagnosis, such as alarms based on a fixed current >15A, the false alarm rate is high, making accurate fault diagnosis difficult to achieve. Summary of the Invention

[0004] The main objective of this application is to provide an adaptive industrial robot fault diagnosis method, apparatus, computer-readable storage medium, and industrial robot, so as to at least solve the problem of low accuracy in industrial robot fault diagnosis using static threshold models in the prior art.

[0005] To achieve the above objectives, according to one aspect of this application, an adaptive fault diagnosis method for an industrial robot is provided, comprising: acquiring the vibration spectrum, current harmonics, and temperature gradient of the industrial robot to obtain a multimodal data set; determining an equipment fault index of the industrial robot based on the multimodal data set, wherein the equipment fault index is used to quantify the performance loss of the industrial robot; acquiring the cumulative runtime and fault history of the industrial robot, and determining the aging degree of the industrial robot based on the cumulative runtime and fault history to obtain an aging correction amount; correcting a preset fault threshold using the aging correction amount to obtain a dynamic fault threshold; and generating an alarm message when the equipment fault index is greater than the dynamic fault threshold, wherein the alarm message is used to indicate that the industrial robot has a fault risk.

[0006] Optionally, the equipment failure index of the industrial robot is determined based on the multimodal data set, including: obtaining historical vibration spectrum, historical current harmonics and historical temperature gradient from the equipment log of the industrial robot to obtain a historical operating data set; determining the equipment failure index corresponding to the historical operating data set based on the equipment log to obtain a historical failure index; training a long short-term memory network with the historical operating data set as input data and the historical failure index as label data to obtain a target scoring network; and inputting the multimodal data set into the target scoring network to obtain the equipment failure index.

[0007] Optionally, the aging degree of the industrial robot is determined based on the cumulative runtime and fault history to obtain an aging correction amount, including: determining the impact of faults on the performance of the industrial robot based on the fault history to obtain a correction coefficient; and calculating the aging correction amount based on the correction coefficient and the cumulative runtime, wherein the calculation formula for the aging correction amount is HI1=a1*(1-e -0.00t1 In the formula, HI1 is the aging correction amount, a1 is the correction coefficient, and t is the cumulative running time.

[0008] Optionally, the preset fault threshold is corrected by aging correction to obtain a dynamic fault threshold, including: weighted summation of aging correction and preset fault threshold to obtain dynamic fault threshold.

[0009] Optionally, when the equipment failure index is greater than a dynamic failure threshold, an alarm message is generated, including: when the equipment failure index is greater than the dynamic failure threshold but less than or equal to a first threshold, a first alarm message is generated, which is used to indicate an adjustment to the production schedule of the industrial robot, and the ratio of the first threshold to the dynamic failure threshold is a first preset ratio; when the equipment failure index is greater than the first threshold but less than or equal to a second threshold, a second alarm message is generated, which is used to indicate that the industrial robot should stop operating after completing the processing of the current workpiece, and the ratio of the second threshold to the dynamic failure threshold is a second preset ratio; when the equipment failure index is greater than the second threshold, a third alarm message is generated, which is used to indicate that the industrial robot should stop operating immediately.

[0010] Optionally, the vibration spectrum, current harmonics, and temperature gradient of the industrial robot are acquired to obtain a multimodal data set, including: acquiring the timestamp of a preset master clock to obtain a first timestamp; extracting the timestamps of the vibration spectrum, current harmonics, and temperature gradient to obtain a second, third, and fourth timestamp; correcting the second, third, and fourth timestamps based on the first timestamp; and filling the data gaps generated during the correction process using linear interpolation based on the corrected second, third, and fourth timestamps to obtain the multimodal data set.

[0011] Optionally, determining the equipment failure index of the industrial robot based on the multimodal data set includes: identifying missing values ​​in the multimodal data set; using time series interpolation to determine the data values ​​corresponding to the missing values, and using the data values ​​to fill in the missing values, wherein the time series interpolation includes linear interpolation and polynomial interpolation; and determining the equipment failure index of the industrial robot based on the multimodal data set after filling in the missing values.

[0012] According to another aspect of this application, an adaptive industrial robot fault diagnosis device is provided. The device includes: an acquisition unit for acquiring the vibration spectrum, current harmonics, and temperature gradient of the industrial robot to obtain a multimodal data set; a first determination unit for determining an equipment fault index of the industrial robot based on the multimodal data set, wherein the equipment fault index is used to quantify the performance loss of the industrial robot; a second determination unit for acquiring the cumulative running time and fault history of the industrial robot, and determining the aging degree of the industrial robot based on the cumulative running time and fault history to obtain an aging correction amount; and a generation unit for correcting a preset fault threshold using the aging correction amount to obtain a dynamic fault threshold, wherein when the equipment fault index is greater than the dynamic fault threshold, an alarm message is generated to indicate that the industrial robot has a fault risk.

[0013] According to another aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform any of the methods described.

[0014] According to another aspect of this application, an industrial robot is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include methods for performing any one of the methods described.

[0015] Applying the technical solution of this application, in the aforementioned adaptive industrial robot fault diagnosis method, firstly, the vibration spectrum, current harmonics, and temperature gradient of the industrial robot are acquired to obtain a multimodal data set; then, the equipment fault index of the industrial robot is determined based on the multimodal data set, and the equipment fault index is used to quantify the performance loss of the industrial robot; subsequently, the cumulative runtime and fault history of the industrial robot are acquired, and the aging degree of the industrial robot is determined based on the cumulative runtime and fault history to obtain an aging correction amount; finally, the aging correction amount is used to correct a preset fault threshold to obtain a dynamic fault threshold. When the equipment fault index is greater than the dynamic fault threshold, an alarm message is generated, and the alarm message is used to indicate that the industrial robot has a fault risk. This application introduces the impact of equipment aging on the performance of industrial robots, quantifies it through the aging impact of cumulative runtime and faults, corrects the fault diagnosis threshold, and achieves adaptive fault diagnosis. By using a static threshold model, false alarms after equipment aging are eliminated, ensuring the accuracy of fault diagnosis and solving the problem of low accuracy in industrial robot fault diagnosis using static threshold models in the prior art. Attached Figure Description

[0016] Figure 1A hardware structure block diagram of a mobile terminal for an adaptive industrial robot fault diagnosis method provided in an embodiment of this application is shown.

