A bus duct line management system and method based on artificial intelligence
Through the bus duct line management system based on artificial intelligence, the maintenance behavior characteristics of the bus duct line network are analyzed and the smooth curve function is generated, which solves the blindness problem in the maintenance of the bus duct line network in the existing technology, and achieves targeted and efficient maintenance. Strategy to adapt to changes in the entire life cycle of the bus duct.
Patent Information
- Application Number
- CN202410113067.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-26
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-01-26
AI Technical Summary
The existing technology is blind in the maintenance of bus duct line networks and lacks targeted strategies, which leads to the increase of invalid work content in maintenance work and is difficult to deal with changes in different operating conditions throughout the entire life cycle of bus ducts.
Using a bus trough line management system based on artificial intelligence, through data source processing, data relationship analysis, feature sequence generation and iterative analysis modules, we build maintenance objects, generate smooth curve functions, analyze synchronous maintenance solutions between maintenance objects, and realize targeting Sexual maintenance strategies.
We have implemented targeted maintenance work content strategies based on the changes in the entire life cycle and operating conditions of the bus duct line network, which improves the pertinence and efficiency of maintenance.
Smart Images

Figure CN118036948B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of bus ducts, and in particular to a bus duct line management system and method based on artificial intelligence. Background Art
[0002] As an important component of the power system, the operating status of the bus duct directly affects the stability and reliability of power transmission; however, since the bus duct is in a high-current, high-voltage working environment for a long time, its lines and components are easily affected by factors such as electrochemical corrosion, dust accumulation, and changes in ambient temperature and humidity, which can lead to failures; therefore, bus duct maintenance is essential to ensure its normal operation.
[0003] In the prior art, measures such as regular inspection of the bus duct appearance, fasteners and insulating materials, as well as cleaning of dust and dirt are generally adopted to timely discover potential problems and take corresponding measures to repair them to avoid the occurrence of faults; and, for faults that have already occurred, targeted repairs are required, such as replacing damaged parts, repairing line faults or repainting the bus duct; at the same time, infrared thermometers, partial discharge detectors and other equipment are used to monitor the operating status of the bus duct in real time and diagnose faults;
[0004] Then, for the complex bus duct line network, the types and methods of faults and maintenance in each bus duct line segment have different maintenance methods as the bus duct changes throughout its life cycle or due to different operating conditions. If existing technologies are used for detection and maintenance, there is a lot of blindness, which makes the maintenance work lack targeted strategies and adds a lot of invalid work content. Summary of the invention
[0005] The purpose of the present invention is to provide a bus duct line management system and method based on artificial intelligence to solve the problems raised in the above background technology.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:
[0007] A bus duct line management system based on artificial intelligence, the system includes: a data source processing module, a data relationship analysis module, a feature sequence generation module and an iteration analysis module;
[0008] The data source processing module is used to sort out each branch node in the bus duct line network, form a bus duct line segment between any two adjacent branch nodes, and create a maintenance object based on the bus duct line segment; allocate a service data cluster, record and store a maintenance log of the maintenance object;
[0009] The data relationship analysis module is used to extract all branch nodes adjacent to the central node in the bus duct line network with the branch node as the central node to form a node-centric feature set; take the synchronous maintenance of the bus duct line network as a unified scale, extract the service label set of the maintenance object recorded during the different maintenance of the bus duct line network in the service data cluster, take the node-centric feature set as a discretization guide, count all the behavior features formed during the different maintenance of the bus duct line network, and generate a maintenance behavior feature set;
[0010] The feature sequence generation module analyzes the continuous behavior characteristics of the bus duct line network maintenance according to the maintenance behavior feature set, calculates the first service feature value of the maintenance object; sorts the first service feature values according to the continuous behavior sequence of the bus duct line network maintenance, and generates a first service feature value sequence;
[0011] The iterative analysis module is used to construct a two-dimensional coordinate system, map the first service characteristic value sequence generated corresponding to each maintenance object into the two-dimensional coordinate system in the form of point coordinates, and generate a smooth curve function; analyze the second service characteristic value between any two smooth curve functions, and generate a synchronous secondary maintenance plan between maintenance objects through function translation and iterative analysis.
