Predictive maintenance method and system

Abstract

Embodiments of the present application provide a predictive maintenance method and system. The method comprises: a first device respectively acquires first running state data of each device in a system based on a first task queue; the first device determines whether a corresponding device is running according to the first running state data, and if so, sends a corresponding second sampling task to a second device, a sampling duration of the second sampling task being greater than a sampling duration of the corresponding first sampling task; the second device receives the second sampling task, samples the corresponding device based on the second sampling task, and feeds back second running state data of the sampled corresponding device to the first device; and the first device determines whether the corresponding device is faulty according to the second running state data. Embodiments of the present application can perform predictive maintenance on randomly running devices, and have high reliability.

Description

Technical Field

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

[0002] Existing production systems (such as production line systems) typically achieve production through the coordinated operation of various different devices, and the operating time of some devices in the system is irregular, with many operating randomly. Currently, predictive maintenance can only be performed on devices that are in operation; there are no existing methods for predictive maintenance of randomly operating devices.

[0003] Moreover, existing technologies typically use one device to predictively maintain one device at a time, which is costly and inefficient for systems with many devices. Furthermore, existing technologies lack an effective control scheme for one device to perform predictive maintenance on multiple devices simultaneously. Summary of the Invention

[0004] In view of this, the present invention provides a predictive maintenance method and system for at least partially solving the above-mentioned technical problems.

[0005] Firstly, a predictive maintenance method is provided, the method comprising:

[0006] The first device acquires the first operating status data of each device in the system at regular intervals based on the first task queue, wherein the first task queue includes the first sampling task corresponding to the sampling timing time of each device, and the first sampling task includes the first sampling signal having a first sampling duration and a first sampling precision.

[0007] The first device determines whether the corresponding device is running based on the first operating status data. If so, it sends a corresponding second sampling task to a second device. The second sampling task includes a second sampling signal with a second sampling duration and a second sampling precision. The second sampling duration is greater than the first sampling duration.

[0008] The second device that receives the second sampling task samples the corresponding device based on the second sampling task, and feeds back the sampled second operating status data of the corresponding device to the first device;

[0009] The first device determines whether the corresponding device is faulty based on the second operating status data.

[0010] In one possible implementation, the first device periodically acquires first operating status data of each device in the system based on a first task queue, further including:

[0011] The first device sends a first sampling task corresponding to the first device to at least one second device based on the first task queue;

[0012] The second device that receives the first sampling task samples the corresponding device based on the first sampling task and feeds back the first operating status data of the sampled device to the first device.

[0013] In one possible implementation, the first device periodically acquires first operating status data of each device in the system based on a first task queue, further including:

[0014] The first device directly collects the first operating status data of each device in the system at regular intervals based on the first task queue, and when it detects that any first sampling task in the first task queue is delayed, it distributes the first sampling task of the corresponding device to at least one second device based on the first task queue.

[0015] The second device that receives the first sampling task samples the corresponding device based on the first sampling task and feeds back the first operating status data of the sampled device to the first device.

[0016] In one possible implementation, the second sampling precision is greater than the first sampling precision;

[0017] The first device determines whether the corresponding device is running based on the first operating status data. If so, it sends a second sampling task to a second device. The method further includes: the first device determines whether the corresponding device is running and whether it is in a predetermined working condition based on the first operating status data. If the corresponding device is running and in a predetermined working condition, it sends a second sampling task matching the predetermined working condition to a second device.

[0018] The first device's determination of whether the corresponding device is faulty based on the second operating status data further includes: the first device comparing the second operating status data with target status data that matches the predetermined operating condition, and determining whether the corresponding device is faulty based on the comparison result.

[0019] In one possible implementation, the second sampling precision is less than the first sampling precision;

[0020] The first device determines whether the corresponding device is running based on the first operating status data. If so, it sends a second sampling task to a second device. The method further includes: the first device determines whether the corresponding device is running and whether there is an abnormality based on the first operating status data. If the device is running and there is an abnormality, it sends a second sampling task to a second device.

