Fault location method and system for power optical transmission system based on decision tree

Through the fault location method based on the decision tree, the fault segments are quickly located using the characteristic attribute data of the OTN device, which solves the problem of fault location difficulties in power OTN system and improves the troubleshooting efficiency and system stability.

CN115833935BActive Publication Date: 2024-11-08INFORMATION COMM COMPANY STATE GRID SHANDONG ELECTRIC POWER +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202211344898.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-10-31
Publication Date
2024-11-08
Estimated Expiration
2042-10-31

AI Technical Summary

Technical Problem

The fault location of the power OTN system is difficult and the technical requirements of the operation and maintenance personnel are high, resulting in low troubleshooting efficiency.

Method used

The fault location method based on the decision tree is adopted, and the feature vector is constructed by obtaining the characteristic attribute data of the OTN device and inputting it into the trained OTN fault location decision tree, and the fault segment is quickly located using the decision tree's judgment logic.

Benefits of technology

It realizes the rapid and accurate positioning of the OTN system fault location, reduces the technical requirements of operation and maintenance personnel, and enhances the business performance stability of the power OTN system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115833935B_ABST
    Figure CN115833935B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and system for fault location of an electric power optical transmission system based on a decision tree, comprising: obtaining characteristic attribute data of an electric power OTN device and constructing a characteristic vector; inputting the constructed characteristic vector into a trained OTN fault location decision tree and outputting a fault location section; wherein the OTN fault location decision tree comprises a root node, a plurality of internal branch nodes and a plurality of leaf nodes; each leaf node corresponds to a located fault section, and each other internal branch node corresponds to a characteristic state of the OTN device, the root node contains a full set of samples, and the path from the root node to each leaf node corresponds to a fault location judgment logic. The present invention can quickly locate the fault location of the OTN optical transmission system, and solves the problem that the OTN system is difficult to analyze and determine the fault location, and has high technical requirements for operation and maintenance personnel.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of optical transmission system fault location, and in particular to a method and system for locating faults in an electric power optical transmission system based on a decision tree. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] In the process of building smart grids, the use of advanced technologies can improve the operating efficiency of power systems and achieve innovative development of power grids. As a representative technology, OTN optical transmission plays a positive role in the working performance of power communication networks and effectively promotes the stable and good operation of power systems.

[0004] The transmission capacity of the power optical transmission system (hereinafter referred to as OTN) is greatly improved compared with the traditional SDH transmission system, but the resulting large number of pigtail connections and complex signal flows have greatly increased the difficulty of daily operation and maintenance. When a power OTN fails, it is difficult to quickly locate and troubleshoot the fault, which requires a high level of technical skills from the operation and maintenance personnel. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a decision tree-based power optical transmission system fault location method and system, which can quickly locate the fault location of the power OTN, solving the problem that it is difficult to analyze and determine the fault location of the power OTN and has high technical requirements for operation and maintenance personnel.

[0006] In some embodiments, the following technical solutions are adopted:

[0007] A method for locating faults in a power optical transmission system based on a decision tree, comprising:

[0008] Obtain the input optical power, output optical power, MUT_LOS, R_LOS, OSC_LOS, HARD_BARD, IN_PWR_LOW, NE_NOT_LOGIN, NE_COMMU_BREAK alarm, DOWN and UP status characteristic attribute data of the power OTN equipment, and construct a feature vector;

[0009] Input the constructed feature vector into the trained OTN fault location decision tree and output the fault location section;

[0010] The OTN fault location decision tree includes a root node, several internal branch nodes and several leaf nodes; each leaf node corresponds to a located fault section, and each other internal branch node corresponds to a characteristic state of the OTN device. The root node contains the full set of samples, and the path from the root node to each leaf node corresponds to a fault location decision logic.