[0017] Figure 2 A schematic flowchart of an adaptive industrial robot fault diagnosis method according to an embodiment of this application is shown.

[0018] Figure 3 A structural block diagram of an adaptive industrial robot fault diagnosis device provided according to an embodiment of this application is shown.

[0019] The above figures include the following reference numerals:

[0020] 102. Processor; 104. Memory; 106. Transmission device; 108. Input / output device. Detailed Implementation

[0021] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0023] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0024] As described in the background section, existing fault prediction technologies use static threshold models for fault diagnosis, such as fixed current >15A alarms, which have a high false alarm rate and are difficult to achieve accurate fault diagnosis. To solve the problem of low accuracy in fault diagnosis of industrial robots using static threshold models in existing technologies, embodiments of this application provide an adaptive fault diagnosis method, device, computer-readable storage medium, and industrial robot for industrial robots.

[0025] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0026] The methods and embodiments provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for an adaptive industrial robot fault diagnosis method according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0027] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the adaptive industrial robot fault diagnosis method in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the aforementioned networks may include wireless networks provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0028] This embodiment provides a fault diagnosis method for an adaptive industrial robot that runs on a mobile terminal, computer terminal, or similar computing device. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Also, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0029] Figure 2 This is a flowchart of an adaptive industrial robot fault diagnosis method according to an embodiment of this application. Figure 2 As shown, the method includes the following steps:

[0030] Step S201: Obtain the vibration spectrum, current harmonics, and temperature gradient of the industrial robot to obtain a multimodal data set;

[0031] Specifically, vibration spectrum analysis uses a triaxial vibration accelerometer (10kHz sampling rate) to monitor vibration signals generated by the joint movements of the industrial robot, and extracts spectral features using FFT (Fast Fourier Transform). Current harmonics analysis employs a Hall effect sensor (accuracy ±0.5%) to monitor motor current in real time, calculating the total harmonic distortion (THD) of the current to detect electrical system anomalies. Temperature gradient analysis utilizes an infrared thermal imager (temperature range -20℃ to 300℃) to monitor the temperature distribution in the joint area and identify thermal faults such as overheating.

[0032] Step S202: Determine the equipment failure index of the industrial robot based on the multimodal data set. The equipment failure index is used to quantify the performance loss of the industrial robot.

[0033] Specifically, this application sets up an LSTM network to analyze the temporal changes of multimodal data sets in order to determine the performance loss of industrial robots and obtain the aforementioned equipment failure index (HI value).

[0034] Step S203: Obtain the cumulative runtime and fault history of the industrial robot, and determine the aging degree of the industrial robot based on the cumulative runtime and fault history to obtain the aging correction amount.

[0035] Specifically, through statistical analysis, the relationship between the cumulative operating time of the equipment and its fault history is determined, and a coefficient representing the degree of equipment aging is calculated, namely the aforementioned aging correction amount. This coefficient reflects the gradual deterioration of the equipment over time and is used as a key parameter for correcting the preset fault threshold.

[0036] Step S204: The preset fault threshold is corrected by aging correction to obtain a dynamic fault threshold. When the equipment fault index is greater than the dynamic fault threshold, an alarm message is generated. The alarm message is used to indicate that there is a fault risk in the industrial robot.

[0037] Specifically, by determining and correcting the impact of aging on the performance of industrial robots, the threshold for fault diagnosis is changed to avoid misjudgment due to performance loss. This results in the aforementioned dynamic fault threshold. Furthermore, by comparing the equipment fault index with the aforementioned dynamic fault threshold, an alarm message is generated to determine the existence of a fault risk when the fault index is greater than the fault threshold.

[0038] This embodiment first acquires the vibration spectrum, current harmonics, and temperature gradient of the industrial robot to obtain a multimodal data set. Then, based on the multimodal data set, it determines the equipment failure index of the industrial robot, which quantifies the performance loss. Next, it acquires the cumulative runtime and fault history of the industrial robot, and determines the aging degree based on these data to obtain an aging correction amount. Finally, it uses the aging correction amount to correct a preset fault threshold, obtaining a dynamic fault threshold. When the equipment failure index exceeds the dynamic fault threshold, an alarm is generated to indicate a potential fault risk in the industrial robot. This application introduces the impact of equipment aging on industrial robot performance, quantifies it through cumulative runtime and the aging effect of faults, corrects the fault diagnosis threshold, and achieves adaptive fault diagnosis. By employing a static threshold model, it eliminates false alarms after equipment aging, ensuring the accuracy of fault diagnosis and solving the problem of low accuracy in fault diagnosis using static threshold models in existing technologies.

[0039] To determine the aforementioned equipment failure index, in one optional implementation, step S202 includes:

[0040] Step S2021: Obtain historical vibration spectrum, historical current harmonics and historical temperature gradient from the equipment log of the industrial robot to obtain historical operating data set;

[0041] Specifically, vibration spectrum, current harmonics, and temperature gradient data recorded during past operations are extracted from the equipment logs of industrial robots to cover the equipment's performance under various normal and fault conditions, resulting in the aforementioned historical operating data set.

[0042] Step S2022: Determine the equipment failure index corresponding to the historical operating data group based on the equipment log to obtain the historical failure index;

[0043] Specifically, based on historical fault records, fault index labels are assigned to the aforementioned historical operating data groups. For example, the HI value of the equipment before a minor fault is labeled as 1.05, while it may be labeled as 1.3 or higher before a serious fault.

[0044] Step S2023: Using historical operation data sets as input data and historical failure indices as labeled data, train a long short-term memory network to obtain the target scoring network;

[0045] Specifically, historical vibration spectrum, current harmonics, and temperature gradient data are standardized to eliminate the influence of dimensions. Using the processed historical operating data set as input and historical failure index as output label, an LSTM network is trained to predict the equipment failure index of the equipment based on multimodal data.

[0046] Step S2024: Input the multimodal data set into the target scoring network to obtain the equipment failure index.