[0021] In one possible implementation, before the first device periodically acquires the first operating status data of each device in the system based on the first task queue, the method further includes:

[0022] When the first device detects that the device is accessing the system for the first time, it sends a third sampling task to a second device. The third sampling task includes a third sampling signal with a third sampling duration and a third sampling precision. The third sampling duration is longer than the second sampling duration, and the third sampling precision is greater than the second sampling precision and greater than the first sampling precision.

[0023] The second device samples the corresponding device based on the third sampling task and stores the operating status data of all the sampled operating conditions of the corresponding device in the database;

[0024] The first device generates, in the database, the first sampling task, the second sampling task, and the target status data corresponding to different operating conditions based on the operating status data of all operating conditions of the device.

[0025] In one possible implementation, there are multiple second devices, each of which is configured with a corresponding second task queue;

[0026] When the first device sends the first sampling task, the second sampling task, or the third sampling task to the second device, the method further includes: the first device preferentially sends the first sampling task, the second sampling task, or the third sampling task to the second device with the fewest tasks in the second task queue;

[0027] When the second device receives the first sampling task, the second sampling task, or the third sampling task, the method further includes: the second device adding the received first sampling task, the second sampling task, or the third sampling task to a second task queue and reordering them according to the priority level of each sampling task in the second task queue, wherein the priority of the second sampling task is higher than the priority level of the first sampling task and the priority of the third sampling task.

[0028] In one possible implementation, the first device periodically acquires first operating status data of each device in the system based on a first task queue, further including:

[0029] For each of the aforementioned devices, the first device is based on the sampling timing t of the first task queue in the current sampling. n Acquire the first operating status data of the device in the system;

[0030] According to the sampling timing t of the current sampling of the device n Calculate the sampling timing t for the next sampling based on the current historical maintenance status information. n+1 The historical maintenance status information includes at least the device’s historical usage time, expected remaining lifespan, continuous trouble-free operation time, mean time between failures (MTBF), and post-maintenance usage time.

[0031] Based on the calculated sampling timing t of the device for the next sampling. n+1 Add the corresponding first sampling task to the first task queue.

[0032] In one possible implementation, the sampling timing time of the next sampling... Where a is the set value and a≥ the minimum sampling frequency of the device, b is the historical usage time of the device, c is the expected remaining lifespan of the device, d is the continuous fault-free operation time of the device, e is the historical average fault-free operation time of the device, g is the post-maintenance usage time of the device, and w1, w2, w3 represent weighting coefficients and w1+w2+w3=1.

[0033] In one possible implementation, the method further includes:

[0034] When the first device detects that the device is connected to the system for the first time, it obtains the device's historical fault time, historical maintenance time, maintenance cycle, and historical usage time to obtain initial historical maintenance status information. Among them, the initial historical mean time between failures is calculated based on the historical fault time, the initial continuous fault-free operation time is calculated based on the current time and the historical fault time, the initial expected remaining life is calculated based on the theoretical life and the historical usage time, and the initial post-maintenance usage time is calculated based on the historical maintenance time and the maintenance cycle.

[0035] In one possible implementation, the method further includes:

[0036] The historical usage time, expected remaining lifespan, continuous trouble-free operation time, and post-maintenance usage time of each device are periodically updated to update the historical maintenance status information, wherein the timing of the periodic update for each device is determined based on the corresponding sampling timing time; and...

[0037] When the device is determined to be faulty, the continuous fault-free operating time in the historical maintenance status information is cleared, and the historical average fault-free operating time is recalculated using the current fault time to update the historical maintenance status information.

[0038] Secondly, a predictive maintenance system is provided, comprising a first device and at least one second device; wherein,

[0039] The first device includes:

[0040] First interface unit;

[0041] The first sampling unit is configured to periodically acquire first operating status data of each device in the system based on a first task queue via the first interface unit. When a determination result indicating that a corresponding device is running is obtained, the unit sends a corresponding second sampling task to a second device via the first interface unit and acquires second operating status data of the corresponding device from the second device via the first interface unit. The first task queue includes first sampling tasks with corresponding sampling timings for each device. The first sampling task includes a first sampling signal with a first sampling duration and a first sampling precision. The second sampling task includes a second sampling signal with a second sampling duration and a second sampling precision. The second sampling duration is longer than the first sampling duration.