[0011] As a further solution, the training process of the OTN fault location decision tree includes:

[0012] (1) Based on the fault elimination records during the daily operation and maintenance of the power OTN equipment, a training sample set D1 is constructed; the training sample set includes the feature vectors of the power OTN equipment and the corresponding fault sections;

[0013] (2) Select relevant features when power OTN equipment fails and construct attribute set A;

[0014] (3) Generate the root node T. If all instances in the training sample set D1 belong to the same class, then T is a single-node tree, and the instance class is used as the class label of node T. Otherwise, calculate the information gain of each feature in the attribute set A, and select the feature attribute a with the largest information gain. * ;

[0015] (4) If a * If a is not greater than the set threshold, then T is a single-node tree, and the category C with the largest number of instances in the training sample set D1 is recorded as the class label of the node; if a * is greater than the set threshold, for a * Every value of Let D v Indicates that the training sample set D1 is in a * The upper value is A sample subset; each D v Generate a branch node of node T;

[0016] (5) If Mark the branch node as a leaf node, and use the category C with the largest number of instances in the training sample set D1 as the class label of the node; otherwise, use D v As the starting node, calculate the attribute set A except the feature attribute a * The feature attribute with the largest information gain other than the one with the largest information gain is selected, and the process of (4) to (5) is repeated until all the feature attributes in the attribute set A are judged. The trained OTN fault location decision tree is obtained.

[0017] As a further solution, the feature vector of the power OTN device includes input optical power, output optical power, MUT_LOS, R_LOS, OSC_LOS, HARD_BARD, IN_PWR_LOW, NE_NOT_LOGIN, NE_COMMU_BREAK alarm, DOWN and UP state feature attribute data of the power OTN device.

[0018] As a further solution, it also includes: constructing a verification set D2 based on the fault elimination records during the daily operation and maintenance of some power OTN equipment; after obtaining the trained OTN fault location decision tree, using the verification set D2 to verify the decision accuracy H1 of each leaf node on the verification set; then verifying the decision accuracy H2 of the upper-level node of each leaf node on the verification set, if the decision accuracy H2 ≥ H1, then pruning the leaf node to obtain the pruned OTN fault location decision tree as the final decision tree for fault location.

[0019] As a further solution, the information gain is calculated as follows:

[0020] The information entropy of the training sample set D1 is defined as is the proportion of samples of the kth category in the training sample set D1;

[0021] Assume that the feature attribute a has V possible values ​​{a 1 ,a 2 ,...,a V}, if a is used to divide the training sample set D1, V branch nodes will be generated, where the vth branch node contains all the training sample set D1 with a value of a on attribute a. v The sample is counted as D v ;

[0022] The information gain obtained by dividing the training sample set D1 with attribute a is calculated as:

[0023]

[0024] As a further solution, if all instances in the training sample set D1 have no features Or if the samples in the training sample set D1 have the same value on the attribute set A, then T is a single-node tree, and the category C with the largest number of instances in the training sample set D1 is used as the class label of the node.

[0025] As a further solution, if The branch node is marked as a leaf node, and the category C with the largest number of instances in the training sample set D1 is used as the class label of the node.

[0026] In other embodiments, the following technical solutions are adopted:

[0027] A fault location system for an electric power optical transmission system based on a decision tree, comprising:

[0028] The data acquisition module is used to obtain the input optical power, output optical power, MUT_LOS, R_LOS, OSC_LOS, HARD_BARD, IN_PWR_LOW, NE_NOT_LOGIN, NE_COMMU_BREAK alarm, DOWN and UP status characteristic attribute data of the power OTN equipment and construct a characteristic vector;

[0029] A fault section location module is used to input the constructed feature vector into the trained OTN fault location decision tree and output the fault location section;

[0030] The OTN fault location decision tree includes a root node, several internal branch nodes and several leaf nodes; each leaf node corresponds to a located fault section, and each other internal branch node corresponds to a characteristic state of the OTN device. The root node contains the full set of samples, and the path from the root node to each leaf node corresponds to a fault location decision logic.

[0031] In other embodiments, the following technical solutions are adopted:

[0032] A terminal device comprises a processor and a memory, wherein the processor is used to implement various instructions; and the memory is used to store a plurality of instructions, wherein the instructions are suitable for being loaded by the processor and executing the above-mentioned decision tree-based power optical transmission system fault location method.

