A method and system for predicting the life of an optical cable based on big data

Abstract

The application provides a kind of optical cable life prediction method and system based on big data, first gather multi-source optical cable operation associated data, including historical operation state, laying environment and maintenance record data, then carry out partition associated processing to multi-source optical cable operation associated data, obtain road section associated data set, then based on the road section associated data set, execute associated rule mining, generate road section associated rule set, according to road section associated rule set and preset parameter, generate each road section life influence factor set, finally combine influence factor set and road section associated data set, generate optical cable life prediction result containing each road section residual life interval, so as to comprehensively consider various influence factors, accurately predict optical cable life for different road sections, improve optical cable communication network reliability and stability, reduce operating cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of big data, in particular to a cable life prediction method and system based on big data. BACKGROUND

[0002] At present, with the large-scale construction and application of optical cable communication networks, the reliability and service life of optical cables as the key infrastructure for information transmission are directly related to the stable operation of communication networks. Accurate prediction of the service life of optical cables is of great significance for reasonable arrangement of optical cable maintenance and replacement plans, reduction of communication network failure risks, and cost savings.

[0003] At present, the traditional optical cable life prediction method has many limitations. Some methods only predict based on historical operating status data of optical cables, but the actual service life of optical cables is affected by a variety of factors, and only considering historical operating status data cannot fully reflect the true condition of optical cables. For example, factors such as temperature, humidity, and chemical corrosion in the laying environment can accelerate the aging process of optical cables, and the number of repairs and repair methods in the maintenance record data also have an important impact on the service life of optical cables. In addition, some methods use simple linear models for prediction, without considering the complex nonlinear relationship between different factors, resulting in a large deviation between the prediction results and the actual situation. Moreover, most existing methods do not process the optical cable data in different regions, ignoring the differences in the environment of optical cables in different laying sections, and cannot accurately predict the characteristics of optical cables in different sections. SUMMARY

[0004] In view of the above-mentioned problems, in combination with the first aspect of the present application, the present application embodiment provides a cable life prediction method based on big data, which comprises:

[0005] Gathering multi-source optical cable operation associated data, the multi-source optical cable operation associated data comprising historical operating status data, laying environment data and maintenance record data of optical cables;

[0006] Performing data partitioning and association processing on the multi-source optical cable operation associated data, dividing the multi-source optical cable operation associated data into a plurality of section data groups according to the laying sections of optical cables, obtaining a section associated data set, each section data group containing historical operating status data, laying environment data and maintenance record data of the corresponding section;

[0007] Performing association rule mining processing based on the section associated data set, extracting the association characteristics between historical operating status data and laying environment data, maintenance record data in different sections, and generating a section association rule set, each section association rule in the section association rule set indicating the influence relationship between different data items;

[0008] generate a set of life influence factors of each section according to the set of section association rules and a preset optical cable life benchmark parameter, each factor in the set of life influence factors corresponding to an action strength of an association relationship influencing the life of the optical cable;

[0009] generate an optical cable life prediction result containing a residual life interval of each section in combination with the set of life influence factors of each section and the set of section association data.

[0010] In still another aspect, an embodiment of the present application further provides an optical cable life prediction system based on big data, comprising a processor, a machine readable storage medium, the machine readable storage medium being connected with the processor, the machine readable storage medium being used for storing programs, instructions or codes, and the processor being used for executing the programs, instructions or codes in the machine readable storage medium to realize the above method.

[0011] Based on the above aspects, an embodiment of the present application collects multi-source optical cable operation association data, covers historical operation state, laying environment and maintenance record and other information of the optical cable, performs partition association processing on the multi-source optical cable operation association data, divides data groups according to optical cable laying sections, fully considers the difference of environments of optical cables in different sections, performs association rule mining processing based on the set of section association data, can deeply extract association characteristics between different data items, the generated set of section association rules accurately indicates the influence relationship of each factor on the life of the optical cable, generates a set of life influence factors according to the set of section association rules and a preset optical cable life benchmark parameter, accurately quantifies the action strength of different association relationships on the life of the optical cable, finally generates a prediction result containing a residual life interval of each section in combination with the set of life influence factors and the set of section association data, realizes accurate prediction of the life of optical cables in different sections, effectively improves the reliability and stability of the optical cable communication network, and reduces the operation cost. BRIEF DESCRIPTION OF DRAWINGS

[0012] Figure 1 is a flowchart of an execution process of the optical cable life prediction method based on big data provided by an embodiment of the present application.

[0013] Figure 2 is a schematic diagram of exemplary hardware and software components of the optical cable life prediction system based on big data provided by an embodiment of the present application. DETAILED DESCRIPTION

[0014] The present application will be described in detail below with reference to the accompanying drawings, Figure 1 is a flowchart of an execution process of the optical cable life prediction method based on big data provided by an embodiment of the present application.

[0015] Step S110: gather multi-source optical cable operation related data, including historical operation state data, laying environment data and maintenance record data of the optical cable.

[0016] Step S111: obtain real-time operation state data transmitted by the optical cable monitoring system, including transmission performance data and physical state data of the current period.

[0017] Next, first, real-time operation state data needs to be obtained. The optical cable monitoring system will continuously monitor the optical cable and transmit real-time data. For example, the current period is T0 to T1, the transmission performance data can be the signal attenuation degree of the optical cable at different frequencies f1, f2, f3, which are A1, A2, A3, respectively, and the transmission rate fluctuation, such as the difference between the transmission rate at different time points t0, t0.5, t1 and the standard rate V1, V2, V3. The physical state data includes the damage of the optical cable sheath at different detection points P1, P2, P3, such as the damage area ratio S1, S2, S3, and the bending state of the internal optical fiber at different positions W1, W2, W3, such as the bending angle B1, B2, B3.

[0018] Step S112: retrieve historical operation state data stored in the database, including transmission performance data and physical state data of past periods, including signal attenuation degree and transmission rate fluctuation, and optical cable sheath damage and internal optical fiber bending state.

[0019] Then, historical operation state data needs to be retrieved from the database. These data are accumulated in the past periods, which can reflect the change process of the optical cable operation state. For example, the past periods include T-1 to T0, T-2 to T-1 and other periods. In the period of T-1 to T0, the signal attenuation degree in the transmission performance data at frequencies f1, f2, f3 is A1-1, A2-1, A3-1, respectively, and the transmission rate fluctuation at time points t-1, t-0.5, t0 is V1-1, V2-1, V3-1; the damage area ratio of the optical cable sheath in the physical state data at detection points P1, P2, P3 is S1-1, S2-1, S3-1, and the bending angle of the internal optical fiber at positions W1, W2, W3 is B1-1, B2-1, B3-1. The historical data of other periods also exist in a similar form, which together constitute the set of historical operation state data.

[0020] Step S113: Collect the along-line geographical environment data provided by the geographical information system, the along-line geographical environment data including the terrain type and soil characteristics of the laying area, and obtain the climate data provided by the weather monitoring station, the climate data including the temperature variation range and humidity variation.

[0021] In this embodiment, in addition to the running state data of the optical cable itself, the environmental data where the optical cable is located is also crucial. The terrain type of the optical cable laying area can be obtained from the geographical information system. For example, different road sections on the optical cable laying path may belong to different terrain types such as plains, hills, and mountains. The corresponding terrain parameters can be terrain gradient variation values G1, G2, G3, etc., and altitude variation values H1, H2, H3, etc. As for soil characteristics, through analysis of soil samples along the line, the content of various components in the soil of different road sections is obtained, such as the concentration of corrosive components C1, C2, C3, etc., to reflect the corrosiveness and other characteristics of the soil. At the same time, the climate data is obtained from the weather monitoring station. The temperature variation range can be the difference between the maximum and minimum temperatures in different seasons or months, such as the temperature variation range R1, R2, …, R12 in each month within the time period T-12 to T0. The humidity variation includes humidity values and their durations at different times of the day, such as humidity values M1, M2, …, Mn at different times of the day, each humidity value corresponding to a duration D1, D2, …, Dn.

[0022] Step S114: Extract the maintenance record data in the maintenance management system, the maintenance record data including the maintenance operation type and corresponding maintenance time information of each maintenance, the maintenance operation type including optical cable repair and optical cable reinforcement.

[0023] Then, the maintenance record data needs to be extracted from the maintenance management system. These data record the past maintenance of the optical cable, for example, in the past time, the optical cable has been maintained at positions L1, L2, …, Lk respectively. The operation type of each maintenance can be optical cable repair or optical cable reinforcement, and the corresponding maintenance time information includes the start time and end time, such as the maintenance at position L1, the start time is T1s, the end time is T1e, and the operation type is optical cable repair; the maintenance at position L2, the start time is T2s, the end time is T2e, and the operation type is optical cable reinforcement, etc. These data record the specific circumstances of the maintenance in detail, which is of great significance for analyzing the influence of maintenance on the service life of the optical cable.

[0024] Step S115: The real-time running state data, the historical running state data, the along-line geographical environment data, the climate data, and the maintenance record data are summarized and integrated to form multi-source optical cable running correlation data, each data in the multi-source optical cable running correlation data has a corresponding optical cable identifier and a time marker.

[0025] Then, the real-time running state data, the historical running state data, the along-line geographical environment data, the climate data and the maintenance record data obtained above are integrated. In the integration process, in order to ensure the accuracy and traceability of the data, each item of data is provided with corresponding cable identification, such as the unique cable number ID1, ID2 and the like, and time mark, such as the specific time stamp of data generation or record. For example, the signal attenuation degree A1 in the real-time running state data corresponds to the cable identification ID001 and the time mark T0.2; the transmission rate fluctuation difference V1-1 in the historical running state data corresponds to the cable identification ID001 and the time mark T-0.8; the terrain slope change value G1 in the along-line geographical environment data corresponds to the cable identification ID001 and the time mark Tf of the laying time; the temperature change range R1 in the climate data corresponds to the cable identification ID001 and the time mark T-11; the maintenance at the position L1 in the maintenance record data corresponds to the cable identification ID001 and the time mark T1s to T1e and the like. Through the above integration, the multi-source cable running correlation data is formed.