[0042] The first processing unit is used to determine whether the corresponding device is running based on the first operating status data, and to determine whether the device is malfunctioning based on the second operating status data of the corresponding device.

[0043] The second device includes:

[0044] Second interface unit;

[0045] The second sampling unit is used to sample the corresponding device based on the received second sampling task via the second interface unit to obtain the second operating status data of the corresponding device, and to feed back the second operating status data to the first device via the second interface unit.

[0046] In the embodiments of this application, by using a first device and at least one second device, for each device in the system, a short-term predictive sampling is performed before fault judgment sampling, and the decision on whether to perform fault judgment sampling is made based on the short-term predictive sampling result. This enables predictive maintenance of randomly operating devices with reliable results. Specifically, during short-term predictive sampling, the first device periodically acquires the first operating status data of each device in the system based on a first task queue, determines whether the device is running based on the first operating status data, and immediately sends a second sampling task to the second device to perform fault judgment sampling if the device is determined to be running. Furthermore, considering that different devices have different operating conditions and the same device may operate under different operating conditions at different times, before sending the sampling task, the corresponding sampling task is dynamically matched and sent to the second device based on the current different operating condition characteristics of the device. This enables more efficient predictive maintenance of randomly operating devices with higher reliability.

[0047] Furthermore, a new formula for calculating the sampling timing time is proposed. For each device, the sampling timing time of the next sampling is calculated based on the current sampling timing time using this formula. The first sampling task of the next sampling is then added to the first task queue according to the calculated sampling timing time. This allows for adaptive configuration of the sampling timing time for different devices, enabling orderly and efficient predictive maintenance of randomly operating devices with high reliability. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating a predictive maintenance method according to an embodiment of the present invention.

[0049] Figure 2 A schematic diagram of the structure of a predictive maintenance system according to an embodiment of the present invention is shown.

[0050] Figure 3 A flowchart illustrating a sampling process of a device by a first device through a second device is shown in one embodiment.

[0051] Figure 4 A flowchart illustrating the sampling process of a first device through a second device for a device accessing the system for the first time, according to one embodiment, is shown.

[0052] Figure 5 This is a schematic diagram of a predictive maintenance system according to an embodiment of the present invention.

[0053] List of reference numerals in the attached diagram:

[0054] 210: First device; 211: First interface unit

[0055] 212: First sampling unit; 213: First processing unit

[0056] 220: Second device; 221: Second interface unit

[0057] 222: Second sampling unit 231~23N: Device 1~Device N

[0058] 240: Database Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided with reference to the accompanying drawings and embodiments. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. All other technical solutions obtained by those skilled in the art based on the embodiments of this application fall within the scope of protection of this application.

[0060] Figure 1 A schematic flowchart of a predictive maintenance method according to an embodiment of the present invention is shown; Figure 2 A schematic diagram of the predictive maintenance system architecture according to an embodiment of the present invention is shown, as follows: Figure 2 As shown, the system includes a first device 210 and at least one second device 220. Figure 2 As shown, the first device 210 can establish communication connections with each device (231-23N) to directly sample the operating status data of each device (231-23N) in the system, and can also establish communication connections with at least one second device 220 to sample the operating status data of each device (231-23N) in the system through the second device 220. Optionally, the system may further include a database 240, which is communicatively connected to the first device 210 and the second device 220 respectively, for storing large-scale sampled data from the first device 210 and / or the second device 220.

[0061] The following is based on Figure 2 Detailed description of predictive maintenance systems Figure 1 The specific content of predictive maintenance methods, such as Figure 1 As shown, the method includes:

[0062] S110: The first device 210 acquires the first operating status data of each device (231-23N) in the system at regular intervals based on the first task queue. The first task queue includes the first sampling task of each device (231-23N) with a corresponding sampling time, and the first sampling task includes a first sampling signal with a first sampling duration and a first sampling precision.

[0063] Optionally, in the aforementioned step S110, the first device 210 may directly sample the first operating status data of each device (231 to 23N) in the system at regular intervals based on the first task queue, for example, when any first sampling task in the current first task queue will not be delayed.