[0033] In other embodiments, the following technical solutions are adopted:

[0034] A computer-readable storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded by a processor of a terminal device and executing the above-mentioned decision tree-based power optical transmission system fault location method.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] (1) The present invention introduces a decision tree machine learning algorithm, which can quickly locate the fault location of the OTN optical transmission system through continuous training and learning, solving the problem that it is difficult to analyze and determine the fault location of the OTN system and has high technical requirements for operation and maintenance personnel. It further enhances the robustness of the service performance of the power OTN optical transmission system and ensures the normal and stable operation of the power grid service.

[0037] Other features and advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is the structural diagram of the two communication sites A and B in the power optical transmission system. DETAILED DESCRIPTION

[0039] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0040] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0041] Embodiment 1

[0042] In one or more embodiments, a method for locating a fault in a power optical transmission system based on a decision tree is disclosed, which specifically includes the following process:

[0043] (1) Obtain the input optical power, output optical power, MUT_LOS, R_LOS, OSC_LOS, HARD_BARD, IN_PWR_LOW, NE_NOT_LOGIN, NE_COMMU_BREAK alarm, DOWN and UP status feature attribute data of the power OTN equipment and construct a feature vector;

[0044] (2) Inputting the constructed feature vector into the trained OTN fault location decision tree and outputting the fault location segment; wherein the OTN fault location decision tree includes a root node, a plurality of internal branch nodes and a plurality of leaf nodes; each leaf node corresponds to a located fault segment, and each of the other internal branch nodes corresponds to a characteristic state of the OTN device. The root node contains the full set of samples, and the path from the root node to each leaf node corresponds to a fault location decision logic.

[0045] Specifically, the structures of the two communication sites A and B in the power OTN system are abstracted as Figure 1 As shown, the entire power OTN system can be divided into five parts:

[0046] The five possible fault sections are the service equipment of station A - the OTU board of OTN equipment of station A, the OTU board of OTN equipment of station A - the FIU board of OTN equipment of station A, the FIU board of OTN equipment of station A - the FIU board of OTN equipment of station B, the OTU board of OTN equipment of station B - the FIU board of OTN equipment of station B, and the service equipment of station B - the OTU board of OTN equipment of station B.

[0047] Each fault segment is recorded as vector Y, and the i-th fault record is y i express.

[0048] Then select the service equipment, OTU (optical conversion unit) board, OMU (optical multiplexing unit) board, ODU (optical splitter unit) board, SC2 (optical monitoring channel unit) board, BA (preamplifier) ​​board, PA (postamplifier) ​​board, FIU (fiber line interface unit) board, etc. Figure 1 The input optical power, output optical power, MUT_LOS (loss of combined signal), R_LOS (loss of signal on the receiving line side), OSC_LOS (loss of optical monitoring signal), HARD_BARD (hardware damage), IN_PWR_LOW (input optical power is too low), NE_NOT_LOGIN (network element not logged in), NE_COMMU_BREAK (network element communication interruption) alarm, DOWN (shutdown), UP (in operation) status, etc. of the listed devices when a fault occurs are used as feature attribute data and recorded as vector X. Each fault record is denoted by x. i It means that through the fault elimination records in the daily operation and maintenance process, the training test sample set D = {(x1, y1), (x2, y2), ..., (x m ,y m )}.

[0049] Randomly select 70% of the samples in the sample set D as the training set: D1 = {(x1, y1), (x2, y2), ..., (x n ,y n )}.

[0050] The remaining 30% of the samples are used to obtain the validation set: D2 = {(x1, y1), (x2, y2), ..., (x s ,y s )}.

[0051] In this embodiment, a decision tree includes a root node, several internal branch nodes and several leaf nodes. The leaf node corresponds to the decision result, that is, the fault section of the located power optical transmission system, and each other node corresponds to an attribute test, that is, the characteristic state of the OTN equipment board, such as the input optical power of the SC2 board. The sample set contained in each node is divided into sub-nodes according to the result of the attribute test, and the root node contains the full set of samples. The path from the root node to each leaf node corresponds to a fault location judgment logic, which determines the fault location step by step through different characteristics of the input optical power of the SC2 board, the OSC_LOS of the SC2 board, the MUT_LOS of the BA board, the MUT_LOS of the PA board, the DOWN and UP states of the service equipment, etc.

