Fault prediction method and device, nonvolatile storage medium and electronic equipment

Abstract

The invention discloses a fault prediction method and device, a nonvolatile storage medium and electronic equipment. The method comprises the following steps: acquiring system data of a target optical transmission system; the fault type of a target device in the target optical transmission system is determined, fault related data of the target device is determined in system data, the fault type comprises a single-influence-factor fault and a multi-influence-factor fault, and the fault related data comprises data used for determining whether the target device breaks down or not; the single-influence-factor fault corresponds to one type of fault related data type, and the multi-influence-factor fault corresponds to multiple types of fault related data type; and determining a fault prediction model according to the fault type, and determining a fault prediction result of the target equipment by adopting the fault prediction model and the fault related data. The technical problem that the operation efficiency and stability of the optical transmission system are affected due to the fact that the fault of the optical transmission system cannot be accurately predicted in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the field of operation and maintenance, and specifically, to a fault prediction method, device, non-volatile storage medium and electronic equipment. Background Art

[0002] In the related art, it is impossible to accurately predict various types of faults in optical transmission systems in advance, and the faults can only be repaired after they occur, which affects the operating efficiency and stability of the optical transmission system.

[0003] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0004] The embodiments of the present application provide a fault prediction method, device, non-volatile storage medium and electronic device to at least solve the technical problem that the operating efficiency and stability of the optical transmission system are affected due to the inability to accurately predict the faults of the optical transmission system in the related art.

[0005] According to one aspect of an embodiment of the present application, a fault prediction method is provided, comprising: obtaining system data of a target optical transmission system, wherein the system data includes system operation data and hardware resource data; determining a fault type of a target device in the target optical transmission system, and determining fault-related data of the target device in the system data, wherein the fault type includes a single-influencing factor fault and a multi-influencing factor fault, and the fault-related data includes data used to determine whether the target device is faulty, a single-influencing factor fault corresponds to one type of fault-related data type, and a multi-influencing factor fault corresponds to multiple types of fault-related data types; determining a fault prediction model based on the fault type, and using the fault prediction model and the fault-related data to determine a fault prediction result for the target device.

[0006] Optionally, determining the fault prediction model based on the fault type includes: when the fault type is a single-influencing factor fault, determining the fault prediction model to be a linear regression model, and the fault prediction result includes the predicted value of the fault-related data; when the fault type is a multi-influencing factor fault, determining the fault prediction model to be a fault prediction model, and the fault prediction result includes whether the target device fails within a preset time period, wherein the fault prediction model includes multiple decision trees, and each decision tree corresponds to a type of fault-related data.

[0007] Optionally, using a fault prediction model and fault-related data to determine a fault prediction result for a target device includes: dividing the fault-related data into multiple sub-data sets according to data acquisition time information; determining the regression coefficient corresponding to each sub-data set in a linear regression model; and using a linear regression model to process the fault-related data based on the correspondence between each sub-data set and the regression coefficient to obtain a predicted value of the fault-related data.

[0008] Optionally, the method also includes: when the fault prediction model is a linear regression model, determining the predicted value of the fault-related data within a preset time period; obtaining the true value of the fault-related data based on the preset time period; determining the determination coefficient of the linear regression model based on the predicted value and the true value; when the determination coefficient is less than a preset determination coefficient threshold, adjusting the regression coefficient of the linear regression model.

[0009] Optionally, using a fault prediction model and fault-related data to determine a fault prediction result for a target device includes: in the case where the fault type is a multi-influencing factor fault, determining the influencing factors corresponding to each type of fault-related data, wherein the influencing factors are used to reflect the fault probability determined based on the fault-related data, and the greater the influencing factors, the greater the fault probability; determining the weight coefficients corresponding to each type of fault-related data based on the fault prediction model, wherein each decision tree in the fault prediction model corresponds to a weight coefficient; and determining whether a fault occurs in the target device within a preset time period based on the influencing factors and the weight coefficients.

[0010] Optionally, determining whether a target device fails within a preset time period based on the impact factor and the weight coefficient includes: determining the probability of each target device failing within the preset time period based on the impact factor and the weight coefficient; and determining the operation and maintenance priority of each target device based on the probability of failure.

[0011] Optionally, determining the fault-related data of the target device in the system data includes: determining the collection identification information of the system data, wherein the collection identification information includes data type information and data collection port information; determining the association relationship between the collection identification information and the identification information of the target device; and determining the fault-related data of the target device in the system data based on the association relationship.

[0012] Optionally, single-influencing factor failures include optical cable failures, optical line protection device sending end switch failures, optical line protection device receiving end switch failures, and connecting line failures between the optical protection device and the transmission device in the sending direction, wherein the optical cable failures include optical cable failures without optical line protection devices installed, and the corresponding fault-related data include the line attenuation of the target optical transmission system; the optical cable failures also include main route optical cable failures with optical line protection devices installed, and the corresponding fault-related data include the main route line loss of the optical line protection device; the optical cable failures also include backup route optical cable failures with optical line protection devices installed, and the corresponding fault-related data include the backup route line loss of the optical line protection device; the fault-related data corresponding to the sending end switch failure of the optical line protection device include the sending end switch insertion loss of the optical line protection device; the fault-related data corresponding to the receiving end switch failure of the optical line protection device include the receiving end switch insertion loss of the optical line protection device; and the fault-related data corresponding to the connecting line failure include the sending direction pigtail attenuation.

[0013] Optionally, the fault prediction method further includes: determining a device model of the target device when the fault type is a multi-influencing factor fault, wherein the multi-influencing factor fault includes an optical board fault; and determining a data type of the fault-related data based on the device model.

[0014] According to another aspect of an embodiment of the present application, a fault prediction device is also provided, including: a first processing module, used to obtain system data of a target optical transmission system, wherein the system data includes system operation data and hardware resource data; a second processing module, used to determine the fault type of a target device in the target optical transmission system, and determine fault-related data of the target device in the system data, wherein the fault type includes a single-influencing factor fault and a multiple-influencing factor fault, and the fault-related data includes data used to determine whether the target device is faulty, a single-influencing factor fault corresponds to one type of fault-related data type, and a multiple-influencing factor fault corresponds to multiple types of fault-related data types; a third processing module, used to determine a fault prediction model based on the fault type, and use the fault prediction model and the fault-related data to determine the fault prediction result of the target device.

[0015] According to another aspect of an embodiment of the present application, a non-volatile storage medium is provided, wherein a program is stored in the non-volatile storage medium, wherein when the program is executed, the device where the non-volatile storage medium is located is controlled to execute a fault prediction method.

[0016] According to another aspect of an embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein the processor is configured to run a program stored in the memory, wherein the fault prediction method is executed when the program is run.

[0017] According to another aspect of an embodiment of the present application, a computer program product is provided, including a computer program, which implements a fault prediction method when executed by a processor.

[0018] In an embodiment of the present application, system data of a target optical transmission system is obtained, wherein the system data includes system operation data and hardware resource data; a fault type of a target device in the target optical transmission system is determined, and fault-related data of the target device is determined in the system data, wherein the fault type includes a single-influencing factor fault and a multi-influencing factor fault, and the fault-related data includes data used to determine whether the target device is faulty, a single-influencing factor fault corresponds to one type of fault-related data type, and a multi-influencing factor fault corresponds to multiple types of fault-related data types; a fault prediction model is determined based on the fault type, and a fault prediction result of the target device is determined using the fault prediction model and the fault-related data. By classifying the fault type and predicting the fault using a fault prediction model corresponding to the fault type, the purpose of improving the accuracy of the fault prediction result is achieved, thereby achieving the technical effect of automatically predicting the possible faults of the target device in the optical transmission system with high accuracy, thereby solving the technical problem that the operating efficiency and stability of the optical transmission system are affected due to the inability to accurately predict the faults of the optical transmission system in the related art. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0020] Figure 1 This is a schematic diagram of the structure of a computer terminal (or mobile device) provided according to an embodiment of the present application;

[0021] Figure 2 This is a flow chart of a fault prediction method provided according to an embodiment of the present application;