[0064] Optionally, the first device 210 may also use the second device 220 to periodically sample the first operating status data of each device (231-23N) in the system based on the first task queue, as described in the following example. Figure 3 Specifically, the process includes: ① A first device 210 sends a first sampling task for the corresponding device 23M (M is an integer ≥1 and ≤N) to a second device 220 based on a first task queue; ② The second device 220, upon receiving the first sampling task, samples the corresponding device 23M based on the first sampling task; ③ The first operating status data of the sampled device is fed back to the first device 210; ④ The first device 210 determines that the corresponding device 23M is running based on the first operating status data and sends a corresponding second sampling task to the second device 220; ⑤ The second device 220, upon receiving the second sampling task, samples the corresponding device 23M based on the second sampling task; ⑥ The second operating status data of the sampled corresponding device 23M is fed back to the first device 210.

[0065] Optionally, the first device 210 may also directly sample the first operating status data of each device (231-23N) in the system in some cases, and sample the first operating status data of each device (231-23N) in the system through the second device 220 in other cases. For example, the first device 210 may directly collect the first operating status data of each device (231-23N) in the system at regular intervals based on the first task queue, and when it is detected that any first sampling task in the first task queue is delayed, the first sampling task of the corresponding device 23M is distributed to at least one second device 220 based on the first task queue; the second device 220 that receives the first sampling task samples the corresponding device 23M based on the first sampling task, and feeds back the sampled first operating status data of the device 23M to the first device 210.

[0066] The aforementioned at least one second device 220 may be present in one or more configurations, and each second device 220 is configured with a corresponding second task queue. Based on this, when the aforementioned first device 210 sends a first sampling task, a second sampling task, or a third sampling task to the second device 220, it prioritizes sending the first sampling task, the second sampling task, or the third sampling task to the second device 220 with the fewest tasks in the second task queue. Furthermore, when the aforementioned second device 220 receives a first sampling task, a second sampling task, or a third sampling task, it adds the received first sampling task, second sampling task, or third sampling task to the second task queue and reorders them according to the priority level of each sampling task in the second task queue, wherein the priority of the second sampling task is higher than the priority level of the first sampling task and the priority of the third sampling task, respectively.

[0067] In some embodiments, see Figure 4 Before the aforementioned step S110, the method further includes: ① the first device 210 detects the device's initial connection to the system; ② when the detection device (231~23N) first connects to the system, it sends a third sampling task to a second device 220, the third sampling task including a third sampling signal with a third sampling duration and a third sampling precision, the third sampling duration being greater than the second sampling duration, and the third sampling precision being greater than the second sampling precision and greater than the first sampling precision; ③ the second device 220 samples the corresponding device 23M based on the third sampling task; ④ the second device 220 stores the operating status data of all operating conditions of the sampled corresponding device 23M in the database 240; ⑤ a notification of the completion of the third sampling task of the corresponding device 23M is sent to the first device 210, and the first device 210 generates the first sampling task, the second sampling task, and the target status data corresponding to different operating conditions in the database 240. It can be understood that in other embodiments, if the local storage space of the first device 210 is sufficient, the data in ④ and ⑥ can also be stored locally.

[0068] In some embodiments, step S110 may further include the following step: For each device 23M in each device (231-23N), the first device 210, based on the sampling timing t of the first task queue in the current sampling, n Acquire the first operating status data of the device in the system (23M); based on the sampling timing t of the current sampling... n Calculate the sampling timing t for the next sampling based on the current historical maintenance status information. n+1 The historical maintenance status information includes at least the historical usage time, expected remaining life, continuous trouble-free operation time, mean time between failures (MTBF), and post-maintenance usage time of the device 23M; and the calculated sampling timing t for the next sampling of the device 23M. n+1The corresponding first sampling task is added to the first task queue. Optionally, the expected remaining lifetime can be obtained by subtracting historical usage time from the theoretical lifetime, or it can be predicted based on the real-time operating status data of the device 23M using a preset lifetime prediction algorithm.