[0052] The purpose of decision tree learning is to generate a decision tree that has a strong ability to handle unseen fault cases. In order to avoid overfitting, after using the training set D1 to generate the decision tree, the OTN fault location decision tree is pruned using the validation set D2 to enhance the generalization ability of the decision tree.

[0053] In this embodiment, the training process of the OTN fault location decision tree is as follows:

[0054] (1) Based on the fault elimination records during the daily operation and maintenance of the power OTN equipment, a training sample set D1 is constructed; the training sample set includes the feature vectors of the power OTN equipment and the corresponding fault sections;

[0055] (2) Select the feature vector when the power OTN equipment fails and construct the attribute set A;

[0056] In this embodiment, attribute set A represents a set of all characteristic attributes. When training the decision tree, all attributes in attribute set A are used for classification. The feature vector of the power OTN equipment includes input optical power, output optical power, MUT_LOS, R_LOS, OSC_LOS, HARD_BARD, IN_PWR_LOW, NE_NOT_LOGIN, NE_COMMU_BREAK alarm, DOWN and UP state features of the power OTN equipment.

[0057] (3) Generate the root node T. If all instances in the training sample set D1 belong to the same class, then T is a single-node tree, and the instance class is used as the class label of node T.

[0058] If all instances in the training sample set D1 have no features Or if the samples in D have the same value on A, then T is a single-node tree, which records the category C with the largest number of instances in the training sample set D1 as the class label of the node and returns T; this process is mainly to prevent errors in the program code during training.

[0059] Otherwise, calculate the information gain of each feature in attribute set A and select the feature attribute a with the largest information gain * = arg max a∈A Gain(D,a); In this embodiment, the information gain Gain(D,a) is calculated as follows:

[0060] The proportion of samples of the kth category in the sample set D2 is p k (k=1,2,...,n), then the information entropy of D2 is defined as Assume that attribute a has V possible values ​​{a 1 ,a 2 ,...,a V}, if a is used to divide the sample set D2, V branch nodes will be generated, where the vth branch node contains all the nodes in D2 whose value is a on attribute a v The sample is counted as D v The information gain obtained by dividing the sample set D2 by attribute a is calculated as:

[0061]

[0062] (4) If a * If it is not greater than the set threshold δ, then T is a single-node tree, and the category C with the largest number of instances in the training sample set D1 is recorded as the class label of the node;

[0063] If a * is greater than the set threshold δ, for a * Every value of (such as the input optical power of each OTN device), let D v Indicates that the training sample set D1 is in a * The upper value is A sample subset; each D v Generate a branch node of node T;

[0064] Among them, by setting the relevant threshold δ, the depth of the OTN fault location decision tree can be controlled.

[0065] (5) If (i.e., the sample subset is an empty set), mark the branch node as a leaf node, and use the category C with the largest number of instances in the training sample set D1 as the class label of the node; otherwise, use D v As the starting node, calculate the attribute set A except the feature attribute a * The feature attribute with the largest information gain other than the one with the largest information gain is selected, and the process of (4) to (5) is repeated until all the feature attributes in the attribute set A are judged. The trained OTN fault location decision tree is obtained.

[0066] Post-pruning process: For the trained OTN fault location decision tree, a validation set D2 = {(x1, y1), (x2, y2), ..., (x s ,y s )}, test the decision accuracy H1 of the fault location decision tree T. Then, shrink upward from the leaf node on the tree, test the decision accuracy H2 of the previous level node on the validation set, and if the decision accuracy H2 ≥ H1, prune the node to obtain the pruned OTN fault location decision tree T2.

[0067] In the next step, the OTN fault location decision tree is pruned recursively to obtain the OTN fault location decision tree T with the highest accuracy and the strongest generalization ability. i , using the fault location decision tree T i Locate the fault section of the optical transmission system.