[0022] Figure 3 This is a schematic diagram of an interface reference point of a wavelength division multiplexing optical transmission system provided in accordance with an embodiment of the present application;

[0023] Figure 4 This is a schematic diagram of an interface reference point of an optical transmission system equipped with an optical line protection device according to an embodiment of the present application;

[0024] Figure 5 This is a schematic diagram of optical power collection for an optical line protection device provided according to an embodiment of the present application;

[0025] Figure 6This is a schematic diagram of a fault prediction process for a single-influencing factor fault provided in an embodiment of the present application;

[0026] Figure 7 This is a schematic diagram of a fault prediction process for a fault with multiple influencing factors provided in an embodiment of the present application;

[0027] Figure 8 Schematic diagram of a random forest model with two decision trees provided according to an embodiment of the present application;

[0028] Figure 9 Schematic diagram of a random forest model with multiple decision trees provided according to an embodiment of the present application;

[0029] Figure 10 This is a flowchart of a fault prediction process provided according to an embodiment of the present application;

[0030] Figure 11 It is a structural diagram of a fault prediction device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0032] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0033] In order to better understand the embodiments of the present application, the technical terms involved in the embodiments of the present application are explained as follows:

[0034] Optical transmission system: A communications system that uses optical fiber as a transmission medium to transmit information in the form of light signals. This system leverages the high speed and high bandwidth of light to enable long-distance, high-capacity data transmission. Key components of an optical transmission system include a light source (usually a laser), optical modulators, optical fibers, optical receivers, and optical amplifiers.

[0035] Wavelength division multiplexing (WDM) increases the transmission capacity of optical fibers by simultaneously transmitting multiple optical signals of different wavelengths over the same optical fiber. Each wavelength can carry an independent data stream, significantly increasing network bandwidth without increasing the number of optical fibers.

[0036] Optical Line Protection (OLP) is a protection mechanism used in optical transmission networks to improve network reliability and stability. The OLP system provides line protection at the optical layer. When the primary line fails, it automatically switches to a backup line, ensuring service continuity and uninterrupted data transmission.

[0037] Fault prediction: A key technology in predictive maintenance, it leverages historical and real-time monitoring data to predict potential equipment or system failures. By identifying potential failures in advance, preventative measures can be taken to avoid unplanned downtime, reduce maintenance costs, and improve production efficiency and safety.

[0038] Machine learning: An artificial intelligence (AI) technique that enables computers to learn from data and improve their performance without being explicitly programmed. The core idea of machine learning is to build algorithms that can automatically "learn" patterns and regularities from data and then use these patterns to make predictions or decisions. Machine learning can be categorized as supervised learning, unsupervised learning, semi-supervised learning, and reinforcement learning.

[0039] Linear regression: A method for modeling the relationship between independent variables (features) and dependent variables using a linear function. Linear regression is a basic regression algorithm widely used in machine learning. Linear regression with multiple features is called multiple linear regression.

[0040] Linear Normalization: A commonly used data preprocessing method that aims to scale the data so that it falls into a small, specific range, such as [0, 1] or [-1, 1]. This method is very useful for many machine learning algorithms because it ensures that all features are on a similar scale, thus preventing some features from overly affecting model training due to their large numerical range.

[0041] Least Squares Method (LS): A mathematical optimization technique used to find the best function matching data. It estimates unknown data by minimizing the sum of squared errors. It is widely used in data processing fields such as parameter estimation, curve fitting, error estimation, uncertainty analysis, system identification, and forecasting.

[0042] Mean Squared Error (MSE): A commonly used loss function or error metric used in regression problems. It measures the average squared difference between the predicted and actual values.

[0043] The Coefficient of Determination (CDR), often expressed as (R^2) or (r^2), is a statistic used in regression analysis to measure the goodness of fit of a model. It represents the proportion of the variation in the response variable that is explained by the model to the total variation in the response variable.

[0044] Random Forest: An ensemble learning method primarily used for classification and regression tasks. It builds multiple decision trees and combines their results to improve prediction accuracy and stability. Key advantages of Random Forest include strong resistance to overfitting, excellent handling of high-dimensional data, ability to handle missing values and outliers, and ease of parallelization.

[0045] OLP network management: Real-time monitoring of OLP (optical line protection) equipment, collecting equipment alarms, configuration, and performance data.

[0046] Transmission equipment network management: Real-time monitoring of transmission equipment, collection of equipment alarms, configuration, and performance data.

[0047] Integrated transmission network management: realizes centralized management of transmission equipment and OLP equipment network elements, and completes alarm, performance, and configuration data collection.

[0048] Long-haul Resource Management System: This is a full-service long-haul network resource information management system, capable of managing resources such as optical cables, systems, equipment, and services along transmission trunk lines. It interfaces with the integrated transmission network management system, providing naming standards for linking data between the OLP network management system, transmission equipment network management system, and the integrated transmission network management system.

[0049] Possible causes of optical transmission system failures include optical cable failures, optical board failures, and connection issues. Optical cable failures refer to loose, broken, or damaged optical cables that may lead to signal transmission interruption. Depending on whether an OLP system is installed, optical cable failures can be categorized as either optical cable failures in optical transmission systems without OLP installed or failures in optical transmission systems with OLP installed. Optical board failures are key components in optical transmission equipment, responsible for sending and receiving optical signals. Failure of an optical board can prevent normal signal transmission. Connection issues include problems with connecting equipment such as ODFs, DDFs, and cables, which can hinder or interrupt signal transmission.

[0050] However, related technologies cannot accurately predict various types of faults in optical transmission systems in advance, and can only repair the faults after they occur, which affects the operating efficiency and stability of the optical transmission system.

[0051] In order to solve the above problems, relevant solutions are provided in the embodiments of the present application, which are described in detail below.

[0052] According to an embodiment of the present application, a method embodiment of a fault prediction method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0053] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 FIG1 shows a hardware structure block diagram of a computer terminal (or mobile device) for implementing a fault prediction method. Figure 1As shown, the computer terminal 10 (or mobile device 10) may include one or more (illustrated as 102a, 102b, ..., 102n) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0054] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuitry may be a single independent processing module, or may be incorporated in whole or in part into any of the other components of the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0055] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the fault prediction in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, to implement the above-mentioned fault prediction method. The memory 104 may include a high-speed random access memory, and may also include a 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 a memory remotely located relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0056] The transmission device 106 is configured to receive or transmit data via a network. A specific example of the aforementioned network may include a wireless network provided by the communications provider of the computer terminal 10. In one embodiment, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, the transmission device 106 may be a radio frequency (RF) module, which is configured to communicate with the Internet wirelessly.

[0057] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0058] In the above operating environment, the embodiment of the present application provides a fault prediction method, such as Figure 2 As shown, the method includes the following steps:

[0059] Step S202: acquiring system data of the target optical transmission system, wherein the system data includes system operation data and hardware resource data;

[0060] In some embodiments of the present application, Figure 3 As shown in Figure 1, in order to standardize the optical interface parameters in optical transmission systems, ITU-T G.692 and other related documents define all reference points of WDM optical transmission systems. Figure 3 As can be seen in the figure, Tx1, Tx2, and TxN are typically the terminal transmitters of the system. A wavelength division multiplexer (WDM) is used at the transmitting end to combine optical carriers of different wavelengths and transmit them through a single optical fiber for transmission. At the receiving end, a WDM (WDM) separates the optical carriers carrying different signals at different wavelengths. Figure 3 The specific meanings of the reference points corresponding to the various identifiers in are as follows:

[0061] S1…Sn: reference points of channels 1…n on the optical fiber at the transmitter optical output connector;

[0062] RM1…RMn: reference points of channels 1…n on the optical fiber at the optical input connector of the OM / OA;

[0063] MPI-S: Reference point on the optical fiber behind the optical output connector of the OM / OA;

[0064] S': reference point on the optical fiber behind the optical output connector of the line optical amplifier;

[0065] R': reference point on the optical fiber in front of the optical input connector of the line optical amplifier;

[0066] MPI-R: Reference point on the optical fiber in front of the optical input connector of OA / OD;

[0067] SD1…SDn: are the reference points at the optical output connector of OA / OD;

[0068] R1…Rn: Reference points at the receiver optical input connector.