[0069] For example, the sampling timing time of the aforementioned next sampling Where a is a set value and a ≥ the minimum sampling frequency of the device, b is the historical usage time of device 23M, c is the expected remaining lifespan of device 23M, d is the continuous fault-free operation time of device 23M, e is the historical average fault-free operation time of device 23M, g is the post-maintenance usage time of device 23M, and w1, w2, and w3 represent weighting coefficients and w1 + w2 + w3 = 1. Where t0 is the moment when the first device receives notification that the third sampling task of device 23M is completed, or a moment thereafter. It is understood that the present invention is not limited to this example formula; for example, it can also be an equivalent transformation. In this invention, the sampling timing time t of the device... n Calculate the sampling timing t for the next sampling based on all or part of the information in the historical maintenance status information. n+1 All methods described herein are within the scope of protection of this invention.

[0070] The initial historical maintenance status information can be obtained through the following steps: When the first device 210 detects that the device 23M is connected to the system for the first time, it obtains the historical fault time, historical maintenance time, maintenance cycle, and historical usage time of the device 23M. It calculates the initial historical mean time between failures based on the historical fault time, calculates the initial continuous fault-free operation time based on the current time and historical fault time, calculates the initial expected remaining lifespan based on the theoretical lifespan and historical usage time, and calculates the initial post-maintenance usage time based on the historical maintenance time and maintenance cycle to obtain the initial historical maintenance status information.

[0071] Furthermore, the method also includes: periodically updating the historical usage time, expected remaining lifespan, continuous fault-free operation time, and post-maintenance usage time of each device (231-23N) to update the historical maintenance status information, wherein the timing of the periodic update of each device (231-23N) is determined according to the corresponding sampling timing time; and when any device 23M is determined to be faulty, the continuous fault-free operation time in the historical maintenance status information of device 23M is cleared, and the historical average fault-free operation time is recalculated using the current fault time to update the historical maintenance status information.

[0072] More specifically, the aforementioned "periodic updating of the historical usage time, expected remaining lifespan, continuous trouble-free operation time, and post-maintenance usage time of each device (231-23N) to update historical maintenance status information" can further include: during the sampling time t of the current sampling...n Or the sampling time t of the current sampling. n Sampling timing t for the next sampling n+1 The historical usage time, expected remaining life, continuous trouble-free operation time, and post-maintenance usage time of each device (231-23N) are updated periodically at any point in time to update the historical maintenance status information.

[0073] S120: The first device 210 determines whether the corresponding device 23M is running based on the first operating status data. If so, it sends the corresponding second sampling task to a second device 220. The second sampling task includes a second sampling signal with a second sampling duration and a second sampling precision. The second sampling duration is longer than the first sampling duration.

[0074] In some embodiments, in step S120, when the first device 210 determines that the corresponding device 23M is running, it immediately sends the corresponding second sampling task to a second device 220. If, for some reason, the corresponding second sampling task cannot be sent to a second device 220 immediately, the second sampling task triggered by the first sampling task of the current sampling can be sent in turn.

[0075] Considering that the same device may have multiple operating conditions, and that the operating status data under certain operating conditions may not be meaningful for detection, in some embodiments, step S120 may further include: the first device 210 determines whether the corresponding device 23M is running and whether it is in a predetermined operating condition based on the first operating status data; if the corresponding device 23M is running and in a predetermined operating condition, it sends a second sampling task matching the predetermined operating condition to a second device 220. This avoids the secondary sampling process of each device 23M under atypical operating conditions (i.e., subsequent steps S130 and S140), thereby reducing the system load.

[0076] In some embodiments, the second sampling precision is greater than the first sampling precision, so that the subsequent judgment result of the device 23M fault based on the second operating status data obtained by sampling with higher sampling precision is accurate and reliable.

[0077] However, in practical applications, the amount of sampled data is related to the sampling duration and sampling accuracy. Generally, the longer the sampling duration and the higher the sampling accuracy, the more burden the system may be placed if the second sampling accuracy is greater than the first sampling accuracy. Therefore, in some embodiments, the second sampling accuracy can be less than the first sampling accuracy, which can relatively reduce the amount of data in the longer secondary sampling and directly alleviate the system's storage burden. Furthermore, step S120 can further include "the first device 210 determines whether the corresponding device 23M is running and whether there is an abnormality based on the first operating status data. If the device 23M is running and there is an abnormality, a second sampling task is sent to a second device 220. More specifically, the first device 210 determines whether the corresponding device 23M is running based on the first operating status data. If so, it identifies the current operating condition and compares the first operating status data with the target status data that matches the current operating condition to determine whether the corresponding device 23M has an abnormality. If the device 23M is running and there is an abnormality, a second sampling task is sent to a second device 220." In this way, before the fault judgment in the subsequent step S140, a preliminary abnormality detection process can be performed based on the first operating status data of the device 23M obtained by sampling with higher sampling accuracy. Since the first operating status data is the sampling result obtained by sampling with higher sampling accuracy, it can compensate to some extent for the problem of low accuracy caused by the small amount of data.