[0068] Embodiment 2

[0069] In one or more embodiments, a decision tree-based power optical transmission system fault location system is disclosed, comprising:

[0070] The data acquisition module is used to obtain the input optical power, output optical power, MUT_LOS, R_LOS, OSC_LOS, HARD_BARD, IN_PWR_LOW, NE_NOT_LOGIN, NE_COMMU_BREAK alarm, DOWN and UP status characteristic attribute data of the power OTN equipment and construct a characteristic vector;

[0071] A fault section location module is used to input the constructed feature vector into the trained OTN fault location decision tree and output the fault location section;

[0072] The OTN fault location decision tree includes a root node, several internal branch nodes and several leaf nodes; each leaf node corresponds to a located fault section, and each other internal branch node corresponds to a characteristic state of the OTN device. The root node contains the full set of samples, and the path from the root node to each leaf node corresponds to a fault location decision logic.

[0073] It should be noted that the specific implementation methods of the above modules have been described in Example 1 and will not be described in detail here.

[0074] Embodiment 3

[0075] In one or more embodiments, a terminal device is disclosed, including a server, the server including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the decision tree-based power optical transmission system fault location method in Embodiment 1 when executing the program. For the sake of brevity, it will not be described here.

[0076] It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general-purpose processors, digital signal processors DSP, application-specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0077] The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0078] In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in a processor or an instruction in the form of software.

[0079] Embodiment 4

[0080] In one or more embodiments, a computer-readable storage medium is disclosed, in which a plurality of instructions are stored, wherein the instructions are suitable for being loaded by a processor of a terminal device and executing the decision tree-based power optical transmission system fault location method described in Example 1.

[0081] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A method for locating faults in a power optical transmission system based on a decision tree, characterized in that: include: Obtain the input optical power, output optical power, combined signal loss MUT_LOS, receiving line side signal loss R_LOS, optical monitoring signal loss OSC_LOS, hardware damage HARD_BARD, input optical power too low IN_PWR_LOW, network element not logged in NE_NOT_LOGIN, network element communication interruption NE_COMMU_BREAK alarm, shutdown DOWN and running UP state feature attribute data of the power OTN equipment, and construct a feature vector; Input the constructed feature vector into the trained OTN fault location decision tree and output the fault location section; The OTN fault location decision tree includes a root node, several internal branch nodes and several leaf nodes; each leaf node corresponds to a located fault section, and each other internal branch node corresponds to a characteristic state of the OTN device. The root node contains a full set of samples, and the path from the root node to each leaf node corresponds to a fault location decision logic; The fault location section includes: station A service equipment-station A OTN equipment OTU single board, station A OTN equipment OTU single board-station A OTN equipment FIU single board, station A OTN equipment FIU single board-station B OTN equipment FIU single board, station B OTN equipment OTU single board-station B OTN equipment FIU single board, and station B service equipment-station B OTN equipment OTU single board; The training process of the OTN fault location decision tree includes: (1) Construct a training sample set based on the fault elimination records during the daily operation and maintenance of power OTN equipment ; The training sample set includes the feature vector of the power OTN equipment and the corresponding fault section; (2) Select the feature vector when the power OTN equipment fails and construct the attribute set A; (3) Generate the root node T. If the training sample set If all instances in the tree belong to the same class, then T is a single-node tree, and the instance class is used as the class label of node T; otherwise, the information gain of each feature in the attribute set A is calculated, and the feature attribute with the largest information gain is selected. ; (4) If If it is not greater than the set threshold, then T is a single-node tree, recording the training sample set The category C with the largest number of instances in is used as the class label of the node; if is greater than the set threshold, for Every value of ,make Represents the training sample set exist The upper value is A subset of samples; each Generate a branch node of node T; (5) If , mark the branch node as a leaf node, and set the training sample set The category C with the largest number of instances in is used as the class label of the node; otherwise, As the starting node, calculate the attribute set A except the characteristic attribute The feature attribute with the largest information gain other than the one with the largest information gain is selected, and the process of (4) to (5) is repeated until all the feature attributes in the attribute set A are judged. The trained OTN fault location decision tree is obtained.

2. A method for locating faults in a power optical transmission system based on a decision tree as claimed in claim 1, characterized in that: Also includes: A validation set is constructed based on the fault elimination records of some power OTN equipment during daily operation and maintenance. ; After obtaining the trained OTN fault location decision tree, use the validation set Verify the decision accuracy of each leaf node on the validation set ; Then verify the decision accuracy of the previous node of each leaf node on the validation set If the decision accuracy ≥ , then prune the leaf node to obtain the pruned OTN fault location decision tree as the final decision tree for fault location.