[0012] Furthermore, the data source processing module also includes a data object construction unit and a data cluster allocation unit;
[0013] The data object construction unit is used to sort out each branch node in the bus duct line network through the connection tree model; form a bus duct line segment between any two adjacent branch nodes, and create a maintenance object based on the bus duct line segment, and record any maintenance object as I ij , where I ij Indicates the maintenance object corresponding to the bus duct line segment formed between the adjacent i-th node and j-th node in the bus duct line network;
[0014] The data cluster allocation unit is used to build a bus duct line maintenance cloud platform, record and store the maintenance log of the maintenance object, and the maintenance log records different types of maintenance items; allocate service data clusters, and one maintenance object is correspondingly allocated a service data cluster, and any service data cluster corresponding to a maintenance object is recorded as SDC (I ij )={MR v |v∈[1,u]}, where SDC(I ij ) indicates maintenance object I ij Corresponding to the assigned service data cluster, MR v Indicates that during the vth maintenance of the bus duct line network, maintenance object I ij The service label set of MRv ={MP x |x∈[1,w]}, where MP x It indicates the xth type of maintenance item, u indicates the number of maintenance times of the bus duct line network at present, and w indicates the type code number of the maintenance item.
[0015] Furthermore, the data relationship analysis module also includes a node centralization feature unit and a behavior integration unit;
[0016] The node-centric feature unit is used to extract all branch nodes adjacent to the i-th branch node in the bus duct line network with the i-th branch node as the central node to form a node-centric feature set, which is recorded as NC i ={j|j∈[1,J]}, where J represents the maximum code number of the branch nodes adjacent to the i-th branch node when the i-th branch node is the central node;
[0017] The behavior integration unit is used to use the synchronous secondary maintenance of the bus duct line network as a unified standard, in the service data cluster SDC (I ij ), extract the maintenance object I recorded during the vth maintenance of the bus duct line network ij Service label collection MR v , and constitute the behavioral characteristics, denoted as I ij :MR v , and I ij :MR v =MR v ; Node-centered feature set NC i For discretization guidance, all behavioral features formed during the vth maintenance of the bus duct line network are counted, and a maintenance behavior feature set is generated, denoted as MBF v (NC i )={I ij :MR v |j∈NC i}.
[0018] Furthermore, the feature sequence generation module also includes a maintenance flow feature analysis unit and a feature sequence generation unit;
[0019] The maintenance flow feature analysis unit calculates, based on the maintenance behavior feature set, the maintenance object I when the maintenance flow from the vth maintenance of the bus duct line network to the v+1th maintenance of the bus duct line network. ij The first service characteristic value is calculated as follows:
[0020]
[0021] Among them, FE[I ij(v)] indicates that when the maintenance flow from the vth maintenance of the bus duct line network to the v+1th maintenance of the bus duct line network, the maintenance object I ij The first service characteristic value, I ij :MR v+1 Indicates that when the bus duct line network is maintained for the v+1th time, the maintenance object I ij The corresponding behavior characteristics, NUM(I ij :MR v+1 -I ij :MR v ) indicates behavioral characteristics I ij :MR v+1 and behavioral characteristics I ij :MR v The number of maintenance item types included in the difference set, NUM(I ij :MR v+1 ) indicates behavioral characteristics I ij :MR v+1 The number of maintenance item types included in the
[0022] The feature sequence generation unit is used to set v=v+1, return to step S301, and iterate and calculate the maintenance object I in turn. ij When v = u, the iteration stops; obtain the maintenance object I in the process of bus duct line network maintenance flow ij All the first service characteristic values generated are arranged in sequence according to the flow order from first to last to generate a first service characteristic value sequence, which is recorded as AL(I ij ), where AL(I ij ) indicates maintenance object I ij The corresponding first service characteristic value sequence is generated.