[0069] In some embodiments of the present application, an optical transmission system having an OLP installed, such as Figure 4 As shown in the figure, the specific meanings of the reference points corresponding to each mark in the figure are as follows:

[0070] MPI-S: Reference point on the optical fiber behind the optical output connector of the OM / OA;

[0071] MPI-R: Reference point on the optical fiber in front of the optical input connector of OA / OD;

[0072] TX: Transmitter reference point where OLP equipment and WDM equipment are connected;

[0073] RX: The receiving end reference point where the OLP device and the WDM device are connected;

[0074] T1: Transmitter reference point where the OLP device and the primary routing optical cable are connected;

[0075] R1: The receiving end reference point where the OLP device is connected to the main routing optical cable;

[0076] T2: Transmitter reference point where the OLP equipment and the backup routing optical cable are connected;

[0077] R2: The receiving end reference point where the OLP equipment and the backup routing optical cable are connected.

[0078] In the technical solution provided in step S202, the step of determining the fault-related data of the target device in the system data includes: determining the collection identification information of the system data, wherein the collection identification information includes data type information and data collection port information; determining the association relationship between the collection identification information and the identification information of the target device; and determining the fault-related data of the target device in the system data based on the association relationship.

[0079] In some embodiments of the present application, the collected system operation data includes line-side optical power information of the optical disk, optical power data of the optical protection device, and other types of system operation data and resource data collected from the long-distance resource management system through the transmission integrated network management.

[0080] Optionally, when collecting line-side optical power information for optical amplifiers (OAs), the specific port information collected depends on the manufacturer and model of the transmission equipment. For example, for the OAU disk of an OSN9600 device from manufacturer A, the optical power of the OAU-1 (IN) and OAU-4 (OUT) ports must be collected. For the EONA disk of a ZXONE9700 device from manufacturer B, the optical power of the IN and OUT ports must be collected. For the PA disk of a FONST5000 device from manufacturer C, the optical power of the IN port must be collected, and for the OA disk of a FONST5000 device, the optical power of the OUT port must be collected.

[0081] As an optional implementation, Figure 5 As shown in the figure, when collecting optical power data for optical protection devices, a total of 12 optical power values must be collected for each relay segment. The optical power values collected for OLP ports are defined as follows: RX is the OLP receive optical power, R1 is the OLP primary route receive optical power, R2 is the OLP backup route receive optical power, TX is the OLP transmit optical power, T1 is the OLP primary route transmit optical power, and T2 is the OLP backup route transmit optical power. It should be noted that optical protection devices from different manufacturers and product models have different identifiers for the various power values.

[0082] In some embodiments of the present application, the integrated transmission network management system can collect resource data such as stations, computer rooms, racks, sub-racks, boards, optical cables, and optical cable segments from the long-distance resource management system. This information is stored in the relevant database tables of the long-distance resource management system and is associated with each other.

[0083] For the various types of data collected above, the port mapping table for the transmission network management's OTS (Optical Transit Segment) can be used to determine the corresponding relationships between various types of information, such as the equipment room, network element number, subframe, slot, disk name, and port. This mapping can then be further used to organically link alarm and performance data from the transmission equipment network management, optical protection network management, and long-distance resource management system, using the long-distance resource management system's optical cable segment and system segment as the standard. This allows fault prediction to determine the specific fault-related data associated with each device.

[0084] Step S204: determining a fault type of a target device in the target optical transmission system, and determining fault-related data of the target device in the system data, wherein the fault type includes a single-influencing factor fault and a multiple-influencing factor fault, and the fault-related data includes data used to determine whether the target device is faulty. A single-influencing factor fault corresponds to one type of fault-related data, and a multiple-influencing factor fault corresponds to multiple types of fault-related data.

[0085] The above-mentioned single-influencing factor failure refers to a failure that can be determined based on a type of fault-related data corresponding to the failure, and the multiple-influencing factor failure refers to a failure that can be determined based on multiple types of fault-related data corresponding to the failure.

[0086] Optionally, single-factor failures include optical cable failures, optical protection device failures, and connection problems. Optical protection device failures can be predicted using the OLP device's optical switch insertion loss in the transmit and receive directions to predict optical transmission system failures.

[0087] In the technical solution provided in step S204, single-factor failures include optical cable failures, optical line protection device transmitter switch failures, optical line protection device receiver switch failures, and connection line failures between the optical protection device and the transmission device in the transmission direction. Optical cable failures include optical cable failures without an optical line protection device installed, and the corresponding fault-related data includes the line loss of the target optical transmission system. Optical cable failures also include primary routing optical cable failures with an optical line protection device installed, and the corresponding fault-related data includes the primary routing line loss of the optical line protection device. Optical cable failures also include backup routing optical cable failures with an optical line protection device installed, and the corresponding fault-related data includes the backup routing line loss of the optical line protection device. Fault-related data corresponding to a transmitter switch failure of an optical line protection device includes the transmitter switch insertion loss of the optical line protection device. Fault-related data corresponding to a receiver switch failure of an optical line protection device includes the receiver switch insertion loss of the optical line protection device. Fault-related data corresponding to a connection line failure includes the transmission direction pigtail loss.

[0088] In some embodiments of the present application, for an optical cable fault in an optical transmission system without an OLP installed, the line loss of the OTS section of the optical transmission system can be used to predict the fault. The line loss is calculated as follows:

[0089] x=P MPI-S -P MPI-R

[0090] Among them, P MPI-S is the optical power at point MPI-S in the OTS segment, P MPI-R is the MPI-R optical power at the opposite end of the OTS segment, then P MPI-S and P MPI-R The difference is the line loss x of the OTS section.

[0091] If a fault occurs in the main routing optical cable of an optical transmission system with OLP installed, the main routing line loss of the OLP device can be used to predict the fault. The main routing line loss is calculated as follows:

[0092] x=P T1 -P R1

[0093] Let P T1 is the optical power at point T1 in the OLP relay section, P R1 is the optical power at the opposite end R1 of the OLP relay segment, then P T1 and P R1 The difference is the main routing line loss x of the OLP device.

[0094] If a backup route optical cable failure occurs in an optical transmission system with OLP installed, the backup route line loss of the OLP device can be used to predict the failure. The backup route line loss calculation formula is as follows:

[0095] x=P T2 -P R2

[0096] Let P T2 is the optical power at point T2 in the OLP relay section, P R2 is the optical power at point R2 on the opposite end of the OLP relay segment, then P T2 and P R2 The difference x is the backup routing line loss of the OLP equipment.

[0097] If an optical switch failure occurs at the transmitting end of an OLP device, the insertion loss of the optical switch at the transmitting end can be used to predict the failure. The insertion loss calculation formula for the optical switch at the transmitting end is as follows:

[0098] x=P Tx -P T1

[0099] Let P Tx is the optical power at the Tx point of the OLP device, P T1 For the optical power of point T1 on the same OLP device, P Tx and P T1 The difference x is used to predict the optical switch failure at the transmitting end of the OLP device.

[0100] If an optical switch failure occurs at the receiving end of an OLP device, the insertion loss of the optical switch at the receiving end can be used to predict the failure. The insertion loss calculation formula for the optical switch at the receiving end is as follows:

[0101] x=P R1 -P Rx

[0102] Let P R1 is the optical power of OLP device R1, P Rx For the same OLP device Rx point optical power, P R1 and P Rx The difference x is used to predict the optical switch failure at the receiving end of the OLP device.

[0103] For an optical transmission system with OLP installed, the optical transmission system fault caused by the connection line between the optical protection device and the transmission equipment in the sending direction can be predicted by the sending direction fiber pigtail loss. The calculation formula for the sending direction fiber pigtail loss is as follows:

[0104] x=P MPI-S -P Tx

[0105] Let P MPI-S is the optical power of the WDM device MPI-S, P Tx is the optical power at the Tx point of the connected OLP device, then P MPI-S and P Tx The difference x is used to predict the optical transmission system fault caused by the connecting line between the optical protection device and the transmission device in the sending direction.