[0047] Specifically, real-time multimodal data sets are input into a trained LSTM network, which outputs a real-time device failure index to instantly determine the health status of the device.

[0048] Through the above embodiments, by fusing multi-source data (vibration spectrum, current harmonics, temperature gradient), a prediction model for the equipment failure index (HI) is constructed using a long short-term memory network (LSTM). The LSTM neural network is used to perform deep learning on historical operating data to mine the intrinsic relationships between data. Using the historical failure index as training labels, a target scoring network that can evaluate the health status of equipment in real time is trained to more comprehensively and accurately assess the health status of the robot.

[0049] In practical implementation, other types of deep learning models, such as GRU (Gated Recurrent Unit) networks or CNN (Convolutional Neural Network) combined with LSTM, can be introduced to meet the prediction needs of different types of faults.

[0050] To mitigate the effects of equipment aging, in one optional implementation, step S203 includes:

[0051] Step S2031: Determine the degree of impact of the fault on the performance of the industrial robot based on the fault history, and obtain the correction coefficient;

[0052] Specifically, all records related to industrial robot malfunctions are collected, including malfunction type, occurrence time, malfunction duration, and repair cost. Based on the malfunction's impact on equipment performance, a correction factor is assigned to each malfunction type (to measure the loss after repair), and the total correction factor for all malfunctions that have occurred with the equipment is determined.

[0053] Step S2032: Calculate the aging correction amount based on the correction factor and cumulative running time. The formula for calculating the aging correction amount is HI1 = a1 * (1 - e) -0.001t ), where HI1 is the aging correction amount, a1 is the correction coefficient, and t is the cumulative running time.

[0054] Specifically, a mathematical function based on the cumulative operating time of the equipment is introduced to describe the equipment aging process, and the performance degradation caused by faults is further introduced, thus obtaining the above formula.

[0055] Through the above embodiments, by analyzing the cumulative runtime and fault history of the industrial robot, an aging correction amount is determined to dynamically adjust the fault prediction threshold. Specifically, a mathematical model is established to quantify faults and to quantitatively evaluate the performance degradation of the industrial robot, thereby correcting the threshold. The introduction of the aging correction amount allows the fault threshold to adaptively adjust with equipment aging, avoiding the problem of threshold inaccuracy caused by equipment aging and improving the accuracy and reliability of fault prediction.

[0056] To achieve adaptive fault determination threshold, in one optional implementation, step S204 includes:

[0057] Step S2041: The aging correction amount and the preset fault threshold are weighted and summed to obtain the dynamic fault threshold.

[0058] Specifically, in the initial state, the preset fault threshold is a reference value pre-set according to equipment design standards and experimental data, used to determine whether the equipment is in a potential fault state (e.g., 0.75). Then, the preset fault threshold and the above-mentioned aging correction amount are weighted and summed, such as HI_2 = W_1 * HI_base + W_2 * HI1, where W_1 and W_2 are weights, HI_base is the preset fault threshold, and HI_2 is the dynamic fault threshold.

[0059] Through the above embodiments, dynamically adjusted fault thresholds can more accurately reflect the true health status of equipment, avoiding false alarms and missed alarms caused by equipment aging. Accurate fault prediction makes maintenance work more targeted and economical. On the one hand, it reduces unnecessary maintenance operations, directly lowering maintenance costs; on the other hand, it reduces potential downtime losses, indirectly saving production costs.

[0060] To achieve real-time processing, in an optional implementation, step S204 further includes:

[0061] Step S2042: When the equipment failure index is greater than the dynamic failure threshold and less than or equal to the first threshold, a first alarm message is generated. The first alarm message is used to indicate the adjustment of the production schedule of the industrial robot. The ratio of the first threshold to the dynamic failure threshold is a first preset ratio.

[0062] Specifically, three thresholds are preset, including a dynamic fault threshold DT, a first threshold DT1, and a second threshold DT2. When the HI value first exceeds the dynamic fault threshold DT but does not reach DT1, i.e., HI>DT and HI≤DT1, the system generates a first alarm message to prompt the factory manager or automation system to adjust the production schedule to reduce the load on the equipment and prevent further deterioration.

[0063] Step S2043: When the equipment failure index is greater than the first threshold and less than or equal to the second threshold, a second alarm message is generated. The second alarm message is used to indicate that the operation should be stopped when the industrial robot has completed the current workpiece processing. The ratio of the second threshold to the dynamic failure threshold is a second preset ratio.

[0064] Specifically, if the HI value rises above DT1 but does not reach DT2 (i.e., HI>DT1 and HI≤DT2), a second alarm message is generated. At this time, the system will instruct the industrial robot to stop working after completing the processing of the current workpiece, allowing time for equipment inspection and possible maintenance, while minimizing production interruptions.

[0065] In step S2044, if the equipment failure index is greater than the second threshold, a third alarm message is generated, which is used to instruct the industrial robot to stop working immediately.

[0066] Specifically, when the HI value exceeds DT2, i.e. HI>DT2, it means that the equipment may be about to experience a serious malfunction. At this time, the system immediately generates a third alarm message, and the industrial robot must stop working immediately to prevent potential equipment damage and production accidents.

[0067] Through the above embodiments, the tiered early warning mechanism adopts different levels of response measures based on the severity of the fault index, avoiding over-maintenance or ignoring critical warnings, and improving the accuracy of early warnings. Through minor and moderate warnings, the system can adjust production schedules and operating modes in advance, reducing unplanned downtime, maintenance costs, and production losses. Moderate warnings allow for the safe completion of processing of the current workpiece, reducing production interruptions and improving production line continuity and efficiency. Emergency warnings can prevent serious equipment damage, effectively extending equipment lifespan through timely intervention.

[0068] To achieve multimodal data fusion, in one optional implementation, step S201 includes:

[0069] Step S2011: Obtain the timestamp of the preset master clock to get the first timestamp;

[0070] Specifically, in the monitoring system of an industrial robot, a high-precision and stable preset master clock (e.g., the real-time clock (RTC) of an embedded system or an external NTP server) is selected as the reference for all timestamps, i.e., the aforementioned master clock. At the beginning of each data acquisition cycle, the current time of the master clock is recorded to obtain the first timestamp.