[0026] Step S120: performing data partition correlation processing on the multi-source cable running correlation data, dividing the multi-source cable running correlation data into a plurality of road section data groups according to the laying sections of the cable, to obtain a road section correlation data set, each road section data group containing the historical running state data, laying environment data and maintenance record data of the corresponding road section.

[0027] In this embodiment, since the laying path of the cable is usually long, the environment and running conditions of the cable in different road sections may be different, and therefore the multi-source cable running correlation data needs to be processed by partition correlation. By dividing the data groups according to the laying road sections, the conditions of each road section can be more accurately analyzed.

[0028] Step S121: analyzing the cable identification in the multi-source cable running correlation data to determine the overall laying path of the cable and the road section division nodes on the path, the road section division nodes being determined based on the geographical boundaries of the laying area and the maintenance responsibility partition.

[0029] First, the cable identifier in the multi-source cable operation correlation data is parsed, such as through the cable identifier ID001, it can be determined that the overall laying path of the cable is from the starting point S to the ending point E. Then, according to the geographical boundaries of the laying area, such as natural or artificial boundaries of rivers, mountains, roads, etc., and the maintenance responsibility partition, such as the area responsible by different maintenance teams, the road section division nodes are determined. For example, on the overall laying path, node N1 is located on the south bank of river A, node N2 is located on the east side of road B, node N3 is located at the responsibility boundary of maintenance team A and maintenance team B, etc., these nodes divide the overall laying path into road sections S-N1, N1-N2, N2-N3, N3-E.

[0030] Step S122: According to the road section division nodes, each item of data in the multi-source cable operation correlation data is allocated to the corresponding road section partition, and a preliminary road section data group is generated, each road section data group contains the historical operation state data, laying environment data and maintenance record data of the road section.

[0031] Then, according to the determined road section division nodes, each item of data in the multi-source cable operation correlation data is allocated to the corresponding road section partition. For example, for the road section S-N1, the data allocated to this road section includes the historical operation state data within this road section, such as signal attenuation degree, transmission rate fluctuation, etc.; laying environment data, such as the terrain type of this road section, soil characteristics, temperature variation range, humidity variation, etc.; maintenance record data, such as the type of maintenance operation performed on this road section and the maintenance time information, etc. Similarly, other road sections also allocate data in a similar manner to generate preliminary road section data groups, such as road section S-N1 data group, road section N1-N2 data group, etc.

[0032] Step S123: In each road section data group, an association index between the historical operation state data and the laying environment data is established, which is used to indicate the corresponding relationship between the transmission performance data, physical state data and the temperature variation range, humidity variation, terrain type, soil characteristics in the same time interval.

[0033] In this embodiment, in each road section data packet, in order to clarify the relationship between the historical running state data and the laying environment data, an association index needs to be established. For example, in the road section S-N1 data group, the time interval is T-3 to T-2, the signal attenuation degree in the transmission performance data is A1-3, A2-3, and A3-3 at frequencies f1, f2, and f3, respectively, and the transmission rate fluctuation difference is V1-3, V2-3, and V3-3; the outer skin damage area ratio in the physical state data is S1-3, S2-3, and S3-3, and the internal optical fiber bending angle is B1-3, B2-3, and B3-3. In the corresponding laying environment data, the temperature change range is R-3, the humidity value in the humidity change is M1-3 and M2-3, and the duration is D1-3 and D2-3; the terrain type is plain, the terrain slope change value is G-3, and the altitude change value is H-3; the corrosion component concentration in the soil property is C-3. The established association index will associate these data in the same time interval, so as to analyze the influence of the environment on the optical cable running state in subsequent.

[0034] Step S124: Establish an association index between the historical running state data and the maintenance record data, the association index being used to indicate the corresponding relationship between the maintenance operation type, the maintenance time information, and the transmission performance data and the physical state data before and after the maintenance.

[0035] Then, an association index between the historical running state data and the maintenance record data is established. For example, in the road section N1-N2 data group, there is a maintenance operation, the maintenance operation type is optical cable reinforcement, the maintenance start time is Tms, and the end time is Tme. The time interval before the maintenance is Tms-Δt to Tms, the signal attenuation degree in the transmission performance data is Ams-1, Ams-2, and Ams-3, and the transmission rate fluctuation difference is Vms-1, Vms-2, and Vms-3; the outer skin damage area ratio in the physical state data is Sms-1, Sms-2, and Sms-3, and the internal optical fiber bending angle is Bms-1, Bms-2, and Bms-3. The time interval after the maintenance is Tme to Tme+Δt, and the transmission performance data and the physical state data are Ame-1, Ame-2, Ame-3, Vme-1, Vme-2, Vme-3, Sme-1, Sme-2, Sme-3, Bme-1, Bme-2, and Bme-3, respectively. The association index will correspond the maintenance operation type, the maintenance time information, and these data before and after the maintenance, and accurately show the influence of the maintenance on the optical cable running state.

[0036] Step S125: Integrate the plurality of road section data groups with the association index to form a road section association data set, the road section association data set being able to reflect the time correlation and influence correlation between different types of data in each road section.

[0037] Finally, all the road section data groups associated with the indexes are integrated to form a road section association data set. The data set integrates all the relevant data of each road section and their association, and can accurately reflect the time association and influence association between the historical running state data, laying environment data and maintenance record data in each road section. For example, through the road section association data set, it can be quickly queried that in the road section N2-N3, how the transmission performance data of the optical cable changes under the high temperature and high humidity environment in a time period, and the improvement of the physical state data after the optical cable repair operation, etc.

[0038] Step S130: performing association rule mining processing based on the road section association data set, extracting the association characteristics between the historical running state data and the laying environment data and the maintenance record data in different road sections, and generating a road section association rule set, each road section association rule in the road section association rule set being used to indicate the influence relationship between different data items.

[0039] In this embodiment, after obtaining the road section association data set, it needs to be subjected to association rule mining processing to find out the influence relationship between different data items. By mining these association rules, the influence of laying environment and maintenance operation on the running state of the optical cable can be deeply known.

[0040] Step S131: selecting a road section data group from the road section association data set as a current processing object, and extracting the historical running state data, laying environment data and maintenance record data in the road section data group.

[0041] Firstly, a road section data group is selected from the road section association data set, for example, the road section S-N1 data group is selected as the current processing object. Then, the historical running state data is extracted from the data group, such as the signal attenuation degree, transmission rate fluctuation, optical cable outer skin damage, internal optical fiber bending state, etc. in the past multiple time periods; the laying environment data, such as the temperature variation range, humidity variation, terrain type, soil characteristics, etc. of the road section; the maintenance record data, such as the maintenance operation type, maintenance time information, etc. performed on the road section.

[0042] Step S132: performing state change trend extraction on the transmission performance data and the physical state data in the historical running state data, and determining the change trend of the transmission performance data and the change trend of the physical state data.

[0043] Then, the state change trend extraction is performed on the transmission performance data and the physical state data in the extracted historical running state data.

[0044] Step S1321: extracting the transmission performance data from the historical running state data, the transmission performance data including the signal attenuation degree and the transmission rate fluctuation at multiple time points.

[0045] Transmission performance data is extracted from the historical operating status data set for the road section S-N1. This data includes signal attenuation and transmission rate fluctuations at multiple time points. For example, the signal attenuation at time points t1, t2, ..., tn is A1, A2, ..., An (corresponding to different frequencies), and the transmission rate fluctuation is V1, V2, ..., Vn (corresponding to the fluctuation difference at different time points).

[0046] Step S1322: Arrange the signal attenuation degrees in chronological order, calculate the change in the signal attenuation degrees between adjacent time points, and determine the change trend of the signal attenuation in the transmission performance data based on the size of the change. If the change is within a preset small range, it is a stable state; if the change is within a preset medium range, it is a slowly decreasing state; if the change exceeds the preset medium range, it is a rapidly decreasing state.

[0047] Arrange the signal attenuation in the time sequence t1, t2, ..., tn to obtain A1, A2, ..., An. Calculate the change between adjacent time points, such as the difference between A2 and A1, ΔA1 = A2 - A1, the difference between A3 and A2, ΔA2 = A3 - A2, and so on, to obtain ΔA1, ΔA2, ..., ΔA(n-1). The preset small range is [ΔAmin1, ΔAmid1), and the preset medium range is [ΔAmid1, ΔAmax1]. When ΔAi is in [ΔAmin1, ΔAmid1), the signal attenuation trend is stable; when ΔAi is in [ΔAmid1, ΔAmax1), it is in a slowly decreasing state; when ΔAi ≥ ΔAmax1, it is in a rapidly decreasing state. For example, if ΔA1 is within a preset small range, the signal attenuation from t1 to t2 is stable; if ΔA2 is within a preset medium range, the signal attenuation from t2 to t3 is slowly decreasing; if ΔA3 exceeds the preset medium range, the signal attenuation from t3 to t4 is rapidly decreasing, and so on.

[0048] Step S1323: Arrange the transmission rate fluctuations in chronological order, calculate the change amplitude of the transmission rate fluctuations between adjacent time points, and determine the change trend of the transmission rate fluctuations in the transmission performance data according to the magnitude of the change amplitude. The change amplitude within the preset small range is a stable state, the change amplitude within the preset medium range is a slowly decreasing state, and the change amplitude exceeding the preset medium range is a rapidly decreasing state.

[0049] Similarly, the transmission rate fluctuation is arranged in time sequence t1, t2, …, tn to obtain V1, V2, …, Vn. The change amplitude between adjacent time points is calculated, such as the difference AV1=V2-V1 between V2 and V1, the difference AV2=V3-V2 between V3 and V2, and so on, to obtain AV1, AV2, …, AV(n-1). A preset small range is [AVmin1, AVmid1), and a preset medium range is [AVmid1, AVmax1). When AVi is in [AVmin1, AVmid1), the change trend of the transmission rate fluctuation is a stable state; when AVi is in [AVmid1, AVmax1), it is a slow decline state; and when AVi≥AVmax1, it is a rapid decline state. For example, AV1 is in the preset small range, and the transmission rate fluctuation is in a stable state during the t1-t2 period; AV2 is in the preset medium range, and the transmission rate fluctuation is in a slow decline state during the t2-t3 period; AV3 exceeds the preset medium range, and the transmission rate fluctuation is in a rapid decline state during the t3-t4 period, and so on.