[0023] Furthermore, the iterative analysis module also includes a function generation unit and a solution decision unit;
[0024] The function generation unit is used to construct a two-dimensional coordinate system, and the first service feature value sequence generated corresponding to each maintenance object is converted into a point coordinate (v, FE[I ij (v)]) in the form of, mapped to a two-dimensional coordinate system, and the first service characteristic value sequence is converted into a smooth curve function by Matlab software; the first service characteristic value sequence AL(I ij ) corresponds to the smooth curve function transformed, denoted as f ij (t), t represents the independent variable of the smooth curve function;
[0025] The solution decision unit is used to obtain any two smooth curve functions, analyze the second service characteristic value between any two smooth curve functions, and calculate the second service characteristic value between any two smooth curve functions. The specific calculation formula is as follows:
[0026]
[0027] Among them, R[f ij (t)→f ab (t)] represents the smooth curve function f ij (t) and the smooth curve function f ab (t) the second service characteristic value between ab (t) represents the first service characteristic value sequence AL(I ab ) corresponds to the smooth curve function transformed, I ab Indicates the maintenance object corresponding to the bus duct line segment formed between the adjacent a-th node and b-th node in the bus duct line network, f ab (t+T) represents the smooth curve function F ab (t) The smooth curve function obtained after the independent variable is translated by step length T. The positive and negative values of T correspond to the smooth curve function f ab (t) translate to the left or right, and ij≠ab;
[0028] Let T = T + 1, and iterate the calculation of the second service characteristic value between any two smooth curve functions until the iteration stops when the second service characteristic value is the maximum. The second service characteristic value corresponding to the maximum is recorded as R max [f ij (t)→f ab (t)];
[0029] Calculate the smooth curve function f ij (t) the maximum second service characteristic value between each smooth curve function;
[0030] The second service characteristic value threshold is preset, and the maximum second service characteristic value R max [f ij (t)→f ab (t)] is greater than or equal to the second service characteristic value threshold, then the maintenance object I is screened out. ab , and maintain object I ab Maintenance Object I ij Perform synchronization maintenance.
[0031] A bus duct line management method based on artificial intelligence, the method comprising the following steps:
[0032] Step S100: combing each branch node in the bus duct line network, forming a bus duct line segment between any two adjacent branch nodes, and creating a maintenance object based on the bus duct line segment; allocating a service data cluster, recording and storing a maintenance log of the maintenance object;
[0033] Step S200: Taking the branch node as the central node, extract all branch nodes adjacent to the central node in the bus duct line network to form a node-centralized feature set; taking the synchronous maintenance of the bus duct line network as a unified scale, extract the service label set of the maintenance object recorded during the different maintenance of the bus duct line network in the service data cluster, taking the node-centralized feature set as a discretization guide, and counting all the behavior features formed during the different maintenance of the bus duct line network to generate a maintenance behavior feature set;
[0034] Step S300: Analyze the continuous behavior characteristics of the bus duct line network maintenance according to the maintenance behavior characteristic set, and calculate the first service characteristic value of the maintenance object; sort the first service characteristic values according to the continuous behavior sequence of the bus duct line network maintenance, and generate a first service characteristic value sequence;
[0035] Step S400: construct a two-dimensional coordinate system, map the first service characteristic value sequence generated for each maintenance object into the two-dimensional coordinate system in the form of point coordinates, and generate a smooth curve function; analyze the second service characteristic value between any two smooth curve functions, and generate a synchronous secondary maintenance plan between maintenance objects through function translation and iterative analysis.
[0036] Furthermore, the specific implementation process of step S100 includes:
[0037] Step S101: comb out each branch node in the bus duct line network through the connection tree model; form a bus duct line segment between any two adjacent branch nodes, and create a maintenance object based on the bus duct line segment, and record any maintenance object as I ij , where I ij Indicates the maintenance object corresponding to the bus duct line segment formed between the adjacent i-th node and j-th node in the bus duct line network;
[0038] Step S102: Building a bus duct line maintenance cloud platform, recording and storing maintenance logs of maintenance objects, wherein the maintenance logs record different types of maintenance items; allocating service data clusters, and one maintenance object is allocated one service data cluster, and any service data cluster allocated to any maintenance object is recorded as SDC (I ij )={MR v |v∈[1,u]}, where SDC(I ij ) indicates maintenance object Iij Corresponding to the assigned service data cluster, MR v Indicates that during the vth maintenance of the bus duct line network, maintenance object I ij The service label set of MR v ={MP x |x∈[1,w]}, where MP x It indicates the xth type of maintenance item, u indicates the number of maintenance times of the bus duct line network at present, and w indicates the type code number of the maintenance item.