[0106] For an optical transmission system with OLP installed, the optical transmission system fault caused by the connection line between the optical protection device and the transmission equipment in the receiving direction is predicted by the receiving direction pigtail loss. The calculation formula for the receiving direction pigtail loss is as follows:

[0107] x=P Rx -P MPI-R

[0108] Let P Rx is the optical power at the Rx point of the OLP device, P Tx is the optical power of the connected WDM device MPI-R, then P MPI-S and P Tx The difference x is used to predict the optical transmission system fault caused by the connecting line between the optical protection device and the transmission device in the sending direction.

[0109] As an optional implementation, the fault prediction method further includes: if the fault type is a multi-factor fault, determining the device model of the target device, where the multi-factor fault includes an optical board fault; and determining the data type of the fault-related data based on the device model. Optical board faults can be predicted using the optical board's performance parameters. Taking the OAU optical board of an OptiX OSN 9600 device from manufacturer A as an example, the performance parameters include board gain, board temperature, laser temperature, laser operating current, laser backlight current, and laser cooling current. It should be noted that the types of fault-related data required for determining optical board faults vary for optical boards of different manufacturers and device models. For example, the fault-related data required for the OA optical board of a FONST 5000 device from manufacturer B include board gain, board temperature, laser operating current, laser bias current, and laser cooling current. The fault-related data required for the EOA optical board of a ZXONE 9700E device from manufacturer C include board gain, laser bias current, and detection point temperature.

[0110] Step S206 , determining a fault prediction model according to the fault type, and determining a fault prediction result of the target device using the fault prediction model and the fault-related data.

[0111] In the technical solution provided in step S206, the step of determining a fault prediction model based on the fault type includes: when the fault type is a single-influencing factor fault, determining the fault prediction model to be a linear regression model, and the fault prediction result includes the predicted value of the fault-related data; when the fault type is a multi-influencing factor fault, determining the fault prediction model to be a fault prediction model, and the fault prediction result includes whether the target device fails within a preset time period, wherein the fault prediction model includes multiple decision trees, and each decision tree corresponds to a type of fault-related data.

[0112] As an optional implementation, the step of using a fault prediction model and fault-related data to determine the fault prediction result of the target device includes: dividing the fault-related data into multiple sub-data sets according to data acquisition time information; determining the regression coefficient corresponding to each sub-data set in the linear regression model; and using the linear regression model to process the fault-related data based on the corresponding relationship between each sub-data set and the regression coefficient to obtain the predicted value of the fault-related data.

[0113] In some embodiments of the present application, when the fault prediction model is a linear regression model, the predicted value of the fault-related data within a preset time period is determined; the true value of the fault-related data is obtained based on the preset time period; the determination coefficient of the linear regression model is determined based on the predicted value and the true value; when the determination coefficient is less than a preset determination coefficient threshold, the regression coefficient of the linear regression model is adjusted.

[0114] In some embodiments of the present application, for a single-factor failure, a linear regression model is used to predict the process as follows: Figure 6 As shown. Figure 6 The prediction process is based on the prediction of optical cable faults in optical transmission systems without OLP installed, and includes the following steps:

[0115] Step S602: data collection.

[0116] First, historical data traffic can be collected from the transmission professional network management and optical protection network management. Taking the optical cable fault prediction of the optical transmission system without OLP installed as an example, it is necessary to collect the optical power P of the MPI-S point in the OTS section. MPI-S , OTS segment peer end MPI-R optical power P MPI-R , then P MPI-S and P MPI-R The difference is the line loss x of the OTS section, which is calculated as follows:

[0117] x=P MPI-S -P MPI-R

[0118] The system can automatically calculate the line loss x of the OTS section as input data for optical network fault prediction based on linear regression.

[0119] The data granularity can be determined based on actual needs and the network management data collection cycle, such as collecting data on a 15-minute, hourly, or daily basis, and recorded as x1, x2, ..., xn, respectively. n is a positive integer greater than 1 and can be adjusted as needed. A larger n indicates a more accurate prediction.

[0120] Step S604: data preprocessing.

[0121] In some embodiments of the present application, linear normalization (also known as minimum-maximum normalization or Min-Max normalization) technology can be used to clean the collected data. The formula of the normalization technology is:

[0122]

[0123] Among them, x is the original OTS line loss data, x min is the minimum line loss of the OTS segment in the data set, x max is the maximum line loss of the OTS segment in the data set, x norm It is the normalized OTS line loss data.

[0124] In addition, as an optional implementation, the NumPy library in Python can also be used to calculate the normalization of the data. The calculation method is:

[0125] normalized_data=(data-data_min) / (data_max-data_min)

[0126] Where (data) is the original data, (data_min) and (data_max) are the minimum and maximum values of the OTS line loss data, respectively.

[0127] Step S606: Establish a linear regression model.

[0128] Based on the selected independent variables and dependent variables, a linear regression equation is established. The linear regression equation is expressed as follows:

[0129] y=f(x)=w1x1+w2x2+...+w n x n +b

[0130] Among them, x1, x2, ..., x n As n features (independent variables), the line loss of the OTS segments from the first 1, 2 to the first n time periods after normalization; the coefficients w1, w2, ..., w n is the weight of the feature (regression coefficient); b is the bias; y is the output variable y.

[0131] Step S608: model training.

[0132] Use historical data to train the linear regression model and estimate the regression coefficients (w1, w2, ..., w n ).

[0133] In some embodiments of the present application, the regression coefficient can be solved by the Ordinary Least Squares (OLS) method to minimize the sum of squared errors between the predicted value and the actual value. The least squares method estimates the regression coefficient by minimizing the sum of squared errors between the predicted value and the actual value, and uses matrix operations and derivation techniques to find the optimal regression coefficient. The NumPy library in Python is used to solve the regression coefficient of the multivariate linear regression. The specific calculation process includes the following steps:

[0134] Prepare data: including the feature matrix X (including the intercept term) and the target variable y.

[0135] Compute the transpose X_transpose of X.

[0136] Computes the product of the transpose of X and X_transpose_dot_X.

[0137] Calculates the inverse matrix X_transpose_dot_X_inv of X_transpose_dot_X.

[0138] Computes the product of the transpose of X and y, X_transpose_dot_y.

[0139] Multiply X_transpose_dot_y by the inverse matrix to get the regression coefficient beta.

[0140] Step S610: model evaluation.

[0141] As an optional implementation, the mean squared error (MSE) can be used to evaluate the performance of the model. MSE is the average of the squares of the differences between the predicted and actual values. It measures the average error in the model's predictions. The smaller the MSE, the better the model's performance. The specific calculation formula is as follows:

[0142]

[0143] Among them, y i is the actual value, is the model's predicted value, and n is the number of samples. The smaller the MSE, the better the model's performance. The mean squared error can be calculated using the NumPy library in Python. The calculation steps include:

[0144] import numpy as np

[0145] errors = y_pred - y_true (calculate the error for each point (predicted value - actual value))

[0146] mse = np.mean(np.square(errors)) (calculate mean square error)

[0147] Step S612, prediction.

[0148] Once the model is trained and evaluated, it can be used to predict the line loss of the future OTS segment. Simply substitute the past independent variable values (x1, x2, ..., xn) into the trained linear regression equation (y = f(x) = w1x1 + w2x2 + ... + wnxn + b) to obtain the predicted line loss value of the OTS segment.

[0149] Step S614: Check whether the prediction result is accurate.

[0150] After the predicted time is reached, the predicted line loss value of the OTS section will be Compared with the line loss value (y) of the actual OTS segment on that day, the determination coefficient (R 2 ) method to test the accuracy of linear regression prediction results. The calculation method is:

[0151] Calculate the Total Sum of Squares (TSS)

[0152] TSS is the sum of the squares of the differences between the observations and their mean, reflecting the total variation in the dependent variable.

[0153]

[0154] Among them, y i is the i-th observation, is the mean of all observations, and n is the number of observations.