[0078] S130: The second device 220, which receives the second sampling task, samples the corresponding device 23M based on the second sampling task, and feeds back the sampled second operating status data of the corresponding device 23M to the first device 210.

[0079] S140: The first device 210 determines whether the corresponding device 23M is faulty based on the second operating status data.

[0080] As previously stated, different operating conditions correspond to different first sampling tasks, second sampling tasks, and target state data. Thus, in some embodiments, step S140 can be further implemented as "the first device 210 compares the second operating state data with the target state data matching the corresponding second sampling task, and determines whether the corresponding device 23M is faulty based on the comparison result." Specifically, before step S140, the current operating condition of the corresponding device 23M can be identified in advance based on the first operating state data, and a second sampling task corresponding to the current operating condition can be sent to a second device 220.

[0081] It should be noted that the devices (231-23N) described in this embodiment can be any device with maintenance requirements. The predictive maintenance object of the predictive maintenance system in this embodiment can be a control system, production system, etc., composed of multiple devices (231-23N). Taking a production line system as an example, the devices (231-23N) include, but are not limited to, one or more of the following: execution devices, transmission devices, control devices, and drive devices. Furthermore, different device types can correspond to different types of operating status data, and the same device type can also correspond to the same or different types of operating status data, which can be selected and configured according to actual needs. Taking a motor as an example, the first and second operating status data can, for example, include one or more of torque, speed, output current, and output voltage, respectively.

[0082] Transfer to Figure 5 It illustrates another embodiment of a predictive maintenance system, which includes Figure 2 The system's first device 210 and second device 220 are shown. (Example) Figure 5 As shown, the first device 210 further includes a first interface unit 211, a first sampling unit 212 and a first processing unit 213, and the second device 220 further includes a second interface unit 221 and a second sampling unit 222.

[0083] The system comprises a first sampling unit 212 acquiring first operating status data of each device (231-23N) in the system at regular intervals via a first interface unit 211 based on a first task queue. A first processing unit 213 determines whether the corresponding device 23M is running based on the first operating status data. If the first processing unit 213 determines that the corresponding device 23M is running, the first sampling unit 212 sends a corresponding second sampling task to a second device 220 via the first interface unit 211 and receives second operating status data of the corresponding device 23M from the second device 220. The first processing unit 213 then determines whether the corresponding device 23M is faulty based on the second operating status data. The first task queue includes first sampling tasks with corresponding sampling timings for each device (231-23N). Each first sampling task includes a first sampling signal with a first sampling duration and a first sampling precision. Each second sampling task includes a second sampling signal with a second sampling duration and a second sampling precision, where the second sampling duration is longer than the first sampling duration.

[0084] The second sampling unit 222 samples the corresponding device 23M based on the received second sampling task via the second interface unit 221 to obtain the second operating status data of the corresponding device 23M, and feeds back the second operating status data to the first interface unit 211 via the second interface unit 221.

[0085] It should be noted that, for the sake of brevity, Figure 5The diagram only shows the connection relationships between the units in a first device 210 and the units in a second device 220. The invention is not limited to this; in other embodiments, the predictive maintenance system may further include a first device 210 and multiple second devices 220, and the connection relationships between each unit of each second device 220 and each unit of the first device 210 are the same. Figure 5 As shown.

[0086] It should be noted that, since the foregoing system embodiments are for implementing the foregoing predictive maintenance method embodiments, each unit in the system embodiments is designed to implement each step of the foregoing method. Therefore, the present invention is not limited to the foregoing system embodiments, and any system that can implement the foregoing predictive maintenance method should be included within the protection scope of the present invention. Furthermore, to save space, content identical to the foregoing method is omitted here, and all technical details in the foregoing method are applicable to the foregoing system embodiments.