[0071] Step S2012: Extract the timestamps of vibration spectrum, current harmonics and temperature gradient to obtain the second timestamp, the third timestamp and the fourth timestamp;

[0072] Specifically, the timestamps of vibration spectrum, current harmonics and temperature gradient data are extracted from each sensor module to obtain the second, third and fourth timestamps.

[0073] Step S2013: Based on the first timestamp, correct the second, third, and fourth timestamps;

[0074] Specifically, based on the first timestamp, a precision synchronization protocol (such as PTP, Precision Time Protocol) is used to correct the timestamps of each sensor to ensure time synchronization between data.

[0075] Step S2014: Based on the corrected second, third, and fourth timestamps, linear interpolation is used to fill in the data gaps generated during the correction process to obtain a multimodal data set.

[0076] Specifically, during the data alignment process, due to the different data acquisition cycles of each sensor, a phenomenon similar to "data gaps" may occur. Therefore, in order to maintain the integrity of the dataset, this application uses linear interpolation to fill in these missing values. Here, linear interpolation assumes that unknown data between two known points can be estimated according to a linear relationship, thereby filling in the missing points.

[0077] Through the above embodiments, timestamp alignment ensures accurate correspondence of multi-source data on the same timeline, eliminates analytical errors caused by time deviations, and improves the accuracy of fault prediction models. Linear interpolation fills in missing values ​​caused by data transmission or sensor failures, making the dataset more continuous and complete, reducing interference during model training and operation, and improving the stability of prediction results.

[0078] To ensure the accuracy of multimodal data, in one optional implementation, step S201 includes:

[0079] Step S2015: Identify missing values ​​in the multimodal data set;

[0080] Specifically, during the data acquisition and preprocessing stage, the system performs an integrity check on each set of multimodal data to identify any missing sensor data and obtain the aforementioned missing values.

[0081] Step S2016: Use time series interpolation to determine the data values ​​corresponding to the missing values, and use the data values ​​to fill the missing values. The time series interpolation includes linear interpolation and polynomial interpolation.

[0082] Specifically, the above linear interpolation assumes that the data changes in a linear trend within the time interval before and after the missing value. Therefore, the missing value can be estimated by linear extrapolation using the values ​​of adjacent data points.

[0083] Furthermore, for missing values ​​in a dataset that exhibit regular changes or complex trends, polynomial interpolation approximates the time series data by fitting a polynomial function. Polynomial interpolation can capture the curve changes of the data, thereby estimating missing values ​​more accurately.

[0084] In practice, more advanced data recovery algorithms, such as machine learning-based methods, can be introduced to fill in missing values ​​in a more intelligent way.

[0085] Step S2017: Determine the equipment failure index of the industrial robot based on the multimodal data set after filling in the missing values.

[0086] Specifically, after filling in the identified missing values ​​using time series interpolation techniques, the original dataset is completed, forming a complete time series dataset. The completed multimodal dataset is then input into a trained fault prediction model (such as an LSTM network) to calculate the equipment failure index (HI) of the industrial robot.

[0087] Through the above embodiments, the supplemented multimodal data sets provide more comprehensive equipment status information, reducing prediction errors caused by data gaps. Even if occasional gaps occur in the data stream, time series interpolation can quickly fill them, ensuring the continuity and real-time nature of fault prediction and avoiding the impact of data gaps on system response speed.

[0088] It should be noted that the steps shown in the flowchart in 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 may be executed in a different order than that shown here.

[0089] This application also provides an adaptive industrial robot fault diagnosis device. It should be noted that the adaptive industrial robot fault diagnosis device of this application can be used to execute the fault diagnosis method for adaptive industrial robots provided in this application. This device is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0090] The following describes the adaptive industrial robot fault diagnosis device provided in the embodiments of this application.

[0091] Figure 3 This is a structural block diagram of an adaptive industrial robot fault diagnosis device according to an embodiment of this application. Figure 3As shown, the device includes:

[0092] The acquisition unit 10 is used to acquire the vibration spectrum, current harmonics and temperature gradient of the industrial robot to obtain a multimodal data set;

[0093] Specifically, vibration spectrum analysis uses a triaxial vibration accelerometer (sampling rate 10kHz) to monitor vibration signals generated by the joint movements of the industrial robot, and extracts spectral features using FFT (Fast Fourier Transform). Current harmonics analysis employs a Hall effect sensor (accuracy ±0.5%) to monitor motor current in real time, calculating the total harmonic distortion (THD) of the current to detect electrical system anomalies. Temperature gradient analysis utilizes an infrared thermal imager (temperature range -20℃ to 300℃) to monitor the temperature distribution in the joint area and identify thermal faults such as overheating.

[0094] The first determining unit 20 is used to determine the equipment failure index of the industrial robot based on the multimodal data set. The equipment failure index is used to quantify the performance loss of the industrial robot.

[0095] Specifically, this application sets up an LSTM network to analyze the temporal changes of multimodal data sets in order to determine the performance loss of industrial robots and obtain the aforementioned equipment failure index (HI value).

[0096] The second determining unit 30 is used to acquire the cumulative running time and fault history of the industrial robot, and determine the aging degree of the industrial robot based on the cumulative running time and fault history to obtain the aging correction amount.

[0097] Specifically, through statistical analysis, the relationship between the cumulative operating time of the equipment and its fault history is determined, and a coefficient representing the degree of equipment aging is calculated, namely the aforementioned aging correction amount. This coefficient reflects the gradual deterioration of the equipment over time and is used as a key parameter for correcting the preset fault threshold.

[0098] The generation unit 40 is used to correct the preset fault threshold by aging correction to obtain a dynamic fault threshold. When the equipment fault index is greater than the dynamic fault threshold, alarm information is generated. The alarm information is used to indicate that there is a fault risk in the industrial robot.

[0099] Specifically, by determining and correcting the impact of aging on the performance of industrial robots, the threshold for fault diagnosis is changed to avoid misjudgment due to performance loss. This results in the aforementioned dynamic fault threshold. Furthermore, by comparing the equipment fault index with the aforementioned dynamic fault threshold, an alarm message is generated to determine the existence of a fault risk when the fault index is greater than the fault threshold.