[0155] Calculate the Residual Sum of Squares (RSS)

[0156] RSS is the sum of the squares of the differences between the observed values and the values predicted by the regression model, reflecting the changes that the model cannot explain.

[0157]

[0158] Among them, y i is the i-th observation, is the predicted value of the regression model for the ith observation, and n is the number of observations.

[0159] Calculate the coefficient of determination (R 2 )

[0160] R 2 is the proportion of the variation explained by the model to the total variation.

[0161]

[0162] If the model fits the data perfectly (i.e., all predicted values are equal to the observed values), then the RSS will be 0, and therefore R 2 will be 1. If the model does not explain any variation (i.e., the predicted value is just the mean of the dependent variable), then RSS will be equal to TSS, so R 2 will be 0.

[0163] Usually, R 2 Values of 0.7 to 0.8 and above are considered to be relatively good model fit. 2 The value 0.8 is used as the judgment threshold, that is, R 2 If it is greater than 0.8, the model is considered to be well-fitted and the prediction results are relatively accurate, and the process continues to step S616. Otherwise, the model is considered to be poorly-fitted and the prediction results are inaccurate, and the process proceeds to step S608. This can be represented by the following code:

[0164] Input RSS (total sum of squares), TSS (residual sum of squares)

[0165] R 2 =1-RSS / TSS

[0166] If R 2 ≧0.8

[0167] Then, if the model fit is good and the prediction result is accurate, the process goes to step S616;

[0168] Else, if it is determined that the model fitting is not good and the prediction result is inaccurate, the process goes to step S608.

[0169] End (algorithm ends)

[0170] As an optional implementation, using a fault prediction model and fault-related data to determine a fault prediction result for a target device includes: in the case where the fault type is a multi-influencing factor fault, determining the influencing factors corresponding to each type of fault-related data, wherein the influencing factors are used to reflect the fault probability determined based on the fault-related data, and the greater the influencing factors, the greater the fault probability; determining the weight coefficients corresponding to each type of fault-related data based on the fault prediction model, wherein each decision tree in the fault prediction model corresponds to a weight coefficient; and determining whether a fault occurs in the target device within a preset time period based on the influencing factors and the weight coefficients.

[0171] Optionally, a confidence level for the target device's operational quality within a preset time period can be determined based on the influencing factors and weight coefficients. The confidence level is equal to the weighted sum of the influencing factors determined based on the weight coefficients. A higher confidence level indicates a lower probability of failure, and vice versa. Furthermore, a preset confidence threshold can be set. When the confidence level falls below the threshold, it is determined that the target device will fail within a preset time period. The preset time period is a future time period.

[0172] In some embodiments of the present application, the step of determining whether a target device fails within a preset time period based on an impact factor and a weight coefficient includes: determining the probability of each target device failing within a preset time period based on the impact factor and the weight coefficient; and determining the operation and maintenance priority of each target device based on the probability of failure.

[0173] In some embodiments of the present application, taking the OAU optical board of manufacturer A as an example, the fault-related data types corresponding to the optical board include single board gain, single board temperature, laser temperature, laser operating current, laser backlight current and laser cooling current. In actual prediction, these data will be quantified into influencing factors, and the fault prediction model will be used to determine the weight coefficients of these influencing factors, and then predict whether a fault will occur based on the influencing factors and the corresponding weight coefficients. The specific process is as follows: Figure 7 As shown:

[0174] Step S702: Collect data that affects the operating quality of the OAU optical board

[0175] Data that affects the operation quality of OAU optical boards can be collected from the transmission network management. These data include six types: board gain, board temperature, laser temperature, laser operating current, laser backlight current, and laser cooling current. The network operation quality of each OAU optical board is quantified using the network operation quality confidence δ. The value of this quality confidence δ can be used to determine the probability of optical board failure. The larger the value, the lower the failure probability.

[0176] Confidence level δ of network operation quality:

[0177]

[0178] A i Respectively represent the impact factors corresponding to the 6 types of fault-related data; Bi represents the weights of the 6 impact factors. i With B i The values are between 0 and 1.

[0179] In step S704, the six factors are quantitatively evaluated, and the evaluation results are set as A1, A2, A3, A4, A5, and A6, respectively, with values ranging from 0 to 1. The specific evaluation method is as follows:

[0180] The influence factor of single board gain on the operation quality of OAU optical board is set to A1:

[0181] First, the OAU board gain is calculated based on the total input optical power and total output optical power of the OAU board. Then, based on the OAU board gain at all data collection points, the average gain and standard deviation of the OAU board gain are calculated. Finally, the influence factor of the board gain on the operation quality of the OAU board is obtained. The specific calculation steps are as follows:

[0182] P=P OUT -P IN

[0183] Among them, P OUT is the total output optical power of the OAU optical board, P IN is the total optical power input to the OAU optical board, then P OUT and P IN The difference is OAU light board P GAIN is the OAU optical board gain P.

[0184]

[0185] Where n is the number of data collection points, P i is the single-board gain of the OAU optical board at the i-th data collection point, then the average gain of the OAU optical board is

[0186]

[0187] Among them, P i is the OAU optical board gain at the i-th data collection point, is the average gain of the OAU optical panel; σ k The standard deviation of the OAU optical board gain at the kth data collection point is used to measure whether the OAU optical board gain is stable.

[0188]

[0189] Among them, A1 is the influence factor of single board gain on the operation quality of OAU optical board, σ k is the standard deviation of the OAU optical panel gain at the kth data collection point. k The smaller it is, the more stable the gain of the OAU optical board is, which means the probability of OAU optical board failure determined by the board gain is smaller. On the contrary, if σ k If it gets bigger, it means that the gain of the OAU optical board is unstable, and the probability of the OAU optical board failure determined by the board gain becomes larger. The value range of A1 is between 0 and 1, so if σ k >1, then A1 is 0.

[0190] The impact factor of board temperature on the operation quality of the OAU optical board is A2:

[0191]

[0192] Where n is the number of data collection points, T i is the temperature of the OAU board at the i-th data collection point, then the average temperature of the OAU board is

[0193]

[0194] Among them, T i is the temperature of the OAU optical board at the i-th data collection point, is the average temperature of the OAU light panel; σ k is the standard deviation of the OAU board temperature at the kth data collection point, which is used to measure whether the OAU board temperature is stable.

[0195]

[0196] Among them, A2 is the influence factor of single board temperature on the operation quality of OAU optical board, σ k is the standard deviation of the temperature of the OAU optical board at the kth data collection point. k The smaller it is, the more stable the temperature of the OAU board is, and the smaller the probability of OAU board failure determined by the board temperature is. k If it gets bigger, it means that the temperature of the OAU board is unstable, and the probability of OAU board failure determined by the board temperature increases. The value range of A2 is between 0 and 1, so if σ k >1, then A2 is 0.

[0197] The influence factor of laser temperature on the operation quality of the OAU optical board is set to A3:

[0198]

[0199] Where n is the number of data collection points, T i is the OAU light board laser temperature at the i-th data collection point, then the average laser temperature of the OAU light board is

[0200]

[0201] Among them, T i is the OAU light board laser temperature at the i-th data collection point, is the average laser temperature of the OAU light board; σ k is the standard deviation of the OAU light board laser temperature at the kth data collection point, which is used to measure whether the OAU light board laser temperature is stable.

[0202]

[0203] Among them, A3 is the influence factor of the single-board laser temperature on the operation quality of the OAU optical board, σ k is the standard deviation of the OAU light board laser temperature at the kth data collection point. k The smaller it is, the more stable the OAU optical board laser temperature is, and the smaller the probability of OAU optical board failure determined by the laser temperature is. k If it gets bigger, it means that the temperature of the OAU optical board laser is unstable, and the probability of OAU optical board failure determined by the laser temperature becomes larger. The value range of A2 is between 0 and 1, so if σ k >1, then A3 is 0.