[0039] Furthermore, the specific implementation process of step S200 includes:
[0040] Step S201: Taking the ith branch node as the central node, extract all the branch nodes adjacent to the ith branch node in the bus duct line network to form a node-centralized feature set, denoted as NC i ={j|j∈[1,J]}, where J represents the maximum code number of the branch nodes adjacent to the i-th branch node when the i-th branch node is the central node;
[0041] Step S202: Taking the synchronous maintenance of the bus duct line network as a unified standard, in the service data cluster SDC (I ij ), extract the maintenance object I recorded during the vth maintenance of the bus duct line network ij Service label collection MR v , and constitute the behavioral characteristics, denoted as I ij :MR v , and I ij :MR v =MR v ; Node-centered feature set NC i For discretization guidance, all behavioral features formed during the vth maintenance of the bus duct line network are counted, and a maintenance behavior feature set is generated, denoted as MBF v (NC i )={I ij :MR v |j∈NC i}.
[0042] According to the above method, the bus duct line network is composed of many bus duct lines cross-connected, and its characteristic is a connection tree model composed of node branches, and then a branch node is connected to the next branch node to form each branch segment in the bus duct line network. The present application invention is based on the bus duct line segment to analyze the maintenance plan decision during the maintenance process of the bus duct line network.
[0043] Furthermore, the specific implementation process of step S300 includes:
[0044] Step S301: According to the maintenance behavior feature set, calculate the maintenance object I when the maintenance flow from the vth maintenance of the bus duct line network to the v+1th maintenance of the bus duct line network ij The first service characteristic value is calculated as follows:
[0045]
[0046] Among them, FE[I ij (v)] indicates that when the maintenance flow from the vth maintenance of the bus duct line network to the v+1th maintenance of the bus duct line network, the maintenance object I ij The first service characteristic value, I ij :MR v+1 Indicates that when the bus duct line network is maintained for the v+1th time, the maintenance object I ij The corresponding behavior characteristics, NUM(I ij :MR v+1 -I ij :MR v ) indicates behavioral characteristics I ij :MR v+1 and behavioral characteristics I ij :MR v The number of maintenance item types included in the difference set, NUM(I ij :MR v+1 ) indicates behavioral characteristics I ij :MR v+1 The number of maintenance item types included in the
[0047] Step S302: Let v = v + 1, return to step S301, and iterate and calculate the maintenance object I ij When v = u, the iteration stops; obtain the maintenance object I in the process of bus duct line network maintenance flow ij All the first service characteristic values generated are arranged in sequence according to the flow order from first to last to generate a first service characteristic value sequence, which is recorded as AL(I ij ), where AL(I ij ) indicates maintenance object I ij The corresponding first service characteristic value sequence is generated.
[0048] Furthermore, the specific implementation process of step S400 includes:
[0049] Step S401: construct a two-dimensional coordinate system, and map the first service feature value sequence generated for each maintenance object to point coordinates (v, FE[I ij(v)]) in the form of, mapped to a two-dimensional coordinate system, and the first service characteristic value sequence is converted into a smooth curve function by Matlab software; the first service characteristic value sequence AL(I ij ) corresponds to the smooth curve function transformed, denoted as f ij (t), t represents the independent variable of the smooth curve function;
[0050] Step S402: Obtain any two smooth curve functions, analyze the second service characteristic value between any two smooth curve functions, and calculate the second service characteristic value between any two smooth curve functions. The specific calculation formula is as follows:
[0051]
[0052] Among them, R[f ij (t)→f ab (t)] represents the smooth curve function f ij (t) and the smooth curve function f ab (t) the second service characteristic value between ab (t) represents the first service characteristic value sequence AL(I ab ) corresponds to the smooth curve function transformed, I ab Indicates the maintenance object corresponding to the bus duct line segment formed between the adjacent a-th node and b-th node in the bus duct line network, f ab (t+T) represents the smooth curve function f ab (t) The smooth curve function obtained after the independent variable is translated by step length T. The positive and negative values of T correspond to the smooth curve function f ab (t) translate to the left or right, and ij≠ab;
[0053] Let T = T + 1, and iterate the calculation of the second service characteristic value between any two smooth curve functions until the iteration stops when the second service characteristic value is the maximum. The second service characteristic value corresponding to the maximum is recorded as R max [f ij (t)→f ab (t)];
[0054] Calculate the smooth curve function f ij (t) the maximum second service characteristic value between each smooth curve function;
[0055] The second service characteristic value threshold is preset, and the maximum second service characteristic value R max [f ij (t)→f ab (t)] is greater than or equal to the second service characteristic value threshold, then the maintenance object I is screened out. ab , and maintain object I abMaintenance Object I ij Perform synchronization maintenance.