[0100] In this embodiment, the acquisition unit acquires the vibration spectrum, current harmonics, and temperature gradient of the industrial robot to obtain a multimodal data set; the first determination unit determines the equipment failure index of the industrial robot based on the multimodal data set, and the equipment failure index is used to quantify the performance loss of the industrial robot; the second determination unit acquires the cumulative runtime and fault history of the industrial robot, and determines the aging degree of the industrial robot based on the cumulative runtime and fault history to obtain an aging correction amount; the generation unit uses the aging correction amount to correct a preset fault threshold to obtain a dynamic fault threshold. When the equipment failure index is greater than the dynamic fault threshold, an alarm message is generated to indicate that the industrial robot has a fault risk. This application introduces the impact of equipment aging on the performance of industrial robots, quantifies it through the aging effect of cumulative runtime and faults, corrects the fault diagnosis threshold, achieves adaptive fault diagnosis, and uses a static threshold model to ensure the accuracy of fault diagnosis in the absence of false alarms after equipment aging, thus solving the problem of low accuracy in fault diagnosis of industrial robots using static threshold models in the prior art.

[0101] To determine the aforementioned equipment failure index, in one optional implementation, the first determining unit includes:

[0102] The first acquisition module is used to acquire historical vibration spectrum, historical current harmonics and historical temperature gradient from the equipment log of the industrial robot to obtain historical operating data set;

[0103] Specifically, vibration spectrum, current harmonics, and temperature gradient data recorded during past operations are extracted from the equipment logs of industrial robots to cover the equipment's performance under various normal and fault conditions, resulting in the aforementioned historical operating data set.

[0104] The first determination module is used to determine the equipment failure index corresponding to the historical operating data group based on the equipment log, and obtain the historical failure index.

[0105] Specifically, based on historical fault records, fault index labels are assigned to the aforementioned historical operating data groups. For example, the HI value of the equipment before a minor fault is labeled as 1.05, while it may be labeled as 1.3 or higher before a serious fault.

[0106] The training module is used to train a long short-term memory network with historical operation data sets as input data and historical failure indices as label data to obtain the target scoring network.

[0107] Specifically, historical vibration spectrum, current harmonics, and temperature gradient data are standardized to eliminate the influence of dimensions. Using the processed historical operating data set as input and historical failure index as output label, an LSTM network is trained to predict the equipment failure index of the equipment based on multimodal data.

[0108] The input module is used to input multimodal data sets into the target scoring network to obtain the equipment failure index.

[0109] Specifically, real-time multimodal data sets are input into a trained LSTM network, which outputs a real-time device failure index to instantly determine the health status of the device.

[0110] Through the above embodiments, by fusing multi-source data (vibration spectrum, current harmonics, temperature gradient), a prediction model for the equipment failure index (HI) is constructed using a long short-term memory network (LSTM). The LSTM neural network is used to perform deep learning on historical operating data to mine the intrinsic relationships between data. Using the historical failure index as training labels, a target scoring network that can evaluate the health status of equipment in real time is trained to more comprehensively and accurately assess the health status of the robot.

[0111] In practical implementation, other types of deep learning models, such as GRU (Gated Recurrent Unit) networks or CNN (Convolutional Neural Network) combined with LSTM, can be introduced to meet the prediction needs of different types of faults.

[0112] To mitigate the effects of equipment aging, in one optional implementation, the second determining unit includes:

[0113] The second determination module is used to determine the degree of impact of the fault on the performance of the industrial robot based on the fault history, and to obtain the correction coefficient.

[0114] Specifically, all records related to industrial robot malfunctions are collected, including malfunction type, occurrence time, malfunction duration, and repair cost. Based on the malfunction's impact on equipment performance, a correction factor is assigned to each malfunction type (to measure the loss after repair), and the total correction factor for all malfunctions that have occurred with the equipment is determined.

[0115] The first calculation module is used to calculate the aging correction amount based on the correction coefficient and the cumulative running time. The formula for calculating the aging correction amount is HI1=a1*(1-e -0.001t ), where HI1 is the aging correction amount, a1 is the correction coefficient, and t is the cumulative running time.

[0116] Specifically, a mathematical function based on the cumulative operating time of the equipment is introduced to describe the equipment aging process, and the performance degradation caused by faults is further introduced, thus obtaining the above formula.

[0117] Through the above embodiments, by analyzing the cumulative runtime and fault history of the industrial robot, an aging correction amount is determined to dynamically adjust the fault prediction threshold. Specifically, a mathematical model is established to quantify faults and to quantitatively evaluate the performance degradation of the industrial robot, thereby correcting the threshold. The introduction of the aging correction amount allows the fault threshold to adaptively adjust with equipment aging, avoiding the problem of threshold inaccuracy caused by equipment aging and improving the accuracy and reliability of fault prediction.

[0118] To achieve adaptive fault determination threshold, in one optional implementation, the generation unit includes:

[0119] The second calculation module is used to perform a weighted summation of the aging correction amount and the preset fault threshold to obtain the dynamic fault threshold.

[0120] Specifically, in the initial state, the preset fault threshold is a reference value pre-set according to equipment design standards and experimental data, used to determine whether the equipment is in a potential fault state (e.g., 0.75). Then, the preset fault threshold and the above-mentioned aging correction amount are weighted and summed, such as HI_2=W_1*HI_base+W_2*HI_1.

[0121] Through the above embodiments, dynamically adjusted fault thresholds can more accurately reflect the true health status of equipment, avoiding false alarms and missed alarms caused by equipment aging. Accurate fault prediction makes maintenance work more targeted and economical. On the one hand, it reduces unnecessary maintenance operations, directly lowering maintenance costs; on the other hand, it reduces potential downtime losses, indirectly saving production costs.

[0122] To achieve real-time processing, in one optional implementation, the generation unit further includes:

[0123] The first generation module is used to generate a first alarm message when the equipment failure index is greater than the dynamic failure threshold and less than or equal to the first threshold. The first alarm message is used to indicate the adjustment of the production schedule of the industrial robot. The ratio of the first threshold to the dynamic failure threshold is a first preset ratio.