[0204] The influence factor of the laser operating current on the operation quality of the OAU optical board is set to A4:

[0205]

[0206] Where n is the number of data collection points, C i is the OAU optical panel laser operating current at the i-th data collection point, then the average laser operating current of the OAU optical panel is

[0207]

[0208] Among them, C i is the OAU optical panel laser operating current at the i-th data collection point, is the average laser operating current of the OAU optical panel; σ k is the standard deviation of the OAU optical panel laser operating current at the kth data collection point, which is used to measure whether the OAU optical panel laser operating current is stable.

[0209]

[0210] Among them, A4 is the influence factor of single board temperature on the operation quality of OAU optical board, σ k is the standard deviation of the OAU optical panel laser operating current at the kth data collection point. k The smaller it is, the more stable the OAU optical board laser operating current is, and the smaller the probability of OAU optical board failure determined by the laser operating current is. k If it increases, it means that the working current of the OAU optical board laser is unstable, and the probability of OAU optical board failure determined by the laser working current increases. The value range of A2 is between 0 and 1, so if σ k >1, then A4 is 0.

[0211] The influence factor of laser backlight current on the operation quality of OAU light board is set to A5:

[0212]

[0213] Where n is the number of data collection points, C i is the OAU light panel laser backlight current at the i-th data collection point, then the average laser backlight current of the OAU light panel is

[0214]

[0215] Among them, C i is the OAU light panel laser backlight current at the i-th data collection point, is the average laser backlight current of the OAU light board; σ k is the standard deviation of the OAU light panel laser backlight current at the kth data collection point, which is used to measure whether the OAU light panel laser backlight current is stable.

[0216]

[0217] Among them, A5 is the influence factor of laser backlight current on the operation quality of OAU light board, σ k is the standard deviation of the OAU light panel laser backlight current at the kth data collection point. k The smaller it is, the more stable the OAU optical board laser backlight current is, and the smaller the probability of OAU optical board failure determined by the laser backlight current is. k If it gets bigger, it means that the backlight current of the OAU optical board laser is unstable, and the probability of OAU optical board failure determined by the laser backlight current becomes larger. The value range of A2 is between 0 and 1, so if σ k>1, then A5 is 0.

[0218] The influence factor of the laser cooling current on the operation quality of the OAU optical board is set to A6:

[0219]

[0220] Where n is the number of data collection points, C i is the OAU optical board laser cooling current at the i-th data collection point, then the average laser cooling current of the OAU optical board is

[0221]

[0222] Among them, C i is the cooling current of the OAU optical panel laser at the i-th data collection point, is the average laser cooling current of the OAU optical panel; σ k is the standard deviation of the OAU optical panel laser cooling current at the kth data collection point, which is used to measure whether the OAU optical panel laser cooling current is stable.

[0223]

[0224] Among them, A6 is the influence factor of laser cooling current on the operation quality of OAU optical board, σ k is the standard deviation of the OAU optical panel laser cooling current at the kth data collection point. k The smaller it is, the more stable the OAU optical board laser cooling current is, and the smaller the probability of OAU optical board failure determined by the laser cooling current is. k If it increases, it means that the cooling current of the OAU optical board laser is unstable, and the probability of OAU optical board failure determined by the cooling current of the laser increases. The value range of A6 is between 0 and 1, so if σ k >1, then A6 is 0.

[0225] Step S706, obtain the initial characteristic values of the factors affecting the operation quality of the OAU optical board, B1, B2, B3, B4, B5, B6, all between 0-1, and B1+B2+B3+B4+B5+B6=1, that is, the sum of B1, B2, B3, B4, B5, B6 is always equal to 1.

[0226] Step S708: Use the random forest classifier to adjust the values of B1 to B6.

[0227] Specifically, random forest is an algorithm that integrates multiple trees using the concept of ensemble learning. Its basic unit is the decision tree, and its essence belongs to the ensemble learning method, a major branch of machine learning. The concept of ensemble learning aims to address the inherent shortcomings of a single model or a model with a specific set of parameters, thereby combining more models to complement each other and avoid limitations. Random forest integrates many decision trees into a forest and uses them together to predict the final result.

[0228] There are two keywords in the name of random forest, one is "random" and the other is "forest". From an intuitive point of view, each decision tree is a classifier (assuming that the problem is classification). Then for an input sample, N trees will have N classification results, such as Figure 8 The random forest of two trees shown will have two classification results, Figure 9 Each tree corresponds to a classifier, resulting in N classification results. The random forest model votes on all classification results and assigns the category with the most votes as the final output. Random forest is a flexible and easy-to-use machine learning algorithm that can produce good results in most situations, even without hyperparameter tuning.

[0229] In some embodiments of the present application, the evaluation function of the random forest classifier can use a decision tree BCE function, with inputs A1 to A6 and y, where y is the feedback value of whether the model's evaluation result is reasonable, and outputs B1 to B6. The formula for using the BCE function is as follows:

[0230] BCELoss=(X i ,y i )=-w i [y i logA i +(1-y i )log(1-A i )]

[0231] Supervised learning also requires manual labeling of training sample data. Parameters A1-A6 of the optical transmission system evaluation factors serve as model inputs, and the corresponding labels indicate whether the evaluation results are reasonable (yes: 1, no: 0). The sample data is divided into a training set:test set ratio of 8:2, and a random forest classifier is trained on the training samples. During training, the classifier continuously adjusts the mapping between the learning input parameters A1-A6 and the labels. After training, each classifier model outputs the importance weighting factors (B1-B6) for each input feature value.

[0232] As an optional implementation, based on the random forest model, this application allows each decision tree (classifier model) to process a type of fault-related data, and by training the random forest model, determines the importance weight factor corresponding to each classifier model. The importance weight factor output by the classifier model is the importance weight factor of the fault-related data corresponding to the classifier model.

[0233] Step S710: Obtain the confidence level of the OAU optical board operation quality assessment value.

[0234] Based on the evaluation results of the six factors (A1 to A6) and the eigenvalue weight coefficients (B1 to B6) within the random forest classifier, the confidence δ of the OAU light panel operation quality evaluation value calculated by the random forest classifier can be obtained:

[0235]

[0236] Step S712: Obtain the OAU optical panel operation quality evaluation value recommended by the algorithm.

[0237] The system collects network management alarms, system performance, and the number of network elements (NEs) in the optical transmission system from the transmission network management system to determine the current values of the six factors that affect the operational quality of the OAU optical board. The system then automatically calculates and ranks the operational quality confidence levels of each target device in the optical transmission system based on the aforementioned formula. Lower confidence levels indicate a higher fault severity and therefore a higher maintenance priority.

[0238] Step S714 : Check whether the OAU optical board operation quality evaluation value recommended by the algorithm is consistent with the actual operation quality of the OAU optical board.

[0239] If the OAU board recommended by the algorithm fails, the algorithm's prediction of an OAU board failure is accurate. If the OAU board recommended by the algorithm does not fail, the algorithm's prediction of an OAU board failure is inaccurate. As an optional implementation, if the confidence calculation result is less than a preset threshold, the algorithm's prediction of an OAU board failure can be considered to be successful; otherwise, the algorithm's prediction of an OAU board failure is considered to be successful.

[0240] If the two results are inconsistent, jump to step S708 and use the classifier in the random forest model again to adjust the values of B1 to B6.

[0241] In some embodiments of the present application, there is also provided a Figure 10 The fault prediction process shown includes the following steps:

[0242] Step S1002, using the transmission integrated network management to collect performance data from the transmission equipment network management;

[0243] Optionally, the integrated transmission network management system collects data on equipment performance from the network management systems of various transmission devices, including but not limited to line-side optical power information of optical disks.

[0244] Step S1004, using the transmission integrated network management to collect performance data from the optical protection network management;

[0245] Optionally, the transmission integrated network management system also collects performance data from the network management system of the optical protection (OLP) equipment, including key parameters such as optical power.

[0246] Step S1006, using the integrated transmission network management to collect resource data from the long-distance resource management system;

[0247] Optionally, this step involves obtaining various resource data from the long-distance resource management system, such as information on stations, computer rooms, racks, subframes, boards, optical cables, and optical cable segments. This information can be used for subsequent data association and analysis.