[0056] According to the above method, the maintenance of the bus duct line network includes a variety of maintenance project types. In the prior art, the appearance, high temperature, voltage and current, loose joints and insulation aging of the bus duct are generally monitored. Then, the present invention analyzes the types of maintenance projects implemented in the historical maintenance of the bus duct to fuzzily summarize the maintenance projects that should be performed on the bus duct in different periods, so as to carry out targeted iteration of the maintenance strategy of the bus duct. By using periodic iterative observation to cover the continuous periodic maintenance method of the bus duct, the maintenance projects of the bus duct line network throughout its life cycle are digitally analyzed, and the maintenance of the bus duct is carried out in a segmented and targeted manner. The first service characteristic value of the maintenance object represents the difference in service projects of the same bus duct line segment during the maintenance flow process. Formula I ij :MR v+1 -I ij :MR v It is a set difference operation formula. The set difference operation is a basic set operation, which is used to find the difference between two sets. For two sets Set 1 and Set 2, the difference set 1-Set 2 of Set 1 is a set composed of all elements that belong to Set 1 but not Set 2. The application in this application feeds back the difference in service items taken by the maintenance object when the v-th maintenance flow is transferred to the v+1-th maintenance of the bus duct line network; and then the larger the first service characteristic value of the maintenance object, the more new difference maintenance items will be added during the next maintenance in a short period of time; the second service characteristic value represents the correlation between the long-term maintenance behaviors of the two maintenance objects. The second service characteristic value is calculated based on the first service characteristic value. The larger the second service characteristic value, the greater the correlation between the long-term maintenance behaviors of the two maintenance objects, so that synchronous maintenance can be performed.
[0057] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: in an artificial intelligence-based bus duct line management system and method provided by the present invention, a bus duct line segment is formed between any two adjacent branch nodes, and a maintenance object is created; taking the branch node as the central node, all branch nodes adjacent to the central node are extracted from the bus duct line network, taking the synchronous secondary maintenance of the bus duct line network as the unified scale, extracting the service label set of the maintenance object recorded during different maintenance of the bus duct line network in the service data cluster, taking the node-centric feature set as the discretization guide, and counting all the behavioral features constructed during different maintenance of the bus duct line network; analyzing the continuous behavioral features of the maintenance of the bus duct line network and converting them into smooth curve functions; analyzing the commonality of long-term service behaviors between any two smooth curve functions, and generating synchronous secondary maintenance plans between maintenance objects through function translation and iterative analysis; and then being able to carry out targeted execution of maintenance work content strategies as the changes in the entire life cycle of the bus duct or the different operating conditions. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0059] Figure 1 It is a structural schematic diagram of a bus duct line management system based on artificial intelligence of the present invention;
[0060] Figure 2 It is a schematic diagram of the steps of a bus duct line management method based on artificial intelligence of the present invention. DETAILED DESCRIPTION
[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0062] See also Figure 1-Figure 2 , the present invention provides a technical solution:
[0063] See also Figure 1 , in the first embodiment: a bus duct line management system based on artificial intelligence is provided, the system comprises: a data source processing module, a data relationship analysis module, a feature sequence generation module and an iteration analysis module;
[0064] The data source processing module is used to sort out each branch node in the bus duct line network, form a bus duct line segment between any two adjacent branch nodes, and create a maintenance object based on the bus duct line segment; allocate a service data cluster, record and store the maintenance log of the maintenance object;
[0065] Wherein, the data source processing module also includes a data object construction unit and a data cluster allocation unit;
[0066] The data object construction unit is used to sort out the various branch nodes in the bus duct line network through the connection tree model; between any two adjacent branch nodes, a bus duct line segment is formed, and based on the bus duct line segment, a maintenance object is created, and any maintenance object is recorded as I ij , where I ij Indicates the maintenance object corresponding to the bus duct line segment formed between the adjacent i-th node and j-th node in the bus duct line network;
[0067] The data cluster allocation unit is used to build a bus duct line maintenance cloud platform, record and store the maintenance log of the maintenance object, and the maintenance log records different types of maintenance items; allocate service data clusters, and one maintenance object is allocated one service data cluster, and any service data cluster allocated to a maintenance object is recorded as SDC (I ij )={MR v |v∈[1,u]}, where SDC(I ij ) indicates maintenance object I ij Corresponding to the assigned service data cluster, MR v Indicates that during the vth maintenance of the bus duct line network, maintenance object I ij The service label set of MR v ={MP x |x∈[1,w]}, where MP x It indicates the xth type of maintenance item, u indicates the number of maintenance times of the bus duct line network at present, and w indicates the type code number of the maintenance item.