[0124] Specifically, three thresholds are preset, including a dynamic fault threshold DT, a first threshold DT1, and a second threshold DT2. When the HI value first exceeds the dynamic fault threshold DT but does not reach DT1, i.e., HI>DT and HI≤DT1, the system generates a first alarm message to prompt the factory manager or automation system to adjust the production schedule to reduce the load on the equipment and prevent further deterioration.

[0125] The second generation module is used to generate a second alarm message when the equipment failure index is greater than the first threshold and less than or equal to the second threshold. The second alarm message is used to indicate that the operation should be stopped when the industrial robot has completed the current workpiece processing. The ratio of the second threshold to the dynamic failure threshold is a second preset ratio.

[0126] Specifically, if the HI value rises above DT1 but does not reach DT2 (i.e., HI>DT1 and HI≤DT2), a second alarm message is generated. At this time, the system will instruct the industrial robot to stop working after completing the processing of the current workpiece, allowing time for equipment inspection and possible maintenance, while minimizing production interruptions.

[0127] The third generation module is used to generate a third alarm message when the equipment failure index is greater than the second threshold. The third alarm message is used to instruct the industrial robot to stop working immediately.

[0128] Specifically, when the HI value exceeds DT2, i.e. HI>DT2, it means that the equipment may be about to experience a serious malfunction. At this time, the system immediately generates a third alarm message, and the industrial robot must stop working immediately to prevent potential equipment damage and production accidents.

[0129] Through the above embodiments, the tiered early warning mechanism adopts different levels of response measures based on the severity of the fault index, avoiding over-maintenance or ignoring critical warnings, and improving the accuracy of early warnings. Through minor and moderate warnings, the system can adjust production schedules and operating modes in advance, reducing unplanned downtime, maintenance costs, and production losses. Moderate warnings allow for the safe completion of processing of the current workpiece, reducing production interruptions and improving production line continuity and efficiency. Emergency warnings can prevent serious equipment damage, effectively extending equipment lifespan through timely intervention.

[0130] To achieve multimodal data fusion, in one optional implementation, the acquisition unit includes:

[0131] The second acquisition module is used to acquire the timestamp of the preset master clock to obtain the first timestamp;

[0132] Specifically, in the monitoring system of an industrial robot, a high-precision and stable preset master clock (e.g., the real-time clock (RTC) of an embedded system or an external NTP server) is selected as the reference for all timestamps, i.e., the aforementioned master clock. At the beginning of each data acquisition cycle, the current time of the master clock is recorded to obtain the first timestamp.

[0133] The third acquisition module is used to extract the timestamps of vibration spectrum, current harmonics and temperature gradient to obtain the second timestamp, the third timestamp and the fourth timestamp;

[0134] Specifically, the timestamps of vibration spectrum, current harmonics and temperature gradient data are extracted from each sensor module to obtain the second, third and fourth timestamps.

[0135] The first processing module is used to correct the second, third and fourth timestamps based on the first timestamp.

[0136] Specifically, based on the first timestamp, a precision synchronization protocol (such as PTP, Precision Time Protocol) is used to correct the timestamps of each sensor to ensure time synchronization between data.

[0137] The second processing module is used to fill in the data gaps generated during the correction process using linear interpolation based on the corrected second, third, and fourth timestamps, to obtain a multimodal data set.

[0138] Specifically, during the data alignment process, due to the different data acquisition cycles of each sensor, a phenomenon similar to "data gaps" may occur. Therefore, in order to maintain the integrity of the dataset, this application uses linear interpolation to fill in these missing values. Here, linear interpolation assumes that unknown data between two known points can be estimated according to a linear relationship, thereby filling in the missing points.

[0139] Through the above embodiments, timestamp alignment ensures accurate correspondence of multi-source data on the same timeline, eliminates analytical errors caused by time deviations, and improves the accuracy of fault prediction models. Linear interpolation fills in missing values ​​caused by data transmission or sensor failures, making the dataset more continuous and complete, reducing interference during model training and operation, and improving the stability of prediction results.

[0140] To ensure the accuracy of multimodal data, in one optional implementation, the acquisition unit includes:

[0141] The identification module is used to identify missing values ​​in multimodal datasets;

[0142] Specifically, during the data acquisition and preprocessing stage, the system performs an integrity check on each set of multimodal data to identify any missing sensor data and obtain the aforementioned missing values.

[0143] The third processing module is used to determine the data values ​​corresponding to the missing values ​​using time series interpolation, and to fill the missing values ​​using the data values. The time series interpolation includes linear interpolation and polynomial interpolation.

[0144] Specifically, the above linear interpolation assumes that the data changes in a linear trend within the time interval before and after the missing value. Therefore, the missing value can be estimated by linear extrapolation using the values ​​of adjacent data points.

[0145] Furthermore, for missing values ​​in a dataset that exhibit regular changes or complex trends, polynomial interpolation approximates the time series data by fitting a polynomial function. Polynomial interpolation can capture the curve changes of the data, thereby estimating missing values ​​more accurately.

[0146] In practice, more advanced data recovery algorithms, such as machine learning-based methods, can be introduced to fill in missing values ​​in a more intelligent way.

[0147] The fourth processing module is used to determine the equipment failure index of the industrial robot based on the multimodal data set after filling in the missing values.

[0148] Specifically, after filling in the identified missing values ​​using time series interpolation techniques, the original dataset is completed, forming a complete time series dataset. The completed multimodal dataset is then input into a trained fault prediction model (such as an LSTM network) to calculate the equipment failure index (HI) of the industrial robot.

[0149] Through the above embodiments, the supplemented multimodal data sets provide more comprehensive equipment status information, reducing prediction errors caused by data gaps. Even if occasional gaps occur in the data stream, time series interpolation can quickly fill them, ensuring the continuity and real-time nature of fault prediction and avoiding the impact of data gaps on system response speed.

[0150] The aforementioned adaptive industrial robot fault diagnosis device includes a processor and a memory. The acquisition unit, first determination unit, second determination unit, and generation unit are all stored as program units in the memory. The processor executes these program units stored in the memory to achieve the corresponding functions. All of the above modules reside in the same processor; alternatively, the modules may be located in different processors in any combination.