[0087] It should be noted that in the claims and specification of this patent, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, 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 the element.

[0088] Obviously, those skilled in the art will understand that the modules or steps of the embodiments 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. Optionally, they can be implemented using computer-executable program code, thereby storing them 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 presented 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 embodiments of this specification are not limited to any particular combination of hardware and software.

[0089] Although the invention has been illustrated and described with reference to certain preferred embodiments thereof, those skilled in the art will understand that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.

[0090] The present invention has been shown and described in detail above with reference to the accompanying drawings and preferred embodiments. However, the present invention is not limited to these disclosed embodiments. Based on the above embodiments, those skilled in the art will know that more embodiments of the present invention can be obtained by combining the code review methods in the different embodiments. These embodiments are also within the protection scope of the present invention.

Claims

1. A predictive maintenance method, characterized in that, The method includes: The first device acquires the first operating status data of each device in the system at regular intervals based on the first task queue, wherein the first task queue includes the first sampling task corresponding to the sampling timing time of each device, and the first sampling task includes the first sampling signal having a first sampling duration and a first sampling precision. The first device determines whether the corresponding device is running based on the first operating status data. If so, it sends a corresponding second sampling task to a second device. The second sampling task includes a second sampling signal with a second sampling duration and a second sampling precision. The second sampling duration is greater than the first sampling duration. The second device that receives the second sampling task samples the corresponding device based on the second sampling task, and feeds back the sampled second operating status data of the corresponding device to the first device; The first device determines whether the corresponding device is faulty based on the second operating status data.

2. The predictive maintenance method according to claim 1, characterized in that, The first device periodically acquires the first operating status data of each device in the system based on the first task queue, further including: The first device sends a first sampling task corresponding to the first device to at least one second device based on the first task queue; The second device that receives the first sampling task samples the corresponding device based on the first sampling task and feeds back the first operating status data of the sampled device to the first device.

3. The predictive maintenance method according to claim 1, characterized in that, The first device periodically acquires the first operating status data of each device in the system based on the first task queue, further including: The first device directly collects the first operating status data of each device in the system at regular intervals based on the first task queue, and when it detects that any first sampling task in the first task queue is delayed, it distributes the first sampling task of the corresponding device to at least one second device based on the first task queue. The second device that receives the first sampling task samples the corresponding device based on the first sampling task and feeds back the first operating status data of the sampled device to the first device.

4. The predictive maintenance method according to claim 1, characterized in that, The second sampling precision is greater than the first sampling precision; The first device determines whether the corresponding device is running based on the first operating status data. If so, it sends a second sampling task to a second device. The method further includes: the first device determines whether the corresponding device is running and whether it is in a predetermined working condition based on the first operating status data. If the corresponding device is running and in a predetermined working condition, it sends a second sampling task matching the predetermined working condition to a second device. The first device's determination of whether the corresponding device is faulty based on the second operating status data further includes: the first device comparing the second operating status data with target status data that matches the predetermined operating condition, and determining whether the corresponding device is faulty based on the comparison result.

5. The predictive maintenance method according to claim 1, characterized in that, The second sampling precision is less than the first sampling precision; The first device determines whether the corresponding device is running based on the first operating status data. If so, it sends a second sampling task to a second device. The method further includes: the first device determines whether the corresponding device is running and whether there is an abnormality based on the first operating status data. If the corresponding device is running and there is an abnormality, it sends a second sampling task to a second device.

6. The predictive maintenance method according to claim 4 or 5, characterized in that, Before the first device periodically acquires the first operating status data of each device in the system based on the first task queue, the method further includes: The first device detects that the corresponding device is connected to the system for the first time, and sends a third sampling task to a second device. The third sampling task includes a third sampling signal with a third sampling duration and a third sampling precision. The third sampling duration is greater than the second sampling duration, and the third sampling precision is greater than the second sampling precision and greater than the first sampling precision. The second device samples the corresponding device based on the third sampling task and stores the operating status data of all the sampled operating conditions of the corresponding device in the database; The first device generates, in the database, the first sampling task, the second sampling task, and the target status data corresponding to different operating conditions based on the operating status data of all operating conditions of the corresponding device.