[0050] Step S1324: The change trend of the signal attenuation and the change trend of the transmission rate fluctuation are integrated to determine the overall change trend of the transmission performance data. When both are in a stable state, the overall change trend is stable; when at least one is in a slow decline state and there is no rapid decline state, the overall change trend is a slow decline; and when at least one is in a rapid decline state, the overall change trend is a rapid decline.

[0051] Exemplarily, in the road section S-N1 data set, after the processing of steps S1322 and S1323, the change trend of the signal attenuation is in a stable state during the t1-t2 period, a slow decline state during the t2-t3 period, and a rapid decline state during the t3-t4 period; and the change trend of the transmission rate fluctuation is in a stable state during the t1-t2 period, a stable state during the t2-t3 period, and a slow decline state during the t3-t4 period. Then, during the t1-t2 period, both are in a stable state, so the overall change trend of the transmission performance data is stable; during the t2-t3 period, the signal attenuation is in a slow decline state, and the transmission rate fluctuation is in a stable state, which satisfies at least one in a slow decline state and no rapid decline state, so the overall change trend is a slow decline; and during the t3-t4 period, the signal attenuation is in a rapid decline state, and no matter the state of the transmission rate fluctuation, the overall change trend is a rapid decline. Through the above integrated judgment, the overall change trend of the transmission performance data in different periods can be obtained.

[0052] Step S1325: Physical state data is extracted from the historical running state data, and the physical state data includes the cable outer skin damage condition and the internal optical fiber bending state at multiple time points.

[0053] The physical state data is extracted from the historical running state data of the section S-N1 data set, which contains the cable sheath damage conditions and internal fiber bending states at multiple time points. For example, the cable sheath damage conditions corresponding to the time points t1, t2, …, tn are S1, S2, …, Sn (the damage area proportions of different detection points, respectively), and the internal fiber bending states are B1, B2, …, Bn (the bending angles at different positions, respectively).

[0054] Step S1326: The cable sheath damage conditions are arranged in time sequence, the change of the cable sheath damage degree between adjacent time points is calculated, the change trend of the cable sheath damage in the physical state data is determined according to the change degree, the change degree within the preset small range is the stable state, the change degree within the preset medium range is the slow decline state, and the change degree exceeding the preset medium range is the rapid decline state.

[0055] The cable sheath damage conditions are arranged in time sequence t1, t2, …, tn, and S1, S2, …, Sn are obtained. The change degree between adjacent time points is calculated, such as the difference ΔS1 between S2 and S1, the difference ΔS2 between S3 and S2, and so on, to obtain ΔS1, ΔS2, …, ΔS(n-1). The preset small range is [ΔSmin1, ΔSmid1), and the preset medium range is [ΔSmid1, ΔSmax1). When ΔSi is in [ΔSmin1, ΔSmid1), the change trend of the cable sheath damage is stable; when ΔSi is in [ΔSmid1, ΔSmax1), it is a slow decline state; and when ΔSi≥ΔSmax1, it is a rapid decline state. For example, ΔS1 is within the preset small range, then the cable sheath damage is stable during the period from t1 to t2; ΔS2 is within the preset medium range, then the cable sheath damage is slowly declining during the period from t2 to t3; ΔS3 exceeds the preset medium range, then the cable sheath damage is rapidly declining during the period from t3 to t4, and so on.

[0056] Step S1327: The internal fiber bending state is arranged in time sequence, the change of the internal fiber bending degree between adjacent time points is calculated, the change trend of the internal fiber bending in the physical state data is determined according to the change degree, the change degree within the preset small range is the stable state, the change degree within the preset medium range is the slow decline state, and the change degree exceeding the preset medium range is the rapid decline state.

[0057] Similarly, the internal fiber bending states are arranged in time sequence t1, t2, …, tn to obtain B1, B2, …, Bn. The change degree between adjacent time points is calculated, such as the difference ΔB1 = B2 - B1 between B2 and B1, the difference ΔB2 = B3 - B2 between B3 and B2, and so on, to obtain ΔB1, ΔB2, …, ΔB(n-1). A preset small range is [ΔBmin1, ΔBmid1), and a preset medium range is [ΔBmid1, ΔBmax1). When ΔBi is in [ΔBmin1, ΔBmid1), the change trend of the internal fiber bending is a stable state; when ΔBi is in [ΔBmid1, ΔBmax1), it is a slow decline state; and when ΔBi ≥ ΔBmax1, it is a rapid decline state. For example, ΔB1 is in the preset small range, and the internal fiber bending is in a stable state during the period from t1 to t2; ΔB2 is in the preset medium range, and the internal fiber bending is in a slow decline state during the period from t2 to t3; ΔB3 exceeds the preset medium range, and the internal fiber bending is in a rapid decline state during the period from t3 to t4, and so on.

[0058] Step S1328: The change trends of the optical cable outer skin damage and the internal fiber bending are integrated to determine the overall change trend of the physical state data. When both are in a stable state, the overall change trend is stable; when at least one is in a slow decline state and there is no rapid decline state, the overall change trend is a slow decline; and when at least one is in a rapid decline state, the overall change trend is a rapid decline.

[0059] For example, in the road section S-N1 data set, the change trend of the optical cable outer skin damage is in a stable state during the period from t1 to t2, in a slow decline state during the period from t2 to t3, and in a rapid decline state during the period from t3 to t4; and the change trend of the internal fiber bending is in a stable state during the period from t1 to t2, in a slow decline state during the period from t2 to t3, and in a stable state during the period from t3 to t4. Then, during the period from t1 to t2, both are in a stable state, and the overall change trend of the physical state data is stable; during the period from t2 to t3, both are in a slow decline state, which satisfies that at least one is in a slow decline state and there is no rapid decline state, and the overall change trend is a slow decline; and during the period from t3 to t4, the optical cable outer skin damage is in a rapid decline state, so the overall change trend is a rapid decline. Through the above integration, the overall change trend of the physical state data in different periods is obtained.

[0060] Step S1329: The overall change trend of the transmission performance data and the overall change trend of the physical state data are combined to obtain the change trend of the historical operation state data.

[0061] The overall change trend of the transmission performance data obtained in step S1324 is combined with the overall change trend of the physical status data obtained in step S1328 to obtain the change trend of the historical operating status data. For example, during the period t1-t2, the transmission performance data is generally stable, and the physical status data is generally stable, so the change trend of the historical operating status data is stable; during the period t2-t3, the transmission performance data is generally slowly decreasing, and the physical status data is generally slowly decreasing, so the change trend of the historical operating status data is slowly decreasing; during the period t3-t4, the transmission performance data is generally rapidly decreasing, and the physical status data is generally rapidly decreasing, so the change trend of the historical operating status data is rapidly decreasing. Through this combination, the changes in the historical operating status of the optical cable are fully reflected.

[0062] Step S133: Extract environmental features from the temperature variation range, humidity variation, terrain type, and soil characteristics in the installation environment data to determine the characteristic manifestations of different environmental factors, including temperature fluctuation amplitude, humidity duration interval, terrain stability, and soil corrosivity.

[0063] In this embodiment, in order to deeply analyze the impact of the laying environment on the operating status of the optical cable, it is necessary to extract environmental features from the laying environment data.

[0064] For example, step S1331: extract the temperature variation range from the installation environment data, divide the temperature variation range into multiple continuous temperature intervals in chronological order, calculate the difference between the maximum and minimum temperature in each temperature interval, and obtain the temperature fluctuation amplitude of each temperature interval.

[0065] Extract the temperature variation range from the installation environment data in the road section S-N1 data group. For example, divide time into multiple consecutive intervals, each ΔT, such as t1-t1+ΔT, t1+ΔT-t1+2ΔT, and so on. Calculate the difference between the maximum and minimum temperature values ​​within each interval. For example, in the interval t1-t1+ΔT, the maximum temperature is Tmax1 and the minimum is Tmin1. Then, the temperature fluctuation range in this interval is F1=Tmax1-Tmin1. In the interval t1+ΔT-t1+2ΔT, the temperature fluctuation range is F2=Tmax2-Tmin2. And so on, to obtain multiple temperature fluctuation ranges F1, F2, ..., Fm.

[0066] Step S1332: Count the duration of each temperature fluctuation amplitude, accumulate the time when the same temperature fluctuation amplitude appears continuously, obtain the duration of the temperature fluctuation amplitude, and generate a temperature fluctuation feature based on the temperature fluctuation amplitude and its duration.

[0067] The time length of each temperature fluctuation amplitude is counted, for example, the temperature fluctuation amplitude F1 appears in the time interval t1-t1+AT1, t2-t2+AT2, etc., the time lengths are accumulated to obtain the duration D_F1 of F1; similarly, the duration D_F2 of F2 is obtained, etc. Then, the temperature fluctuation amplitude is combined with the corresponding duration to generate the temperature fluctuation feature, such as (F1, D_F1), (F2, D_F2), …, (Fm, D_Fm), which can reflect the size and duration of the temperature fluctuation.

[0068] Step S1333: Extract the humidity change from the laying environment data, divide it into different humidity levels according to the size of the humidity value, record the time interval of each humidity level, and obtain the humidity duration interval.

[0069] The humidity change is extracted from the laying environment data, and the humidity value is divided into different levels according to certain standards, such as low humidity, medium humidity, and high humidity. For example, the humidity value in [0, 30%) is the low humidity level H_L, [30%, 70%) is the medium humidity level H_M, and [70%, 100%] is the high humidity level H_H. Then, the time interval of each humidity level is recorded, such as the low humidity level duration time interval [t1s, t1e), [t2s, t2e), etc., the medium humidity level duration time interval [t3s, t3e), etc., and the high humidity level duration time interval [t4s, t4e), etc., which constitute the humidity duration interval.

[0070] Step S1334: Analyze the change frequency of the humidity level in the humidity duration interval, calculate the number of conversions between adjacent humidity levels, and generate the humidity change feature.