[0151] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and adjusting kernel parameters can improve the accuracy of fault diagnosis for industrial robots.

[0152] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0153] This invention provides a computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the adaptive industrial robot fault diagnosis method.

[0154] Specifically, adaptive fault diagnosis methods for industrial robots include:

[0155] Step S201: Obtain the vibration spectrum, current harmonics, and temperature gradient of the industrial robot to obtain a multimodal data set;

[0156] Step S202: Determine the equipment failure index of the industrial robot based on the multimodal data set. The equipment failure index is used to quantify the performance loss of the industrial robot.

[0157] Step S203: Obtain the cumulative runtime and fault history of the industrial robot, and determine the aging degree of the industrial robot based on the cumulative runtime and fault history to obtain the aging correction amount.

[0158] Step S204: The preset fault threshold is corrected by aging correction to obtain a dynamic fault threshold. When the equipment fault index is greater than the dynamic fault threshold, an alarm message is generated. The alarm message is used to indicate that there is a fault risk in the industrial robot.

[0159] This invention provides a processor for running a program, wherein the program executes the adaptive industrial robot fault diagnosis method.

[0160] Specifically, adaptive fault diagnosis methods for industrial robots include:

[0161] Step S201: Obtain the vibration spectrum, current harmonics, and temperature gradient of the industrial robot to obtain a multimodal data set;

[0162] Step S202: Determine the equipment failure index of the industrial robot based on the multimodal data set. The equipment failure index is used to quantify the performance loss of the industrial robot.

[0163] Step S203: Obtain the cumulative runtime and fault history of the industrial robot, and determine the aging degree of the industrial robot based on the cumulative runtime and fault history to obtain the aging correction amount.

[0164] Step S204: The preset fault threshold is corrected by aging correction to obtain a dynamic fault threshold. When the equipment fault index is greater than the dynamic fault threshold, an alarm message is generated. The alarm message is used to indicate that there is a fault risk in the industrial robot.

[0165] This invention provides an industrial robot, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs at least the following steps:

[0166] Step S201: Obtain the vibration spectrum, current harmonics, and temperature gradient of the industrial robot to obtain a multimodal data set;

[0167] Step S202: Determine the equipment failure index of the industrial robot based on the multimodal data set. The equipment failure index is used to quantify the performance loss of the industrial robot.

[0168] Step S203: Obtain the cumulative runtime and fault history of the industrial robot, and determine the aging degree of the industrial robot based on the cumulative runtime and fault history to obtain the aging correction amount.

[0169] Step S204: The preset fault threshold is corrected by aging correction to obtain a dynamic fault threshold. When the equipment fault index is greater than the dynamic fault threshold, an alarm message is generated. The alarm message is used to indicate that there is a fault risk in the industrial robot.

[0170] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program having at least the following method steps:

[0171] Step S201: Obtain the vibration spectrum, current harmonics, and temperature gradient of the industrial robot to obtain a multimodal data set;

[0172] Step S202: Determine the equipment failure index of the industrial robot based on the multimodal data set. The equipment failure index is used to quantify the performance loss of the industrial robot.

[0173] Step S203: Obtain the cumulative runtime and fault history of the industrial robot, and determine the aging degree of the industrial robot based on the cumulative runtime and fault history to obtain the aging correction amount.

[0174] Step S204: The preset fault threshold is corrected by aging correction to obtain a dynamic fault threshold. When the equipment fault index is greater than the dynamic fault threshold, an alarm message is generated. The alarm message is used to indicate that there is a fault risk in the industrial robot.

[0175] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0176] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0177] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0178] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0179] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0180] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0181] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0182] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0183] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0184] It should also be noted that 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 process, method, article, or apparatus. Unless otherwise specified, 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 that element.

[0185] As can be seen from the above description, the embodiments of this application achieve the following technical effects:

[0186] 1) The adaptive fault diagnosis method for industrial robots in this application first acquires the vibration spectrum, current harmonics, and temperature gradient of the industrial robot to obtain a multimodal data set. Then, based on the multimodal data set, the equipment fault index of the industrial robot is determined, which is used to quantify the performance loss of the industrial robot. Next, the cumulative runtime and fault history of the industrial robot are acquired, and the aging degree of the industrial robot is determined based on the cumulative runtime and fault history to obtain an aging correction amount. Finally, the aging correction amount is used to correct a preset fault threshold to obtain a dynamic fault threshold. When the equipment fault index is greater than the dynamic fault threshold, an alarm message is generated, which indicates that the industrial robot has a fault risk. This application introduces the impact of equipment aging on the performance of industrial robots, quantifies it through the aging effect of cumulative runtime and faults, corrects the fault diagnosis threshold, and achieves adaptive fault diagnosis. The use of a static threshold model eliminates false alarms after equipment aging, ensuring the accuracy of fault diagnosis and solving the problem of low accuracy in fault diagnosis of industrial robots using static threshold models in existing technologies.

[0187] 2) The adaptive industrial robot fault diagnosis device of this application acquires the vibration spectrum, current harmonics, and temperature gradient of the industrial robot to obtain a multimodal data set; the first determining unit determines the equipment fault index of the industrial robot based on the multimodal data set, which is used to quantify the performance loss of the industrial robot; the second determining unit acquires the cumulative running time and fault history of the industrial robot, and determines the aging degree of the industrial robot based on the cumulative running time and fault history to obtain an aging correction amount; the generating unit uses the aging correction amount to correct a preset fault threshold to obtain a dynamic fault threshold. When the equipment fault index is greater than the dynamic fault threshold, an alarm message is generated to indicate that the industrial robot has a fault risk. This application introduces the impact of equipment aging on the performance of the industrial robot, quantifies it through the aging impact of cumulative running time and faults, corrects the fault diagnosis threshold, and achieves adaptive fault diagnosis. By using a static threshold model, false alarms after equipment aging are eliminated, ensuring the accuracy of fault diagnosis and solving the problem of low accuracy in fault diagnosis of industrial robots using static threshold models in the prior art.