[0068] The data relationship analysis module is used to extract all branch nodes adjacent to the central node in the bus duct line network with the branch node as the central node to form a node-centered feature set; take the synchronous maintenance of the bus duct line network as a unified scale, extract the service label set of the maintenance object recorded during the different maintenance of the bus duct line network in the service data cluster, take the node-centered feature set as a discretization guide, count all the behavioral features formed during the different maintenance of the bus duct line network, and generate a maintenance behavior feature set;
[0069] Among them, the data relationship analysis module also includes a node-centric feature unit and a behavior integration unit;
[0070] The node-centered feature unit is used to extract all branch nodes adjacent to the i-th branch node in the busbar line network with the i-th branch node as the center node to form a node-centered feature set, denoted as NC i ={j|j∈[1,J]}, where J represents the maximum code number of the branch nodes adjacent to the i-th branch node when the i-th branch node is the central node;
[0071] The behavior integration unit is used to unify the synchronous maintenance of the bus duct line network in the service data cluster SDC (I ij ), extract the maintenance object I recorded during the vth maintenance of the bus duct line network ij Service label collection MR v , and constitute the behavioral characteristics, denoted as I ij :MR v , and I ij :MR v =MR v ; Node-centered feature set NC i For discretization guidance, all behavioral features formed during the vth maintenance of the bus duct line network are counted, and a maintenance behavior feature set is generated, denoted as MBF v (NC i )={I ij :MR v |j∈NC i}.
[0072] The feature sequence generation module analyzes the continuous behavior characteristics of the bus duct line network maintenance according to the maintenance behavior feature set, calculates the first service feature value of the maintenance object; sorts the first service feature values according to the continuous behavior sequence of the bus duct line network maintenance, and generates a first service feature value sequence;
[0073] Among them, the feature sequence generation module also includes a maintenance flow feature analysis unit and a feature sequence generation unit;
[0074] The maintenance flow characteristic analysis unit calculates the maintenance object I from the vth maintenance flow of the bus duct line network to the v+1th maintenance of the bus duct line network according to the maintenance behavior characteristic set. ij The first service characteristic value is calculated as follows:
[0075]
[0076] Among them, FE[I ij (v)] indicates that when the maintenance flow from the vth maintenance of the bus duct line network to the v+1th maintenance of the bus duct line network, the maintenance object I ijThe first service characteristic value, I ij :MR v+1 Indicates that when the bus duct line network is maintained for the v+1th time, the maintenance object I ij The corresponding behavior characteristics, NUM(I ij :MR v+1 -I ij :MR v ) indicates behavioral characteristics I ij :MR v+1 and behavioral characteristics I ij :MR v The number of maintenance item types included in the difference set, NUM(I ij :MR v+1 ) indicates behavioral characteristics I ij :MR v+1 The number of maintenance item types included in the
[0077] The feature sequence generation unit is used to set v = v + 1, return to step S301, and iterate and calculate the maintenance object I ij When v = u, the iteration stops; obtain the maintenance object I in the process of bus duct line network maintenance flow ij All the first service characteristic values generated are arranged in sequence according to the flow order from first to last to generate a first service characteristic value sequence, which is recorded as AL(I ij ), where AL(I ij ) indicates maintenance object I ij The corresponding first service characteristic value sequence is generated.