[0071] The change frequency of the humidity level in the humidity duration interval is analyzed, for example, the number of conversions from low humidity level to medium humidity level is C_LM, the number of conversions from medium humidity level to high humidity level is C_MH, the number of conversions from high humidity level to medium humidity level is C_HM, and the number of conversions from medium humidity level to low humidity level is C_ML, etc. The conversion number is combined with the humidity duration interval to generate the humidity change feature, such as (H_L duration interval, C_LM), (H_M duration interval, C_MH, C_ML), (H_H duration interval, C_HM), etc., to reflect the humidity change.

[0072] Step S1335: Extract the terrain type from the laying environment data, determine the terrain stability parameter based on the slope and altitude change of the terrain, and the terrain stability parameter is calculated by the change amount of the terrain slope and the change amount of the altitude in a unit of time.

[0073] The terrain type of the road section S-N1, such as a plain, is extracted from the laying environment data. Then, the stability parameter of the terrain is calculated according to the slope change and the altitude change of the terrain. For example, the change amount of the terrain slope in a unit time Δt is ΔG, the change amount of the altitude is ΔH, and the stability parameter K of the terrain is obtained by normalizing ΔG and ΔH and combining them according to certain weights, such as K = a × (ΔG / ΔGmax) + b × (ΔH / ΔHmax), where a and b are weight coefficients, and a + b = 1, and ΔGmax and ΔHmax are the preset maximum slope change amount and maximum altitude change amount, respectively. Through the above calculation, the stability parameter of the terrain is obtained.

[0074] Step S1336: The terrain stability grade is divided according to the stability parameter of the terrain, and the terrain stability feature is generated.

[0075] The division standard of the preset terrain stability grade is, for example, when K < K1, it is a high stability grade; when K1≤K < K2, it is a medium stability grade; and when K ≥ K2, it is a low stability grade. According to the K value calculated in step S1335, the terrain stability grade of the road section S-N1 is determined, such as a high stability grade, and the terrain stability feature is generated, such as (high stability grade, K value).

[0076] Step S1337: The soil properties are extracted from the laying environment data, and the corrosion parameter of the soil is determined through the component analysis of the soil sample, wherein the corrosion parameter of the soil is determined based on the content and activity degree of the corrosive components in the soil.

[0077] The soil properties of the road section S-N1 are extracted from the laying environment data, and the content of the corrosive components such as sulfate and chloride ions in the soil and the activity degree of these components are determined through the component analysis of the soil sample. For example, the content of the corrosive components is C_s, the activity degree is A_s, and the corrosion parameter P of the soil is obtained by normalizing and combining C_s and A_s, such as P = c × (C_s / C_smax) + d × (A_s / A_smax), where c and d are weight coefficients, and c + d = 1, and C_smax and A_smax are the preset maximum corrosive component content and maximum activity degree, respectively.

[0078] Step S1338: The soil corrosion grade is divided according to the corrosion parameter of the soil, and the soil corrosion feature is generated.

[0079] Preset soil corrosivity classification standards, such as when P <P1时,为低腐蚀性等级;当P1≤P<P2时,为中腐蚀性等级;当P≥P2时,为高腐蚀性等级。根据步骤S1337计算得到的P值,确定路段S-N1的土壤腐蚀性等级,如中腐蚀性等级,生成土壤腐蚀性特征,如(中腐蚀性等级,P值)。

[0080] Step S1339: Integrate temperature fluctuation characteristics, humidity change characteristics, terrain stability characteristics, and soil corrosivity characteristics to obtain characteristic expressions of different environmental factors.

[0081] The temperature fluctuation characteristics obtained in step S1332, the humidity change characteristics obtained in step S1334, the terrain stability characteristics obtained in step S1336, and the soil corrosivity characteristics obtained in step S1338 are integrated to obtain characteristic representations of different environmental factors. For example, the environmental characteristics of section S-N1 are as follows: (temperature fluctuation characteristics: (F1, D_F1), (F2, D_F2); humidity change characteristics: (H_L duration interval, C_LM), (H_M duration interval, C_MH, C_ML); terrain stability characteristics: (high stability level, K value); soil corrosivity characteristics: (medium corrosivity level, P value)).

[0082] Step S134: Extract maintenance features from the maintenance operation type and maintenance time information in the maintenance record data to determine the frequency of the maintenance operation and the duration of the effect of the maintenance operation. The duration of the effect of the maintenance operation is determined by the time that the transmission performance data and physical status data remain stable after maintenance.

[0083] Step S1341: Counting the frequency of maintenance operations from the maintenance record data, where the frequency of maintenance operations is the number of maintenance operations per unit time.

[0084] From the maintenance records for section S-N1, count the number of maintenance operations performed within a unit time ΔT. For example, within the time interval [T_start, T_end], the total duration is T = T_end - T_start, and the number of maintenance operations is N. Therefore, the maintenance operation frequency F = N / T.

[0085] Step S1342: Determine the duration of the maintenance operation effect based on the transmission performance data and physical status data after maintenance, that is, the time interval from the maintenance end time to the first time that the transmission performance data or physical status data appears in an unstable state.

[0086] For example, in the road section S-N1, the end time of a certain maintenance operation is Tme, the transmission performance data and the physical state data remain stable after the maintenance until the time Tchange, at which the transmission performance data first appears a slow decline state, and then the effect duration of the maintenance operation is AT_effect=Tchange-Tme. The effect duration of each maintenance operation is obtained by performing the above calculation on each maintenance operation.

[0087] Step S1343: integrating the maintenance operation type, the frequency of the maintenance operation and the effect duration of the maintenance operation to obtain the maintenance feature.

[0088] The maintenance operation type (such as optical cable repair, optical cable reinforcement), the frequency of the maintenance operation obtained in step S1341 and the effect duration of the maintenance operation obtained in step S1342 are integrated to obtain the maintenance feature. For example, the maintenance features of the road section S-N1 are (maintenance operation type: optical cable repair, frequency: F1, effect duration: AT1), (maintenance operation type: optical cable reinforcement, frequency: F2, effect duration: AT2), etc.

[0089] Step S135: analyzing the mutual influence relationship between the state change trend and the environmental feature performance, determining the law that the state change trend presents a corresponding state when the environmental feature performance is a certain condition, and recording the law as a first association rule.

[0090] For example, in the road section S-N1, it is found by analysis that when the environmental feature performance is that the temperature fluctuation amplitude is large and the duration is long, the humidity is high and the duration is long, the terrain stability is low, and the soil corrosion is high, the change trend of the historical running state data is rapid decline; when the environmental feature performance is that the temperature fluctuation amplitude is small, the humidity is medium, the terrain stability is high, and the soil corrosion is low, the change trend of the historical running state data is stable. These laws are recorded as the first association rule, such as “if the environmental feature performance is (F large, D long, H high, D high, terrain low stability, soil high corrosion), then the historical running state change trend is rapid decline”.

[0091] Step S136: analyzing the mutual influence relationship between the state change trend and the maintenance feature, determining the law that the state change trend presents a corresponding state when the frequency and the effect duration of the maintenance operation are a certain condition, and recording the law as a second association rule.

[0092] For example, in the road section S-N1, it is analyzed that when the maintenance operation type is optical cable reinforcement, the frequency is high, and the effect duration is long, the change trend of the historical running state data is stable; when the maintenance operation type is optical cable repair, the frequency is low, and the effect duration is short, the change trend of the historical running state data is slowly decreasing. These rules are recorded as the second association rule, such as "if the maintenance feature is (reinforcement, high frequency, long effect duration), then the change trend of the historical running state is stable".

[0093] Step S137: Perform the above processing on each road section data group in the road section association data set to obtain the first association rule and the second association rule of each road section.

[0094] For other road section data groups in the road section association data set, such as the road section N1-N2 data group, the road section N2-N3 data group, etc., perform the operations of steps S131 to S136 in the same manner as the road section S-N1 data group. Taking the road section N1-N2 data group as an example, first extract the historical running state data of the road section, wherein the transmission performance data includes signal attenuation degrees A_N1N2_t1, A_N1N2_t2, …, A_N1N2_tn (corresponding to different frequencies f1, f2, f3) and transmission rate fluctuations V_N1N2_t1, V_N1N2_t2, …, V_N1N2_tn at different time points; the physical state data includes optical cable skin damage S_N1N2_t1, S_N1N2_t2, …, S_N1N2_tn and internal optical fiber bending state B_N1N2_t1, B_N1N2_t2, …, B_N1N2_tn. Then extract the change trend of these data to determine the change trend of the transmission performance data and the change trend of the physical state data, wherein the change trend of the transmission performance data may be that it slowly decreases in some time periods due to environmental changes, and becomes stable after maintenance, etc.

[0095] Subsequently, the laying environment data of the road section is subjected to environment feature extraction, and the temperature change range may exhibit seasonal fluctuations, such as larger temperature fluctuation amplitude in summer and smaller temperature fluctuation amplitude in winter; the humidity change may be in a high humidity state during the rainy season; the terrain type is hilly, and the terrain stability parameter shows slight slope change; the soil property has a moderate content of corrosive components. Based on these environment feature performances, analyze the relationship with the change trend, for example, when the temperature is high in summer and the humidity is continuously high, the change trend of the transmission performance data is slowly decreasing, thereby forming the first association rule of the road section.

[0096] Maintenance records indicate that this section of road likely underwent multiple fiber optic cable reinforcement operations, resulting in a high maintenance frequency and a long duration of maintenance effects. Analyzing the relationship between these maintenance characteristics and state change trends reveals that, for example, when maintenance operations are frequent, the trend in physical state data is more likely to remain stable, thus forming a second association rule for this section of road.

[0097] Similarly, when processing the data set for section N2-N3, we can extract the corresponding state change trends, environmental characteristics, and maintenance features based on its historical operating status data, installation environment data (e.g., mountainous terrain with highly corrosive soil), and maintenance records (e.g., frequent optical cable repair operations). This analysis then yields the first and second association rules for this section. By processing each section's data set one by one, we ultimately derive the first and second association rules for all sections.

[0098] Step S138: Summarize the first association rules and the second association rules of all road sections, remove duplicate rule content, and generate a road section association rule set. Each road section association rule in the road section association rule set contains a premise and a conclusion. The premise is an environmental feature performance or maintenance feature, and the conclusion is a corresponding state change trend.