7. The predictive maintenance method according to claim 6, characterized in that, The second device has multiple devices, and each second device is configured with a corresponding second task queue; When the first device sends the first sampling task, the second sampling task, or the third sampling task to the second device, the method further includes: the first device preferentially sends the first sampling task, the second sampling task, or the third sampling task to the second device with the fewest tasks in the second task queue; When the second device receives the first sampling task, the second sampling task, or the third sampling task, the method further includes: the second device adding the received first sampling task, the second sampling task, or the third sampling task to a second task queue and reordering them according to the priority level of each sampling task in the second task queue, wherein the priority of the second sampling task is higher than the priority level of the first sampling task and the priority of the third sampling task.

8. The predictive maintenance method according to claim 1, characterized in that, The first device periodically acquires the first operating status data of each device in the system based on the first task queue, further including: For each of the aforementioned devices, the first device is based on the sampling timing t of the first task queue in the current sampling. n Acquire the first operating status data of the device in the system; Based on the sampling timing t of the current sampling n The sampling timing t for the next sampling is calculated based on the current historical maintenance status information. n+1 The historical maintenance status information includes at least the device’s historical usage time, expected remaining lifespan, continuous trouble-free operation time, mean time between failures (MTBF), and post-maintenance usage time. Based on the calculated sampling timing t of the device for the next sampling. n+1 Add the corresponding first sampling task to the first task queue.

9. The predictive maintenance method according to claim 8, characterized in that, The sampling timing time for the next sampling Where a is the set value and a≥ the minimum sampling frequency of the device, b is the historical usage time of the device, c is the expected remaining lifespan of the device, d is the continuous fault-free operation time of the device, e is the historical average fault-free operation time of the device, g is the post-maintenance usage time of the device, and w1, w2, w3 represent weighting coefficients and w1+w2+w3=1.

10. The predictive maintenance method according to claim 8, characterized in that, The method further includes: When the first device detects that the device is connected to the system for the first time, it obtains the device's historical fault time, historical maintenance time, maintenance cycle, and historical usage time to obtain initial historical maintenance status information. Among them, the initial historical mean time between failures is calculated based on the historical fault time, the initial continuous fault-free operation time is calculated based on the current time and the historical fault time, the initial expected remaining life is calculated based on the theoretical life and the historical usage time, and the initial post-maintenance usage time is calculated based on the historical maintenance time and the maintenance cycle.

11. The predictive maintenance method according to claim 10, characterized in that, The method further includes: The historical usage time, expected remaining lifespan, continuous trouble-free operation time, and post-maintenance usage time of each device are periodically updated to update the historical maintenance status information, wherein the timing of the periodic update for each device is determined based on the corresponding sampling timing time; and... When the device is determined to be faulty, the continuous fault-free operating time in the historical maintenance status information is cleared, and the historical average fault-free operating time is recalculated using the current fault time to update the historical maintenance status information.

12. A predictive maintenance system, characterized in that, It includes a first device (210) and at least one second device (220); wherein, The first device (210) includes: First interface unit (211); The first sampling unit (212) is used to periodically acquire the first operating status data of each device (231-23N) in the system based on the first task queue via the first interface unit (211). When a determination result indicating that the corresponding device is running is obtained, the unit sends the corresponding second sampling task to a second device (220) via the first interface unit (211) and acquires the second operating status data of the corresponding device from the second device (220) via the first interface unit (211). The first task queue includes the first sampling task with the corresponding sampling timing of each device (231-23N). The first sampling task includes a first sampling duration and a... The first sampling signal has a first sampling precision, and the second sampling task includes a second sampling signal having a second sampling duration and a second sampling precision, wherein the second sampling duration is greater than the first sampling duration; The first processing unit (213) is used to determine whether the corresponding device is running based on the first operating status data, and to determine whether the device is faulty based on the second operating status data of the corresponding device. The second device (220) includes: Second interface unit (221); The second sampling unit (222) is used to sample the corresponding device based on the received second sampling task via the second interface unit (221) to obtain the second operating status data of the corresponding device. The second operating status data is fed back to the first device (210) via the second interface unit (221).

Images