3. A method for locating faults in a power optical transmission system based on a decision tree as claimed in claim 1, characterized in that: The information gain is calculated as follows: Training sample set The information entropy is defined as , The training sample set The proportion of samples of the kth category in ; Assume that the feature attribute a has V possible values , if a is used to train the sample set If the partition is performed, V branch nodes will be generated, of which the vth branch node contains the training sample set All the values ​​of attribute a in The sample is counted as ; Calculate the attribute a for the training sample set The information gain obtained by partitioning is: 。 4. A method for locating faults in a power optical transmission system based on a decision tree as claimed in claim 1, characterized in that: If the training sample set All instances in have no features Or training sample set The samples in the training set have the same value on the attribute set A, then T is a single-node tree. The category C with the largest number of instances is used as the class label of the node.

5. The method for locating faults in a power optical transmission system based on a decision tree according to claim 1, characterized in that: like , mark the branch node as a leaf node, and set the training sample set The category C with the largest number of instances is used as the class label of the node.

6. A fault location system for power optical transmission system based on decision tree, characterized in that: include: The data acquisition module is used to acquire the input optical power, output optical power, combined signal loss MUT_LOS, receiving line side signal loss R_LOS, optical monitoring signal loss OSC_LOS, hardware damage HARD_BARD, input optical power too low IN_PWR_LOW, network element not logged in NE_NOT_LOGIN, network element communication interruption NE_COMMU_BREAK alarm, shutdown DOWN and running UP state feature attribute data of the power OTN equipment, and construct a feature vector; A fault section location module is used to input the constructed feature vector into the trained OTN fault location decision tree and output the fault location section; The OTN fault location decision tree includes a root node, several internal branch nodes and several leaf nodes; each leaf node corresponds to a located fault section, and each other internal branch node corresponds to a characteristic state of the OTN device. The root node contains a full set of samples, and the path from the root node to each leaf node corresponds to a fault location decision logic; The fault location section includes: station A service equipment-station A OTN equipment OTU single board, station A OTN equipment OTU single board-station A OTN equipment FIU single board, station A OTN equipment FIU single board-station B OTN equipment FIU single board, station B OTN equipment OTU single board-station B OTN equipment FIU single board, and station B service equipment-station B OTN equipment OTU single board; The training process of the OTN fault location decision tree includes: (1) Construct a training sample set based on the fault elimination records during the daily operation and maintenance of power OTN equipment ; The training sample set includes the feature vector of the power OTN equipment and the corresponding fault section; (2) Select the feature vector when the power OTN equipment fails and construct the attribute set A; (3) Generate the root node T. If the training sample set If all instances in the tree belong to the same class, then T is a single-node tree, and the instance class is used as the class label of node T; otherwise, the information gain of each feature in the attribute set A is calculated, and the feature attribute with the largest information gain is selected. ; (4) If If it is not greater than the set threshold, then T is a single-node tree, recording the training sample set The category C with the largest number of instances in is used as the class label of the node; if is greater than the set threshold, for Every value of ,make Represents the training sample set exist The upper value is A subset of samples; each Generate a branch node of node T; (5) If , mark the branch node as a leaf node, and set the training sample set The category C with the largest number of instances in is used as the class label of the node; otherwise, As the starting node, calculate the attribute set A except the characteristic attribute The feature attribute with the largest information gain other than the one with the largest information gain is selected, and the process of (4) to (5) is repeated until all the feature attributes in the attribute set A are judged. The trained OTN fault location decision tree is obtained.

7. A terminal device, comprising a processor and a memory, wherein the processor is used to implement each instruction; and the memory is used to store multiple instructions, characterized in that: The instructions are suitable for being loaded by a processor and executing the decision tree-based power optical transmission system fault location method described in any one of claims 1-5.

8. A computer-readable storage medium storing a plurality of instructions, characterized in that: The instructions are suitable for being loaded by a processor of a terminal device and executing the decision tree-based power optical transmission system fault location method described in any one of claims 1-5.