[0099] After obtaining the first and second association rules for each road segment, all of these rules are aggregated. During the aggregation process, the rule content can be compared to remove duplicate rules. For example, road segments S-N1 and N1-N2 may have the same first association rule: "When temperature fluctuates significantly and humidity remains high, transmission performance data tends to slowly decline." In this case, only one of this rule needs to be retained.

[0100] The resulting set of road segment association rules includes all unique rules, each with clear preconditions and conclusions. For example, the preconditions for a first association rule are "temperature fluctuations exceeding X, humidity consistently above level Y, terrain stability parameter Z, and soil corrosivity level W," with the conclusion "transmission performance data trend is rapidly declining." A second association rule has the preconditions "maintenance operation type is optical cable repair, maintenance frequency is monthly, and effect duration is 30 days," with the conclusion "physical status data trend is stable." These rules comprehensively reflect the correlation between environmental and maintenance factors and trends in optical cable operating conditions across different road segments.

[0101] Step S140: generating a life impact factor set for each section according to the section association rule set and preset optical cable life reference parameters, wherein each factor in the life impact factor set corresponds to an intensity of an association relationship affecting the optical cable life.

[0102] In this embodiment, after obtaining the set of road segment association rules, the set of life influence factors of each road segment is generated in combination with the preset optical cable life benchmark parameter. The life influence factor can quantify the influence degree of the association relationship indicated by each association rule on the life of the optical cable.

[0103] Step S141: Obtain a preset optical cable life benchmark parameter, which is used to represent the expected service life of the optical cable under ideal environment and no maintenance.

[0104] The preset optical cable life benchmark parameter is determined in an ideal state, that is, the expected service life when the optical cable is in an environment with stable temperature, appropriate humidity, stable terrain, and non-corrosive soil, and no maintenance operation is required. For example, the benchmark parameter can be a time length, such as L0, which is in years, representing the time during which the optical cable can normally operate under ideal conditions.

[0105] Step S142: Select an association rule from the set of road segment association rules as a current processing rule, and analyze the premise condition and conclusion of the current processing rule.

[0106] An association rule is selected from the set of road segment association rules, for example, the rule "when the temperature fluctuation amplitude is large and the humidity is high, the transmission performance data change trend is slow decline" is selected as the current processing rule. The premise condition of the rule is "temperature fluctuation amplitude is large and humidity is high" (belongs to environmental feature performance), and the conclusion is "transmission performance data change trend is slow decline".

[0107] Step S143: According to the environmental feature performance or maintenance feature in the premise condition, in combination with the preset influence strength evaluation standard, the influence strength corresponding to the current processing rule is determined.

[0108] Step S1431: When the premise condition of the current processing rule is environmental feature performance, analyze the temperature change range, humidity change, terrain type, and soil characteristics in the environmental feature performance.

[0109] For the current processing rule, the premise condition is environmental feature performance, and the temperature change range (temperature fluctuation amplitude is large), humidity change (humidity is high and lasts for a long time), terrain type (assuming it is a plain and the terrain is stable), and soil characteristics (assuming the soil corrosion is low) are analyzed.

[0110] Step S1432: According to the preset influence strength evaluation standard, the greater the temperature change range, the higher the corresponding influence strength score; the higher the humidity in the humidity change and the longer the duration, the higher the influence strength score; the more unstable the terrain type, the higher the influence strength score; the stronger the soil corrosion, the higher the influence strength score.

[0111] In the preset influence intensity evaluation standard, different fluctuation amplitudes are set corresponding to scores for temperature change ranges, such as a score of 1 when the temperature fluctuation amplitude is 0-5℃, a score of 2 when the temperature fluctuation amplitude is 5-10℃, and a score of 3 when the temperature fluctuation amplitude is more than 10℃, that is, the greater the temperature change range, the higher the score. For humidity change conditions, a score of 1 is given when the humidity is below 30% and the duration is less than 10 days, a score of 2 is given when the humidity is 30%-60% and the duration is 10-30 days, and a score of 3 is given when the humidity is more than 60% and the duration is more than 30 days, that is, the higher the humidity and the longer the duration, the higher the score. For terrain types, a score of 1 is given for plains (stable), a score of 2 is given for hills (relatively unstable), and a score of 3 is given for mountains (unstable), that is, the more unstable the terrain, the higher the score. For soil properties, a score of 1 is given for low corrosiveness, a score of 2 is given for medium corrosiveness, and a score of 3 is given for high corrosiveness, that is, the stronger the corrosiveness, the higher the score.

[0112] For the premise condition of the current processing rule, the temperature fluctuation amplitude is large (corresponding to a score of 3), the humidity is high (corresponding to a score of 3), the terrain type is plains (a score of 1), and the soil property is low corrosiveness (a score of 1).

[0113] Step S1433: Add the influence intensity scores corresponding to the temperature change range, humidity change condition, terrain type, and soil property to obtain a total influence intensity score corresponding to the environmental feature performance, and the size of the total influence intensity score corresponds to the high and low of the influence intensity.

[0114] Add the scores corresponding to the above environmental feature performances, that is, 3+3+1+1=8, to obtain a total influence intensity score corresponding to the environmental feature performance, which is 8. The higher the total influence intensity score, the higher the influence intensity of the environmental feature performance on the optical cable life.

[0115] Step S1434: When the premise condition of the current processing rule is a maintenance feature, analyze the maintenance operation type, the frequency of the maintenance operation, and the effect duration of the maintenance operation in the maintenance feature.

[0116] Suppose another association rule "when the maintenance operation type is optical cable reinforcement, the maintenance frequency is high, and the effect duration is long, the physical state data change trend is stable" is selected as the current processing rule, and the premise condition is the maintenance feature. It is analyzed that the maintenance operation type is optical cable reinforcement, the frequency of the maintenance operation is high, and the effect duration of the maintenance operation is long.

[0117] Step S1435: According to the preset influence intensity evaluation standard, the influence intensity score corresponding to the repair type operation in the maintenance operation type is lower than that of the reinforcement type operation; the higher the frequency of the maintenance operation, the lower the influence intensity score; the longer the effect duration of the maintenance operation, the lower the influence intensity score.

[0118] In the preset influence intensity evaluation standard, in terms of the type of maintenance operation, the score corresponding to cable repair is 2; the score corresponding to cable reinforcement is 3. In terms of the frequency of maintenance operation, the score for low frequency (less than 1 time per year) is 3; the score for medium frequency (1-3 times per year) is 2; the score for high frequency (more than 3 times per year) is 1, and the higher the frequency, the lower the score. In terms of the duration of the effect of maintenance operation, the score for short duration (less than 3 months) is 3; the score for medium duration (3-12 months) is 2; the score for long duration (more than 12 months) is 1, and the longer the duration, the lower the score.

[0119] For the current processing rule, the type of maintenance operation is cable reinforcement (score 3), the frequency of maintenance is high (score 1), and the duration of the effect of maintenance is long (score 1).

[0120] Step S1436: Add the influence intensity scores corresponding to the type of maintenance operation, the frequency of maintenance operation, and the duration of the effect of maintenance operation to obtain the total influence intensity score corresponding to the maintenance feature, and the size of the total influence intensity score corresponds to the high and low of the influence intensity.

[0121] After addition, the total influence intensity score is 3+1+1=5, which reflects the influence intensity corresponding to the maintenance feature.

[0122] Step S1437: Convert the total influence intensity score to an influence intensity level, different total influence intensity score intervals correspond to different influence intensity levels, thereby determining the influence intensity corresponding to the current processing rule.

[0123] The correspondence between the preset total influence intensity score interval and the influence intensity level is that the score 1-3 corresponds to the low influence intensity level, the score 4-6 corresponds to the medium influence intensity level, and the score 7-9 corresponds to the high influence intensity level. For the total influence intensity score 8 of the previous environmental feature, the corresponding influence intensity level is high; for the total influence intensity score 5 of the maintenance feature, the corresponding influence intensity level is medium. By the above-mentioned manner, the influence intensity corresponding to the current processing rule is determined.

[0124] Step S144: According to the state change trend in the conclusion, determine the consumption rate of the optical cable life under the state change trend, the consumption rate represents the reduction amount of the optical cable life per unit time.

[0125] The state change trend in the conclusion is different, and the corresponding optical cable life consumption rate is also different. For example, when the state change trend is stable, the consumption rate is v1 (unit: year / year, i.e. the life consumed per year is v1 years); when it is slowly decreasing, the consumption rate is v2, and v2>v1; when it is rapidly decreasing, the consumption rate is v3, and v3>v2. For the current processing rule with the conclusion "the transmission performance data change trend is slowly decreasing", the corresponding life consumption rate is v2.

[0126] Step S145: multiplying the impact intensity by the consumption rate to obtain a life impact factor corresponding to the current processing rule, the life impact factor being used to quantify the influence degree of the associated relationship indicated by the current processing rule on the optical cable life.

[0127] The impact intensity level can be converted into a corresponding numerical value, such as a low level corresponding to a numerical value k1, a medium level corresponding to a numerical value k2, and a high level corresponding to a numerical value k3, and k1 < k2 < k3. For example, when the impact intensity level is high, k3 = 1.5; the consumption rate v2 = 0.2 year / year. Then the life impact factor F = k3 x v2 = 1.5 x 0.2 = 0.3 (unit: year / year), which quantifies the influence degree of the associated rule on the optical cable life.

[0128] Step S146: performing the above processing on each associated rule in the set of road section associated rules to obtain a life impact factor corresponding to each associated rule.

[0129] According to the processing mode of the above steps S142 to S145, each associated rule in the set of road section associated rules is processed. For example, for the rule "when the terrain is unstable and the soil is highly corrosive, the physical state data change trend is rapidly declining", after analyzing its premise and conclusion, it is determined that the impact intensity level is high (corresponding to k3 = 1.5), the consumption rate is v3 = 0.5 year / year, and the life impact factor F = 1.5 x 0.5 = 0.7 year / year is obtained. Through the processing of all rules, the life impact factor corresponding to each rule is obtained.