[0248] Step S1008, realizing data association among the transmission equipment network management, optical protection network management, and long-distance resource management system;

[0249] Step S1010, optical transmission system fault classification;

[0250] Optical transmission system failures can be divided into single-factor failures and multiple-factor failures.

[0251] Step S1012, optical network fault prediction based on linear regression;

[0252] A linear regression model can be used to predict single-factor failures.

[0253] Step S1014, optical network fault prediction based on random forest;

[0254] Random forest models can be used to predict multi-factor failures.

[0255] Step S1016, outputting the optical network fault prediction result;

[0256] Optionally, the prediction results of optical network fault prediction based on linear regression and the prediction results of optical network fault prediction based on random forest can be summarized and sent to relevant maintenance personnel for processing in the form of fault prediction reports, maintenance work orders, etc., so that maintenance personnel can quickly locate and handle potential faults, thereby reducing maintenance costs and improving user experience.

[0257] Step S1018, end the prediction process.

[0258] By acquiring system data of a target optical transmission system, wherein the system data includes system operation data and hardware resource data; determining the fault type of a target device in the target optical transmission system, and determining fault-related data of the target device in the system data, wherein the fault type includes a single-influencing factor fault and a multi-influencing factor fault, and the fault-related data includes data used to determine whether the target device has a fault, a single-influencing factor fault corresponds to one type of fault-related data type, and a multi-influencing factor fault corresponds to multiple types of fault-related data types; determining a fault prediction model according to the fault type, and using the fault prediction model and the fault-related data to determine the fault prediction result of the target device, the purpose of improving the accuracy of the fault prediction result is achieved by classifying the fault type and using the fault prediction model corresponding to the fault type to predict the fault, thereby achieving the technical effect of automatically predicting the possible faults of the target device in the optical transmission system with high accuracy, thereby solving the technical problem that the operating efficiency and stability of the optical transmission system are affected due to the inability to accurately predict the faults of the optical transmission system in the related art.

[0259] In addition, the maintenance methods of optical transmission systems in related technologies often rely on manual inspections and regular testing, which is not only time-consuming and labor-intensive, but also difficult to detect potential faults in a timely manner. The method provided in the embodiments of the present application uses intelligent means to achieve comprehensive system monitoring and fault warning, allowing maintenance personnel to quickly locate and address potential faults, thereby reducing maintenance costs.

[0260] The fault prediction method provided by the embodiments of this application can accurately predict the future operating status of the system, thereby helping operators to rationally plan and optimize network resources. For example, when a failure of optical transmission equipment is predicted to occur, resource scheduling and backup can be performed in advance to ensure the normal operation of the communication network. This helps to avoid problems such as resource waste and communication interruptions, and improves the utilization efficiency of network resources.

[0261] Furthermore, the fault prediction method provided by the embodiments of this application can promptly detect and address potential faults, thereby avoiding problems such as communication interruptions and service quality degradation, which is of great significance for improving user experience. The fault prediction method provided by the embodiments of this application can promptly detect network risks before a fault occurs, achieving "zero" user awareness of network faults, thereby improving user experience.

[0262] In summary, the fault prediction method provided in the embodiments of the present application plays an important role in improving system stability and reliability, reducing maintenance costs, optimizing network resource utilization, enhancing user experience, and promoting technological innovation and development.

[0263] The present application provides a fault prediction device. Figure 11It is a structural diagram of the device. Figure 11 As can be seen from the figure, the device includes: a first processing module 110, used to obtain system data of the target optical transmission system, wherein the system data includes system operation data and hardware resource data; a second processing module 112, used to determine the fault type of the target device in the target optical transmission system, and determine the fault-related data of the target device in the system data, wherein the fault type includes a single-influencing factor fault and a multi-influencing factor fault, and the fault-related data includes data used to determine whether the target device is faulty, a single-influencing factor fault corresponds to one type of fault-related data type, and a multi-influencing factor fault corresponds to multiple types of fault-related data types; a third processing module 114, used to determine a fault prediction model according to the fault type, and use the fault prediction model and the fault-related data to determine the fault prediction result of the target device.

[0264] In some embodiments of the present application, the step of the first processing module 110 determining the fault-related data of the target device in the system data includes: determining the acquisition identification information of the system data, wherein the acquisition identification information includes data type information and data acquisition port information; determining the association relationship between the acquisition identification information and the identification information of the target device; and determining the fault-related data of the target device in the system data based on the association relationship.

[0265] In some embodiments of the present application, single-influencing factor failures include optical cable failures, optical line protection device sending end switch failures, optical line protection device receiving end switch failures, and connecting line failures between the optical protection device and the transmission device in the sending direction, wherein: optical cable failures include optical cable failures without optical line protection devices installed, and the corresponding fault-related data include the line attenuation of the target optical transmission system. Optical cable failures also include main route optical cable failures with optical line protection devices installed, and the corresponding fault-related data include the main route line loss of the optical line protection device. Optical cable failures also include backup route optical cable failures with optical line protection devices installed, and the corresponding fault-related data include the backup route line loss of the optical line protection device. The fault-related data corresponding to the sending end switch failure of the optical line protection device include the sending end switch insertion loss of the optical line protection device. The fault-related data corresponding to the receiving end switch failure of the optical line protection device include the receiving end switch insertion loss of the optical line protection device. The fault-related data corresponding to the connecting line failure include the sending direction pigtail loss.

[0266] In some embodiments of the present application, the second processing module 112 is also used to: determine the device model of the target device when the fault type is a multi-factor fault, wherein the multi-factor fault includes a light board fault; and determine the data type of the fault-related data based on the device model.

[0267] In some embodiments of the present application, the step of the third processing module 114 determining a fault prediction model based on the fault type includes: when the fault type is a single-influencing factor fault, determining the fault prediction model to be a linear regression model, and the fault prediction result includes the predicted value of the fault-related data; when the fault type is a multi-influencing factor fault, determining the fault prediction model to be a fault prediction model, and the fault prediction result includes whether the target device fails within a preset time period, wherein the fault prediction model includes multiple decision trees, and each decision tree corresponds to a type of fault-related data.

[0268] In some embodiments of the present application, the steps of the third processing module 114 using the fault prediction model and fault-related data to determine the fault prediction result of the target device include: dividing the fault-related data into multiple sub-data sets according to data acquisition time information; determining the regression coefficient corresponding to each sub-data set in the linear regression model; and using the linear regression model to process the fault-related data based on the correspondence between each sub-data set and the regression coefficient to obtain the predicted value of the fault-related data.

[0269] In some embodiments of the present application, the third processing module 114 is also used to: determine the predicted value of the fault-related data within a preset time period when the fault prediction model is a linear regression model; obtain the true value of the fault-related data based on the preset time period; determine the determination coefficient of the linear regression model based on the predicted value and the true value; and adjust the regression coefficient of the linear regression model when the determination coefficient is less than a preset determination coefficient threshold.

[0270] In some embodiments of the present application, the step in which the third processing module 114 uses the fault prediction model and fault-related data to determine the fault prediction result of the target device includes: in the case where the fault type is a multi-influencing factor fault, determining the influencing factors corresponding to each type of fault-related data, wherein the influencing factors are used to reflect the fault probability determined based on the fault-related data, and the larger the influencing factors, the greater the fault probability; determining the weight coefficients corresponding to each type of fault-related data based on the fault prediction model, wherein each decision tree in the fault prediction model corresponds to a weight coefficient; and determining whether the target device fails within a preset time period based on the influencing factors and the weight coefficients.

[0271] In some embodiments of the present application, the step in which the third processing module 114 determines whether a target device fails within a preset time period based on the influencing factor and the weight coefficient includes: determining the probability of each target device failing within a preset time period based on the influencing factor and the weight coefficient; and determining the operation and maintenance priority of each target device based on the probability of failure.

[0272] It should be noted that the various modules in the above-mentioned fault prediction device can be program modules (for example, a set of program instructions that implement a certain specific function) or hardware modules. For the latter, it can be expressed in the following forms, but is not limited to this: the expression form of each of the above-mentioned modules is a processor, or the functions of each of the above-mentioned modules are implemented by a processor.