[0078] The iterative analysis module is used to construct a two-dimensional coordinate system, map the first service characteristic value sequence generated for each maintenance object to the two-dimensional coordinate system in the form of point coordinates, and generate a smooth curve function; analyze the second service characteristic value between any two smooth curve functions, and generate a synchronous secondary maintenance plan between maintenance objects through function translation and iterative analysis;
[0079] Among them, the iterative analysis module also includes a function generation unit and a solution decision unit;
[0080] The function generation unit is used to construct a two-dimensional coordinate system, and the first service feature value sequence generated corresponding to each maintenance object is represented by the point coordinates (v, FE[I ij (v)]) in the form of, mapped to a two-dimensional coordinate system, and the first service characteristic value sequence is converted into a smooth curve function by Matlab software; the first service characteristic value sequence AL(I ij ) corresponds to the smooth curve function transformed, denoted as f ij (t), t represents the independent variable of the smooth curve function;
[0081] The solution decision unit is used to obtain any two smooth curve functions, analyze the second service characteristic value between any two smooth curve functions, and calculate the second service characteristic value between any two smooth curve functions. The specific calculation formula is as follows:
[0082]
[0083] Among them, R[f ij (t)→f ab (t)] represents the smooth curve function f ij (t) and the smooth curve function f ab (t) the second service characteristic value between ab (t) represents the first service characteristic value sequence AL(I ab ) corresponds to the smooth curve function transformed, I ab Indicates the maintenance object corresponding to the bus duct line segment formed between the adjacent a-th node and b-th node in the bus duct line network, f ab (t+T) represents the smooth curve function f ab (t) The smooth curve function obtained after the independent variable is translated by step length T. The positive and negative values of T correspond to the smooth curve function f ab (t) translate to the left or right, and ij≠ab;
[0084] Let T = T + 1, and iterate the calculation of the second service characteristic value between any two smooth curve functions until the iteration stops when the second service characteristic value is the maximum. The second service characteristic value corresponding to the maximum is recorded as R max [f ij (t)→f ab (t)];
[0085] Calculate the smooth curve function f ij (t) the maximum second service characteristic value between each smooth curve function;
[0086] The second service characteristic value threshold is preset, and the maximum second service characteristic value R max [f ij (t)→f ab (t)] is greater than or equal to the second service characteristic value threshold, then the maintenance object I is screened out. ab , and maintain object I ab Maintenance Object I ij Perform synchronization maintenance.
[0087] See also Figure 2 In the second embodiment, a bus duct line management method based on artificial intelligence is provided, and the method comprises the following steps:
[0088] Step S100: combing each branch node in the bus duct line network, forming a bus duct line segment between any two adjacent branch nodes, and creating a maintenance object based on the bus duct line segment; allocating a service data cluster, recording and storing a maintenance log of the maintenance object;
[0089] Specifically, each branch node in the bus duct line network is sorted out through the connection tree model; a bus duct line segment is formed between any two adjacent branch nodes, and a maintenance object is created based on the bus duct line segment, and any maintenance object is recorded as I ij , where I ij Indicates the maintenance object corresponding to the bus duct line segment formed between the adjacent i-th node and j-th node in the bus duct line network;
[0090] Build a bus duct line maintenance cloud platform to record and store maintenance logs of maintenance objects. The maintenance logs record different types of maintenance items. Allocate service data clusters, and one maintenance object corresponds to one service data cluster. The service data cluster corresponding to any maintenance object is recorded as SDC (I ij )={MR v |v∈[1,u]}, where SDC(I ij ) indicates maintenance object I ij Corresponding to the assigned service data cluster, MRv indicates that during the vth maintenance of the bus duct line network, maintenance object I ij The service label set of MR v ={MP x |x∈[1,w]}, where MP x It indicates the xth type of maintenance item, u indicates the number of maintenance times of the bus duct line network at present, and w indicates the type code number of the maintenance item.
[0091] Step S200: Taking the branch node as the central node, extract all branch nodes adjacent to the central node in the bus duct line network to form a node-centralized feature set; taking the synchronous maintenance of the bus duct line network as a unified scale, extract the service label set of the maintenance object recorded during the different maintenance of the bus duct line network in the service data cluster, taking the node-centralized feature set as a discretization guide, and counting all the behavior features formed during the different maintenance of the bus duct line network to generate a maintenance behavior feature set;
[0092] Specifically, taking the ith branch node as the central node, all branch nodes adjacent to the ith branch node are extracted from the busbar line network to form a node-centralized feature set, denoted as NC i={j|j∈[1,J]}, where J represents the maximum code number of the branch nodes adjacent to the i-th branch node when the i-th branch node is the central node;
[0093] Taking the synchronous maintenance of bus duct line network as the unified standard, in the service data cluster SDC (I ij ), extract the maintenance object I recorded during the vth maintenance of the bus duct line network ij Service label collection MR v , and constitute the behavioral characteristics, denoted as I ij :MR v , and I ij :MR v =MR v ; Node-centered feature set NC i For discretization guidance, all behavioral features formed during the vth maintenance of the bus duct line network are counted, and a maintenance behavior feature set is generated, which is recorded as
[0094] MBF v (NC i )={I ij :MR v |j∈NC i}.