[0130] Step S147: grouping the life impact factors according to the road sections, integrating all life impact factors belonging to the same road section to form a life impact factor set of the road section, and each life impact factor set of the road section contains the life impact factors corresponding to all associated rules in the road section.

[0131] The life impact factors are grouped according to the road sections to which the associated rules belong. For example, the life impact factors corresponding to all associated rules of the road section S-N1 are F_SN1_1, F_SN1_2, …, F_SN1_m, which are integrated to form the life impact factor set {F_SN1_1, F_SN1_2, …, F_SN1_m} of the road section. Similarly, the life impact factor set of the road section N1-N2 is {F_N1N2_1, F_N1N2_2, …, F_N1N2_n}, and so on, to obtain the life impact factor set of each road section.

[0132] Step S150: combining the life impact factor sets of the road sections and the set of road section associated data to generate an optical cable life prediction result containing the residual life intervals of the road sections.

[0133] In this embodiment, after obtaining the life influence factor set of each section and the section association data set, the residual life interval of each section is calculated by comprehensive analysis of these data, thereby forming a complete optical cable life prediction result.

[0134] Step S151: Select a section data group and the life influence factor set corresponding to the section from the section association data set.

[0135] The data group of section S-N1 and the life influence factor set {F_SN1_1, F_SN1_2, …, F_SN1_m} corresponding to the section are selected as the processing object.

[0136] Step S152: Extract the latest historical running state data in the section data group, and determine the state of the current transmission performance data and physical state data.

[0137] The latest historical running state data at time point t_current in section S-N1 is extracted, wherein the transmission performance data includes signal attenuation degree A_current and transmission rate fluctuation V_current, and the physical state data includes optical cable skin damage S_current and internal optical fiber bending state B_current. By comparing with the preset normal state parameters, it is determined that the current transmission performance data is in the state of “slight decline”, and the physical state data is in the state of “basically stable”.

[0138] Step S153: According to the state of the current transmission performance data and physical state data, and in combination with the preset life consumption benchmark, the proportion of the life consumed by the optical cable of the section is determined.

[0139] In the preset life consumption benchmark, different states correspond to different life consumption proportions. For example, the transmission performance data “stable” corresponds to the consumption proportion of 0-20%, and the “slight decline” corresponds to the consumption proportion of 20%-40%; the physical state data “basically stable” corresponds to the consumption proportion of 0-20%, and the “slight damage” corresponds to the consumption proportion of 20%-40%. The average value of the consumption proportions of the current transmission performance and physical state is taken, such as (30%+15%) / 2=22.5%, to determine that the proportion of the life consumed by the section is 22.5%.

[0140] Step S154: Extract all life influence factors from the life influence factor set of the section, and determine the weight of each life influence factor in life prediction according to the size of the life influence factor.

[0141] Extract all factors from the set of life impact factors {F_SN1_1, F_SN1_2, …, F_SN1_m}, for example F_SN1_1 = 0.3, F_SN1_2 = 0.5, F_SN1_3 = 0.2. Determine the weight according to the size of the factor, and the weight calculation method can be the ratio of each factor to the sum of all factors. The total factor sum is 0.3 + 0.5 + 0.2 = 1.0, then the weight w1 = 0.3 / 1.0 = 0.3, w2 = 0.5 / 1.0 = 0.5, w3 = 0.2 / 1.0 = 0.2.

[0142] Step S155: Weighted sum of each life impact factor according to its weight, to get the comprehensive life impact coefficient of the section.

[0143] The comprehensive life impact coefficient C = w1 x F_SN1_1 + w2 x F_SN1_2 + w3 x F_SN1_3 = 0.3 x 0.3 + 0.5 x 0.5 + 0.2 x 0.2 = 0.09 + 0.25 + 0.04 = 0.38 (unit: year / year).

[0144] Step S156: According to the preset optical cable life reference parameter, the current consumed life proportion and the comprehensive life impact coefficient, the residual life basic value of the optical cable of the section is calculated.

[0145] The preset optical cable life reference parameter is L0 = 20 years, the current consumed life proportion is 22.5%, that is, the consumed life is L0 x 22.5% = 20 x 0.225 = 4.5 years, and the residual life basic value is preliminarily calculated as L0 - consumed life = 20 - 4.5 = 15.5 years. Combined with the comprehensive life impact coefficient C, the adjusted residual life basic value L_base = 15.5 / (1+C) = 15.5 / (1+0.38) = 15.5 / 1.38 ≈ 11.23 years (the adjustment method here is only an example, and the actual model can be determined according to the specific model).

[0146] Step S157: Analyze the future trend of the laying environment data and the maintenance record data in the section data set, predict the influence of future environmental characteristics and maintenance characteristics on life based on the set of section association rules, and determine the fluctuation range of the residual life.

[0147] Step S1571: Extract the historical change of laying environment data in the section data set, including the historical change of temperature change range, the historical change of humidity change, the historical stability of terrain type, and the historical corrosion of soil characteristics.

[0148] The laying environment data history change of the section S-N1 is extracted. The temperature change range shows an expanding trend year by year in the past 5 years. For example, the fluctuation range is 5℃ in the first year, 6℃ in the second year, 7℃ in the third year, 8℃ in the fourth year, and 9℃ in the fifth year. The humidity change history shows that the time of high humidity during the rainy season increases year by year. For example, the duration of humidity higher than 80% during the rainy season is 10 days in the first year, 12 days in the second year, 15 days in the third year, 18 days in the fourth year, and 20 days in the fifth year. The fluctuation of humidity during the non-rainy season also gradually increases. In terms of the historical stability of the terrain type, the terrain slope change and the altitude change of the section in the past 5 years are monitored. The average annual change of the terrain slope is 0.5 degrees / year in the first year, 0.6 degrees / year in the second year, 0.7 degrees / year in the third year, 0.8 degrees / year in the fourth year, and 0.9 degrees / year in the fifth year. The average annual change of the altitude is 0.3 meters / year in the first year, 0.4 meters / year in the second year, 0.5 meters / year in the third year, 0.6 meters / year in the fourth year, and 0.7 meters / year in the fifth year. Overall, the terrain stability shows a slight downward trend year by year. The historical corrosion of the soil characteristics is reflected by the content and activity level of the corrosion components in the soil. In the past 5 years, the content of corrosion components in the soil increases by an average of 0.02% per year, and the activity level detection value also increases year by year. For example, the activity level detection value is 0.1 in the first year, 0.12 in the second year, 0.15 in the third year, 0.18 in the fourth year, and 0.2 in the fifth year.

[0149] Step S1572: According to the historical change of the laying environment data, combined with the weather prediction data and the geographical stability evaluation data, the temperature change range, humidity change, terrain type change, and soil property change in the future setting time are predicted to obtain the future environmental characteristics.

[0150] In this embodiment, the future time is set to 5 years. Combined with meteorological prediction data, it is predicted that the temperature variation range of the road section S-N1 will continue to expand year by year in the next 5 years. Based on the change rate of historical data, the temperature fluctuation range is predicted to be 10℃ in the first year, 11℃ in the second year, 12℃ in the third year, 13℃ in the fourth year, and 14℃ in the fifth year. For the humidity change, combined with the prediction of the future rainy season by the meteorological department, it is predicted that the duration of humidity above 80% during the rainy season will continue to increase in the next 5 years, 22 days in the first year, 25 days in the second year, 28 days in the third year, 31 days in the fourth year, and 34 days in the fifth year, and the humidity fluctuation during the non-rainy season will also further increase. In terms of terrain type change, according to the geographical stability evaluation data, the annual average change amount of the terrain slope of this road section is predicted to be 1.0 degree / year in the first year, 1.1 degree / year in the second year, 1.2 degree / year in the third year, 1.3 degree / year in the fourth year, and 1.4 degree / year in the fifth year; the annual average change amount of the elevation is predicted to be 0.8 meters / year in the first year, 0.9 meters / year in the second year, 1.0 meters / year in the third year, 1.1 meters / year in the fourth year, and 1.2 meters / year in the fifth year, and the terrain stability continues to decrease slightly. In terms of soil property change, according to the prediction model of soil composition change, the content of corrosive components in the soil is predicted to increase by an average of 0.025% per year, and the activity level detection value is 0.23 in the first year, 0.26 in the second year, 0.29 in the third year, 0.32 in the fourth year, and 0.35 in the fifth year, and the corrosion gradually increases. Based on the above prediction results, the environmental characteristics of the road section S-N1 in the next 5 years are obtained.

[0151] Step S1573: Extract the history of the maintenance record data in the road section data set, including the historical distribution of the maintenance operation type, the historical change of the frequency of the maintenance operation, and the historical change of the effect duration of the maintenance operation.

[0152] The history of the maintenance record data of the road section S-N1 is extracted. The historical distribution of the maintenance operation type shows that among the last 10 maintenances, there were 6 times of optical cable repair and 4 times of optical cable reinforcement. The historical change of the frequency of the maintenance operation is that the average maintenance frequency is 1 time per year in the first 5 years and 2 times per year in the last 5 years, showing a trend of gradually increasing frequency. In terms of the historical change of the effect duration of the maintenance operation, the effect duration after optical cable repair is 180 days in the first year, 170 days in the second year, 160 days in the third year, 150 days in the fourth year, and 140 days in the fifth year; the effect duration after optical cable reinforcement is 360 days in the first year, 350 days in the second year, 340 days in the third year, 330 days in the fourth year, and 320 days in the fifth year, and the overall trend is gradually shortened.

[0153] Step S1574: According to the history of the maintenance record data, combined with the maintenance plan and resource allocation situation, the type of maintenance operation, the frequency of maintenance operation, and the duration of maintenance operation effect in the future are predicted, and the future maintenance characteristics are obtained.