[0273] According to an embodiment of the present application, a non-volatile storage medium is further provided, wherein a program is stored in the non-volatile storage medium. When the program is executed, the device containing the non-volatile storage medium is controlled to execute the following fault prediction method: obtaining system data of a target optical transmission system, wherein the system data includes system operation data and hardware resource data; determining a fault type of a target device in the target optical transmission system, and determining fault-related data of the target device in the system data, wherein the fault type includes a single-influencing factor fault and a multi-influencing factor fault, and the fault-related data includes data used to determine whether the target device has a fault, wherein a single-influencing factor fault corresponds to one type of fault-related data type, and a multi-influencing factor fault corresponds to multiple types of fault-related data types; determining a fault prediction model based on the fault type, and using the fault prediction model and the fault-related data to determine a fault prediction result for the target device.

[0274] According to an embodiment of the present application, an electronic device is also provided, including a memory and a processor, the processor being used to run a program stored in the memory, wherein the following fault prediction method is executed when the program is running: obtaining system data of a target optical transmission system, wherein the system data includes system operation data and hardware resource data; determining a fault type of a target device in the target optical transmission system, and determining fault-related data of the target device in the system data, wherein the fault type includes a single-influencing factor fault and a multiple-influencing factor fault, and the fault-related data includes data used to determine whether the target device is faulty, a single-influencing factor fault corresponds to one type of fault-related data type, and a multiple-influencing factor fault corresponds to multiple types of fault-related data types; determining a fault prediction model based on the fault type, and using the fault prediction model and the fault-related data to determine a fault prediction result of the target device.

[0275] According to an embodiment of the present application, a computer program product is also provided, including a computer program, which implements the following fault prediction method when executed by a processor: obtaining system data of a target optical transmission system, wherein the system data includes system operation data and hardware resource data; determining a fault type of a target device in the target optical transmission system, and determining fault-related data of the target device in the system data, wherein the fault type includes a single-influencing factor fault and a multiple-influencing factor fault, and the fault-related data includes data used to determine whether the target device is faulty, a single-influencing factor fault corresponds to one type of fault-related data type, and a multiple-influencing factor fault corresponds to multiple types of fault-related data types; determining a fault prediction model based on the fault type, and using the fault prediction model and the fault-related data to determine a fault prediction result for the target device.

[0276] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0277] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0278] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0279] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0280] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the relevant technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0281] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A fault prediction method, characterized in that: include: Acquiring system data of a target optical transmission system, wherein the system data includes system operation data and hardware resource data; Determining a fault type of a target device in the target optical transmission system, and determining fault-related data of the target device in the system data, wherein the fault type includes a single-influencing factor fault and a multiple-influencing factor fault, and the fault-related data includes data used to determine whether the target device is faulty, the single-influencing factor fault corresponds to one type of fault-related data, and the multiple-influencing factor fault corresponds to multiple types of fault-related data; A fault prediction model is determined according to the fault type, and a fault prediction result of the target device is determined using the fault prediction model and the fault-related data.

2. The fault prediction method according to claim 1, characterized in that: Determining a fault prediction model according to the fault type includes: In a case where the fault type is the single-influencing factor fault, determining that the fault prediction model is a linear regression model, and the fault prediction result includes a predicted value of the fault-related data; In the case where the fault type is the multi-factor fault, the fault prediction model is determined to be a fault prediction model, and the fault prediction result includes whether the target device fails within a preset time period, wherein the fault prediction model includes multiple decision trees, and each decision tree corresponds to one type of fault-related data.

3. The fault prediction method according to claim 2, characterized in that: Determining a fault prediction result of the target device using the fault prediction model and the fault-related data includes: Dividing the fault-related data into multiple sub-data sets according to data collection time information; Determine the regression coefficient corresponding to each of the sub-data sets in the linear regression model; According to the corresponding relationship between each of the sub-data sets and the regression coefficient, the linear regression model is used to process the fault-related data to obtain a predicted value of the fault-related data.

4. The fault prediction method according to claim 2, characterized in that: The method further comprises: In a case where the fault prediction model is the linear regression model, determining a predicted value of the fault-related data within a preset time period; Obtaining a true value of the fault-related data according to the preset time period; Determining the coefficient of determination of the linear regression model based on the predicted value and the true value; When the determination coefficient is less than a preset determination coefficient threshold, the regression coefficient of the linear regression model is adjusted.

5. The fault prediction method according to claim 2, characterized in that: Determining a fault prediction result of the target device using the fault prediction model and the fault-related data includes: In the case where the fault type is the multi-influencing factor fault, determining the influencing factors corresponding to each type of the fault-related data, wherein the influencing factors are used to reflect the fault probability determined based on the fault-related data, and the greater the influencing factors, the greater the fault probability; Determining weight coefficients corresponding to various types of fault-related data based on the fault prediction model, wherein each decision tree in the fault prediction model corresponds to one weight coefficient; Determine whether the target device fails within the preset time period based on the impact factor and the weight coefficient.

6. The fault prediction method according to claim 5, characterized in that: Determining whether a fault occurs in the target device within the preset time period according to the influencing factor and the weight coefficient includes: Determining the probability of each target device failing within the preset time period based on the influencing factor and the weight coefficient; The operation and maintenance priority of each of the target devices is determined according to the occurrence probability.

7. The fault prediction method according to claim 1, characterized in that: Determining the fault-related data of the target device in the system data includes: Determining collection identification information of the system data, wherein the collection identification information includes data type information and data collection port information; Determining an association relationship between the acquisition identification information and the identification information of the target device; The fault-related data of the target device is determined in the system data according to the association relationship.

8. The fault prediction method according to claim 1, characterized in that: The single-factor failure includes an optical cable failure, a switch failure at the transmitting end of an optical line protection device, a switch failure at the receiving end of an optical line protection device, and a connection line failure between the optical protection device and the transmission device in the transmitting direction. The optical cable fault includes an optical cable fault in which the optical line protection device is not installed, and the corresponding fault-related data includes the line attenuation of the target optical transmission system; The optical cable fault also includes a fault of a main routing optical cable in which the optical line protection device is installed, and the corresponding fault-related data includes a loss of the main routing line of the optical line protection device; The optical cable fault also includes a backup route optical cable fault where the optical line protection device is installed, and the corresponding fault-related data includes a backup route line loss of the optical line protection device; The fault-related data corresponding to the optical line protection device transmitting end switch fault includes the transmitting end switch insertion loss of the optical line protection device; The fault-related data corresponding to the optical line protection device receiving end switch fault includes the receiving end switch insertion loss of the optical line protection device; The fault-related data corresponding to the connection line fault includes the pigtail loss in the sending direction.

9. The fault prediction method according to claim 1, characterized in that: The fault prediction method further includes: In a case where the fault type is the multi-influencing factor fault, determining the device model of the target device, wherein the multi-influencing factor fault includes an optical board fault; The data type of the fault-related data is determined according to the device model.

10. A fault prediction device, characterized in that: include: A first processing module is configured to obtain system data of a target optical transmission system, wherein the system data includes system operation data and hardware resource data; a second processing module, configured to determine a fault type of a target device in the target optical transmission system, and to determine fault-related data of the target device in the system data, wherein the fault type includes a single-influencing factor fault and a multiple-influencing factor fault, and the fault-related data includes data used to determine whether the target device is faulty, the single-influencing factor fault corresponds to one type of fault-related data, and the multiple-influencing factor fault corresponds to multiple types of fault-related data; The third processing module is configured to determine a fault prediction model according to the fault type, and determine a fault prediction result of the target device using the fault prediction model and the fault-related data.

11. A non-volatile storage medium, characterized in that: The non-volatile storage medium stores a program, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute the fault prediction method according to any one of claims 1 to 9.

12. An electronic device, characterized in that: include: A memory and a processor, wherein the processor is configured to run a program stored in the memory, wherein the fault prediction method according to any one of claims 1 to 9 is executed when the program is run.

13. A computer program product, characterized in that The method comprises a computer program, which implements the fault prediction method according to any one of claims 1 to 9 when executed by a processor.

Images