[0188] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A failure diagnosis method of an adaptive industrial robot, characterized by, The method comprises the following steps: obtaining a vibration spectrum, a current harmonic and a temperature gradient of an industrial robot to obtain a multi-modal data set; determining a device fault index of the industrial robot according to the multi-modal data set, the device fault index being used to quantify a performance loss of the industrial robot; obtaining a cumulative running time and a fault history of the industrial robot, and determining an aging degree of the industrial robot according to the cumulative running time and the fault history to obtain an aging correction quantity; correcting a preset fault threshold by using the aging correction quantity to obtain a dynamic fault threshold, and generating an alarm information in a case that the device fault index is greater than the dynamic fault threshold, the alarm information being used to indicate that the industrial robot has a fault risk.

2. The method of claim 1, wherein, The method comprises the following steps: obtaining a vibration spectrum, a current harmonic and a temperature gradient of an industrial robot to obtain a multi-modal data set; determining a device fault index of the industrial robot according to the multi-modal data set, the device fault index being used to quantify a performance loss of the industrial robot; obtaining a cumulative running time and a fault history of the industrial robot, and determining an aging degree of the industrial robot according to the cumulative running time and the fault history to obtain an aging correction quantity; correcting a preset fault threshold by using the aging correction quantity to obtain a dynamic fault threshold, and generating an alarm information in a case that the device fault index is greater than the dynamic fault threshold, the alarm information being used to indicate that the industrial robot has a fault risk.

3. The method of claim 1, wherein, The method comprises the following steps: obtaining a vibration spectrum, a current harmonic and a temperature gradient of an industrial robot to obtain a multi-modal data set; According to the correction coefficient and the cumulative running time, the aging correction amount is calculated, wherein a calculation formula of the aging correction amount is HI1=a1*(1-e -0.001t ), wherein HI1 is the aging correction amount, a1 is the correction coefficient, and t is the cumulative running time.

4. The method of claim 1, wherein, determining a device fault index of the industrial robot according to the multi-modal data set, the device fault index being used to quantify a performance loss of the industrial robot; obtaining a cumulative running time and a fault history of the industrial robot, and determining an aging degree of the industrial robot according to the cumulative running time and the fault history to obtain an aging correction quantity; 5. The method according to any one of claims 1 to 4, characterized in that, correcting a preset fault threshold by using the aging correction quantity to obtain a dynamic fault threshold, and generating an alarm information in a case that the device fault index is greater than the dynamic fault threshold, the alarm information being used to indicate that the industrial robot has a fault risk. The method comprises the following steps: obtaining a vibration spectrum, a current harmonic and a temperature gradient of an industrial robot to obtain a multi-modal data set; determining a device fault index of the industrial robot according to the multi-modal data set, the device fault index being used to quantify a performance loss of the industrial robot; 6. The method according to any one of claims 1 to 4, characterized in that, obtaining a cumulative running time and a fault history of the industrial robot, and determining an aging degree of the industrial robot according to the cumulative running time and the fault history to obtain an aging correction quantity; correcting a preset fault threshold by using the aging correction quantity to obtain a dynamic fault threshold, and generating an alarm information in a case that the device fault index is greater than the dynamic fault threshold, the alarm information being used to indicate that the industrial robot has a fault risk. The method comprises the following steps: obtaining a vibration spectrum, a current harmonic and a temperature gradient of an industrial robot to obtain a multi-modal data set; determining a device fault index of the industrial robot according to the multi-modal data set, the device fault index being used to quantify a performance loss of the industrial robot; obtaining a cumulative running time and a fault history of the industrial robot, and determining an aging degree of the industrial robot according to the cumulative running time and the fault history to obtain an aging correction quantity; correcting a preset fault threshold by using the aging correction quantity to obtain a dynamic fault threshold, and generating an alarm information in a case that the device fault index is greater than the dynamic fault threshold, the alarm information being used to indicate that the industrial robot has a fault risk. The method comprises the following steps: obtaining a vibration spectrum, a current harmonic and a temperature gradient of an industrial robot to obtain a multi-modal data set; determining a device fault index of the industrial robot according to the multi-modal data set, the device fault index being used to quantify a performance loss of the industrial robot; obtaining a cumulative running time and a fault history of the industrial robot, and determining an aging degree of the industrial robot according to the cumulative running time and the fault history to obtain an aging correction quantity; correcting a preset fault threshold by using the aging correction quantity to obtain a dynamic fault threshold, and generating an alarm information in a case that the device fault index is greater than the dynamic fault threshold, the alarm information being used to indicate that the industrial robot has a fault risk. correcting the second timestamp, the third timestamp and the fourth timestamp based on the first timestamp; filling in data missing in the correction process based on the corrected second timestamp, the third timestamp and the fourth timestamp by using a linear interpolation method to obtain the multi-modal data set.

7. The method according to any one of claims 1 to 4, characterized in that, determining the equipment failure index of the industrial robot according to the multi-modal data set, including: identifying missing values in the multi-modal data set; determining data values corresponding to the missing values by using time series interpolation, and filling the missing values with the data values, wherein the time series interpolation includes linear interpolation and polynomial interpolation; determining the equipment failure index of the industrial robot according to the multi-modal data set after filling in the missing values.

8. An adaptive failure diagnosis device for an industrial robot, characterized by The apparatus includes: an acquisition unit configured to acquire a vibration frequency spectrum, a current harmonic and a temperature gradient of an industrial robot to obtain a multi-modal data set; a first determination unit configured to determine an equipment failure index of the industrial robot according to the multi-modal data set, the equipment failure index being used to quantify performance loss of the industrial robot; a second determination unit configured to acquire a cumulative running time and a failure history of the industrial robot, and determine an aging degree of the industrial robot according to the cumulative running time and the failure history to obtain an aging correction amount; a generation unit configured to correct a preset failure threshold by using the aging correction amount to obtain a dynamic failure threshold, and generate an alarm information in a case where the equipment failure index is greater than the dynamic failure threshold, the alarm information being used to indicate that the industrial robot has a failure risk.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program controls a device where the computer-readable storage medium is located to execute the method of any one of claims 1 to 7 when the program is running.

10. An industrial robot, characterized by including: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include a program for executing the method of any one of claims 1 to 7.

Images