[0154] In this embodiment, combined with the maintenance plan and resource allocation situation of the road section, the type of maintenance operation in the next 5 years is predicted, and it is predicted that the proportion of optical cable repair and optical cable reinforcement will remain similar to the historical distribution, that is, about 6 times of optical cable repair and 4 times of optical cable reinforcement in the next 10 maintenance operations. The frequency of maintenance operation is predicted to be 2 times / year in the first year, 2 times / year in the second year, 3 times / year in the third year, 3 times / year in the fourth year, and 3 times / year in the fifth year, continuing to maintain the increasing trend. In terms of the duration of maintenance operation effect, the duration of optical cable repair effect is predicted to be 130 days in the first year, 120 days in the second year, 110 days in the third year, 100 days in the fourth year, and 90 days in the fifth year; the duration of optical cable reinforcement effect is predicted to be 310 days in the first year, 300 days in the second year, 290 days in the third year, 280 days in the fourth year, and 270 days in the fifth year, and will still gradually shorten. Thus, the maintenance characteristics of road section S-N1 in the next 5 years are obtained.

[0155] Step S1575: Substitute the future environmental characteristics and future maintenance characteristics into the road section association rule set to find the corresponding association rules, and determine the change trend of the historical running state data corresponding to the future environmental characteristics and future maintenance characteristics.

[0156] Substitute the future environmental characteristics and maintenance characteristics of road section S-N1 into the road section association rule set to find the corresponding association rules. For example, in the future environmental characteristics, the temperature variation range is expanded, the high humidity time is increased, the terrain stability is decreased, and the soil corrosion is enhanced, combined with the maintenance characteristics such as increased maintenance frequency and shortened effect duration, the corresponding rules are found in the road section association rule set, such as "when the temperature variation range is 10-14℃, the humidity is higher than 80% for 22-34 days, the terrain slope annual change amount is 1.0-1.4 degrees / year, the soil activity degree is 0.23-0.35, and the maintenance frequency is 2-3 times / year, the repair effect duration is 90-130 days, and the reinforcement effect duration is 270-310 days, the change trend of transmission performance data is rapid decline, and the change trend of physical state data is rapid decline", so that the change trend of the historical running state data of the road section in the future is determined to be rapid decline.

[0157] Step S1576: According to the determined change trend of the historical running state data, combined with the consumption rate of the life under the change trend, the consumption of the life in the future is predicted.

[0158] It is known that in the historical operating state data is a rapidly declining trend, the corresponding life consumption rate is 0.15 times the optical cable life benchmark parameter consumed per year. The preset optical cable life benchmark parameter of the section S-N1 is L0, then the life consumption in the next 5 years is expected to be: the first year consumes 0.15×L0, the second year consumes 0.15×L0, the third year consumes 0.16×L0 (since the trend may further intensify), the fourth year consumes 0.16×L0, and the fifth year consumes 0.17×L0.

[0159] Step S1577: Determine the fluctuation range of the remaining life based on the proportional relationship between the future life consumption and the remaining life base value.

[0160] The total life consumption in the next 5 years is calculated as 0.15×L0+0.15×L0+0.16×L0+0.16×L0+0.17×L0=0.79×L0. The remaining life base value of the section S-N1 is L_base, and the ratio of future life consumption to remaining life base value is 0.79×L0 / L_base. According to this proportional relationship, the fluctuation range of the remaining life is determined, for example, when the ratio is 0.79×L0 / L_base, the maximum reduction of the remaining life is 0.2×L_base, and the maximum increase is 0.1×L_base (considering possible maintenance optimization and other factors).

[0161] Step S1578: Determine the fluctuation range of the remaining life based on the remaining life base value, combined with the maximum reduction and the maximum increase, the fluctuation range of the remaining life is the interval between the remaining life base value minus the maximum reduction and the remaining life base value plus the maximum increase.

[0162] Taking the remaining life base value L_base of the section S-N1 as the reference, combined with the determined maximum reduction 0.2×L_base and the maximum increase 0.1×L_base, the fluctuation range of the remaining life of the section is determined to be (L_base-0.2×L_base) to (L_base+0.1×L_base), i.e. the interval between 0.8×L_base and 1.1×L_base.

[0163] Step S158: Generate the remaining life interval of the section based on the remaining life base value and the fluctuation range.

[0164] The remaining life base value of the section S-N1 is L_base, and the fluctuation range is 0.8×L_base to 1.1×L_base, so the remaining life interval of the section is [0.8×L_base, 1.1×L_base].

[0165] Step S159: performing the above-mentioned processing on each road section in the road section associated data set to obtain the residual life interval of each road section, and arranging the residual life interval of each road section in the order of road sections to form the optical cable life prediction result containing the residual life interval of each road section.

[0166] The other road sections in the road section associated data set, such as N1-N2, N2-N3, N3-E, etc., are operated according to the processing mode of the above-mentioned steps S151 to S158. For example, for the road section N1-N2, the residual life interval thereof obtained through processing is [0.9xL0', 1.2xL0']; the residual life interval of the road section N2-N3 is [0.7xL0'', 1.0xL0'']; the residual life interval of the road section N3-E is [1.0xL0''', 1.3xL0'''], etc. Then, the residual life intervals of these road sections are arranged in the order of S-N1, N1-N2, N2-N3, N3-E to form the optical cable life prediction result containing the residual life interval of each road section, which can accurately show the residual life of each road section in the entire optical cable line.

[0167] Figure 2 A schematic diagram of exemplary hardware and software components of the optical cable life prediction system 100 based on big data that can implement the idea of the present application is shown. For example, the processor 120 can be used in the optical cable life prediction system 100 based on big data and used to perform the functions in the present application.

[0168] For example, the optical cable life prediction system 100 based on big data can include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and different forms of storage media 140, such as a disk, a ROM, or a RAM, or any combination thereof. Exemplarily, the optical cable life prediction system 100 based on big data can also include program instructions stored in a ROM, a RAM, or other types of non-transitory storage media, or any combination thereof. The method of the present application can be implemented according to these program instructions. The optical cable life prediction system 100 based on big data also includes an I / O interface 150 between the computer and other input / output devices.

[0169] In addition, the present application also provides a readable storage medium, in which computer executable instructions are pre-set, and when the processor executes the computer executable instructions, the above-mentioned optical cable life prediction method based on big data is realized.

[0170] It should be noted that the foregoing description of embodiments of the application has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the application to the precise form described, and many modifications, variations, and alternatives are possible.

Claims

1. A method for predicting optical cable life based on big data, characterized in that: The method comprises: Aggregating multi-source optical cable operation-related data, wherein the multi-source optical cable operation-related data includes historical operation status data, laying environment data, and maintenance record data of the optical cables; performing data partitioning and association processing on the multi-source optical cable operation-related data, dividing the multi-source optical cable operation-related data into a plurality of section data groups according to the laying sections of the optical cables, to obtain section-related data sets, each section data group containing historical operation status data, laying environment data, and maintenance record data of the corresponding section; performing association rule mining based on the road section association data set to extract association features between historical operating status data, installation environment data, and maintenance record data in different road sections, and generate a road section association rule set, wherein each road section association rule in the road section association rule set is used to indicate an influence relationship between different data items; Generate a life impact factor set for each section according to the section association rule set and the preset optical cable life benchmark parameter, wherein each factor in the life impact factor set corresponds to the strength of an association relationship that affects the optical cable life; Combining the life influencing factor set of each section and the section-related data set, generating an optical cable life prediction result including the remaining life interval of each section; The association rule mining process is performed based on the road section association data set to extract the association features between the historical operation status data and the laying environment data and maintenance record data in different road sections, and generate a road section association rule set, including: Select a road section data group from the road section associated data set as a current processing object, and extract historical operation status data, laying environment data and maintenance record data from the road section data group; Extracting status change trends of transmission performance data and physical status data in historical operation status data to determine the change trends of transmission performance data and physical status data; Extract environmental features from the temperature variation range, humidity variation, terrain type, and soil characteristics in the laying environment data to determine the characteristic manifestations of different environmental factors, including temperature fluctuation amplitude, humidity duration range, terrain stability, and soil corrosivity; Extract maintenance features from the maintenance operation type and maintenance time information in the maintenance record data to determine the frequency of the maintenance operation and the duration of the maintenance operation effect. The duration of the maintenance operation effect is determined by the time that the transmission performance data and physical status data remain stable after maintenance. Analyze the mutual influence relationship between the state change trend and the environmental characteristics, determine the law that when the environmental characteristics show a certain situation, the state change trend presents a corresponding state, and record the law as the first association rule; Analyze the mutual influence between the state change trend and the maintenance characteristics, determine the law that the state change trend presents the corresponding state when the frequency and duration of the maintenance operation are certain, and record this law as the second association rule; Performing the above processing on each road segment data group in the road segment association data set to obtain a first association rule and a second association rule for each road segment; Summarize the first association rules and the second association rules of all road sections, remove duplicate rule contents, and generate a road section association rule set. Each road section association rule in the road section association rule set contains a premise and a conclusion. The components are environmental characteristics or maintenance characteristics, and the conclusions are corresponding state change trends; The generating of a life impact factor set for each section according to the section association rule set and preset optical cable life benchmark parameters includes: Obtaining preset optical cable life reference parameters, where the optical cable life reference parameters are used to represent the expected service life of the optical cable under ideal conditions and without maintenance; Selecting an association rule from the road section association rule set as a current processing rule, and analyzing the premise and conclusion of the current processing rule; Determine the impact intensity corresponding to the current processing rule based on the environmental characteristic performance or maintenance characteristic in the prerequisite and a preset impact intensity evaluation standard, wherein the impact intensity evaluation standard is used to measure the degree of influence of different environmental characteristic performances and maintenance characteristics on the life of the optical cable; Determining, based on the state change trend in the conclusion, a consumption rate of the optical cable life under the state change trend, wherein the consumption rate represents a reduction in the optical cable life per unit time; Multiplying the impact intensity by the consumption rate to obtain a lifespan impact factor corresponding to the current processing rule, wherein the lifespan impact factor is used to quantify the degree of impact of the association relationship indicated by the current processing rule on the lifespan of the optical cable; The above process is performed on each association rule in the road section association rule set to obtain the life impact factor corresponding to each association rule; The lifespan influencing factors are grouped according to the road sections, and all the lifespan influencing factors belonging to the same road section are integrated to form the lifespan influencing factor set of the road section. The lifespan influencing factor set of each road section contains the lifespan influencing factors corresponding to all association rules in the road section. The determining of the impact intensity corresponding to the current processing rule based on the environmental characteristic performance or maintenance characteristic in the precondition and a preset impact intensity evaluation standard includes: When the premise of the current processing rule is environmental characteristics, analyzing the temperature variation range, humidity variation, terrain type, and soil characteristics in the environmental characteristics; According to the preset impact intensity assessment criteria, the greater the temperature change range, the higher the corresponding impact intensity score; in the case of humidity change, the higher the humidity and the longer the duration, the higher the impact intensity score; the more unstable the terrain type, the higher the impact intensity score; the more corrosive the soil characteristics, the higher the impact intensity score; Add the impact intensity scores corresponding to the temperature range, humidity, terrain type, and soil characteristics to obtain the total impact intensity score corresponding to the environmental characteristics. The size of the total impact intensity score corresponds to the level of impact intensity. When the prerequisite of the current processing rule is a maintenance feature, parsing the maintenance operation type, maintenance operation frequency, and maintenance operation effect duration in the maintenance feature; According to the preset impact intensity assessment standard, the impact intensity score corresponding to the repair type of maintenance operation is lower than that of the reinforcement type. The higher the frequency of maintenance operations, the lower the impact intensity score. The longer the duration of the maintenance operation, the lower the impact intensity score. Add the impact intensity scores corresponding to the maintenance operation type, maintenance operation frequency, and maintenance operation effect duration to obtain the total impact intensity score corresponding to the maintenance feature. The size of the total impact intensity score corresponds to the level of impact intensity. Convert the total impact intensity score into an impact intensity level, where different total impact intensity score intervals correspond to different impact intensity levels, thereby determining the impact intensity corresponding to the current processing rule; The generating of the optical cable life prediction result including the remaining life interval of each section by combining the life influencing factor set of each section and the section-related data set includes: Selecting a road section data group and a life impact factor set corresponding to the road section from the road section associated data set; Extract the latest historical operating status data from the road section data group to determine the current status of the transmission performance data and physical status data; Based on the current transmission performance data and physical status data, combined with the preset life consumption benchmark, determine the current proportion of the life of the optical cable in the section; Extract all life-influencing factors from the life-influencing factor set of the road section, and determine the weight of each life-influencing factor in the life-influencing prediction according to its size; The lifespan impact factors are weighted and summed according to their weights to obtain the comprehensive lifespan impact coefficient of the road section; Calculate the remaining life of the optical cable in this section based on the preset optical cable life benchmark parameters, the current consumed life ratio and the comprehensive life impact coefficient; Analyze the future change trends of the laying environment data and maintenance record data in the road section data group, predict the impact of future environmental characteristics and maintenance characteristics on the service life based on the road section association rule set, and determine the fluctuation range of the remaining service life; Combining the remaining life base value and the fluctuation range, the remaining life interval of the road section is generated; The above processing is performed on each section in the section-related data set to obtain the remaining life interval of each section, and the remaining life interval of each section is sorted according to the sequence of the sections to form an optical cable life prediction result containing the remaining life interval of each section.

2. The optical cable life prediction method based on big data according to claim 1, characterized in that: The aggregated multi-source optical cable operation-related data includes: Acquire real-time operating status data transmitted by the optical cable monitoring system, wherein the real-time operating status data includes transmission performance data and physical status data of the current period; Retrieving historical operating status data stored in a database, the historical operating status data including transmission performance data and physical status data for multiple past time periods, the transmission performance data including signal attenuation and transmission rate fluctuations, and the physical status data including cable sheath damage and internal fiber bending conditions; Collect geographical environment data along the route provided by the geographic information system, including the terrain type and soil characteristics of the laying area, and obtain climate data provided by the meteorological monitoring station, including the temperature range and humidity changes; Extracting maintenance record data from a maintenance management system, wherein the maintenance record data includes a maintenance operation type and corresponding maintenance time information for each maintenance operation, wherein the maintenance operation type includes optical cable repair and optical cable reinforcement; The real-time operation status data, the historical operation status data, the geographical environment data along the line, the climate data and the maintenance record data are aggregated and integrated to form multi-source optical cable operation related data, and each data in the multi-source optical cable operation related data is accompanied by a corresponding optical cable identification and time stamp.

3. The optical cable life prediction method based on big data according to claim 1, characterized in that: The step of performing data partition association processing on the multi-source optical cable operation associated data, dividing the multi-source optical cable operation associated data into a plurality of section data groups according to the optical cable laying sections, and obtaining a section associated data set, includes: Parsing the optical cable identifiers in the multi-source optical cable operation association data to determine the overall laying path of the optical cable and the segment division nodes on the path, wherein the segment division nodes are determined based on the geographical boundaries of the laying area and the maintenance responsibility zones; Allocating each data item in the multi-source optical cable operation-related data to a corresponding road section partition according to the road section division node to generate preliminary road section data groups, each road section data group containing historical operation status data, laying environment data, and maintenance record data of the road section; In each road section data group, an association index is established between historical operating status data and laying environment data. The association index is used to indicate the corresponding relationship between transmission performance data, physical status data and temperature variation range, humidity variation, terrain type, and soil characteristics within the same time interval; Establishing an association index between historical operating status data and maintenance record data, wherein the association index is used to indicate the correspondence between the maintenance operation type, maintenance time information, and transmission performance data and physical status data before and after maintenance; A plurality of road segment data groups with associated indexes are integrated to form a road segment associated data set, and the road segment associated data set can reflect the time correlation and impact correlation between different types of data in each road segment.

4. The optical cable life prediction method based on big data according to claim 3, characterized in that: The extracting of status change trends of the transmission performance data and the physical status data in the historical operation status data to determine the change trends of the transmission performance data and the physical status data includes: Extracting transmission performance data from historical operating status data, wherein the transmission performance data includes signal attenuation and transmission rate fluctuation at multiple time points; Arrange the signal attenuation levels in chronological order, calculate the changes in signal attenuation between adjacent time points, and determine the trend of signal attenuation changes in the transmission performance data based on the magnitude of the changes. A change within a preset small range indicates a stable state, a change within a preset medium range indicates a slowly decreasing state, and a change exceeding the preset medium range indicates a rapidly decreasing state. Arrange the transmission rate fluctuations in chronological order, calculate the amplitude of the transmission rate fluctuations between adjacent time points, and determine the transmission rate fluctuation trend in the transmission performance data based on the amplitude of the amplitude. If the amplitude is within a small preset range, it is considered stable; if the amplitude is within a medium preset range, it is considered slowly decreasing; and if the amplitude exceeds the medium preset range, it is considered rapidly decreasing. The overall trend of transmission performance data is determined by combining the trend of signal attenuation and the trend of transmission rate fluctuation. When both are in a stable state, the overall trend is stable; when at least one is in a slowly decreasing state and there is no rapidly decreasing state, the overall trend is slowly decreasing; when at least one is in a rapidly decreasing state, the overall trend is rapidly decreasing. Extracting physical status data from historical operating status data, wherein the physical status data includes damage to the outer sheath of the optical cable and bending status of the internal optical fiber at multiple time points; Arrange the damage of the optical cable sheath in chronological order, calculate the change in the degree of damage of the optical cable sheath between adjacent time points, and determine the change trend of the optical cable sheath damage in the physical state data according to the degree of change. If the degree of change is within a preset small range, it is a stable state; if the degree of change is within a preset medium range, it is a slowly decreasing state; if the degree of change exceeds the preset medium range, it is a rapidly decreasing state; Arrange the internal fiber bending states in chronological order, calculate the change in the degree of internal fiber bending between adjacent time points, and determine the change trend of the internal fiber bending in the physical state data based on the degree of change. If the degree of change is within a preset small range, it is a stable state; if the degree of change is within a preset medium range, it is a slowly decreasing state; and if the degree of change exceeds the preset medium range, it is a rapidly decreasing state. The overall change trend of the physical status data is determined by combining the change trend of the cable sheath damage and the change trend of the internal optical fiber bending. When both are in a stable state, the overall change trend is stable; when at least one is in a slowly decreasing state and there is no rapidly decreasing state, the overall change trend is slowly decreasing; when at least one is in a rapidly decreasing state, the overall change trend is rapidly decreasing. The overall change trend of the transmission performance data and the overall change trend of the physical status data are combined to obtain the change trend of the historical operation status data.

5. The optical cable life prediction method based on big data according to claim 4, characterized in that: The analyzing of future change trends of the installation environment data and maintenance record data in the road section data group, predicting the impact of future environmental characteristics and maintenance characteristics on the service life based on the road section association rule set, and determining the fluctuation range of the remaining service life includes: Extracting historical changes in the laying environment data in the road section data group, wherein the historical changes include historical changes in temperature range, historical changes in humidity, historical stability of terrain type, and historical corrosiveness of soil characteristics; Based on the historical changes in the laying environment data, combined with meteorological forecast data and geographical stability assessment data, the temperature change range, humidity change, terrain type change, and soil property change within the future set time are predicted to obtain future environmental characteristics; Extracting historical status of maintenance record data in the road segment data group, wherein the historical status includes historical distribution of maintenance operation types, historical changes in frequency of maintenance operations, and historical changes in duration of effects of maintenance operations; Based on the historical maintenance record data, combined with the maintenance plan and resource allocation, the maintenance operation type, frequency, and duration of maintenance operation effects within a set time period in the future are predicted to obtain future maintenance characteristics. Substituting the future environmental characteristics and the future maintenance characteristics into the road section association rule set, searching for corresponding association rules, and determining the change trend of the historical operating status data corresponding to the future environmental characteristics and the future maintenance characteristics; Based on the determined trend of historical operating status data and the life consumption rate under the trend, the life consumption within a set time in the future is predicted; Based on the proportional relationship between future life consumption and the basic value of remaining life, determine the maximum possible reduction and maximum increase in remaining life; Taking the basic value of remaining life as the benchmark, the fluctuation range of remaining life is determined in combination with the maximum decrease and the maximum increase. The fluctuation range of remaining life is the interval between the basic value of remaining life minus the maximum decrease and the basic value of remaining life plus the maximum increase.

6. A big data-based optical cable life prediction system, characterized in that: The invention comprises a processor and a memory, wherein the memory is connected to the processor, the memory is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the memory to implement the optical cable life prediction method based on big data as described in any one of claims 1 to 5.

Images