[0095] Step S300: Analyze the continuous behavior characteristics of the bus duct line network maintenance according to the maintenance behavior characteristic set, and calculate the first service characteristic value of the maintenance object; sort the first service characteristic values according to the continuous behavior sequence of the bus duct line network maintenance, and generate a first service characteristic value sequence;
[0096] Specifically, according to the maintenance behavior feature set, the maintenance object I is calculated when the maintenance flow from the vth maintenance of the bus duct line network to the v+1th maintenance of the bus duct line network. ij The first service characteristic value is calculated as follows:
[0097]
[0098] Among them, FE[i ij (v)] indicates that when the maintenance flow from the vth maintenance of the bus duct line network to the v+1th maintenance of the bus duct line network, the maintenance object I ij The first service characteristic value, I ij :MR v+1 Indicates that when the bus duct line network is maintained for the v+1th time, the maintenance object I ij The corresponding behavior characteristics, NUM(I ij :MR v+1 -I ij :MR v ) indicates behavioral characteristics Iij :MR v+1 and behavioral characteristics I ij :MR v The number of maintenance item types included in the difference set, NUM(I ij :MR v+1 ) indicates behavioral characteristics I ij :MR v+1 The number of maintenance item types included in the
[0099] Let v = v + 1, return to step S301, and iterate and calculate the maintenance object I ij When v = u, the iteration stops; obtain the maintenance object I in the process of bus duct line network maintenance flow ij All the first service characteristic values generated are arranged in sequence according to the flow order from first to last to generate a first service characteristic value sequence, which is recorded as AL(I ij ), where AL(I ij ) indicates maintenance object I ij The corresponding first service characteristic value sequence is generated.
[0100] Step S400: construct a two-dimensional coordinate system, map the first service characteristic value sequence generated for each maintenance object to the two-dimensional coordinate system in the form of point coordinates, and generate a smooth curve function; analyze the second service characteristic value between any two smooth curve functions, and generate a synchronous secondary maintenance plan between the maintenance objects through function translation and iterative analysis;
[0101] Specifically, a two-dimensional coordinate system is constructed, and the first service feature value sequence generated corresponding to each maintenance object is represented by the point coordinates (v, FE[I ij (v)]) in the form of, mapped to a two-dimensional coordinate system, and the first service characteristic value sequence is converted into a smooth curve function by Matlab software; the first service characteristic value sequence AL(I ij ) corresponds to the smooth curve function transformed, denoted as f ij (t), t represents the independent variable of the smooth curve function;
[0102] Obtain any two smooth curve functions, analyze the second service characteristic value between any two smooth curve functions, and calculate the second service characteristic value between any two smooth curve functions. The specific calculation formula is as follows:
[0103]
[0104] Among them, R[f ij (t)→f ab (t)] represents the smooth curve function f ij (t) and the smooth curve function fab (t) between the second service characteristic value, → represents the first service characteristic value sequence AL(I ab ) corresponds to the smooth curve function transformed, I ab Indicates the maintenance object corresponding to the bus duct line segment formed between the adjacent a-th node and b-th node in the bus duct line network, f ab (t+T) represents the smooth curve function f ab (t) The smooth curve function obtained after the independent variable is translated by step length T. The positive and negative values of T correspond to the smooth curve function f ab (t) translate to the left or right, and ij≠ab;
[0105] Let T = T + 1, and iterate the calculation of the second service characteristic value between any two smooth curve functions until the iteration stops when the second service characteristic value is the maximum. The second service characteristic value corresponding to the maximum is recorded as R max [f ij (t)→f ab (t)];
[0106] Calculate the smooth curve function f ij (t) the maximum second service characteristic value between each smooth curve function;
[0107] The second service characteristic value threshold is preset, and the maximum second service characteristic value R max [f ij (t)→f ab (t)] is greater than or equal to the second service characteristic value threshold, then the maintenance object I is screened out. ab , and maintain object I ab Maintenance Object I ij Perform synchronization maintenance.
[0108] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0109] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A bus duct line management method based on artificial intelligence, characterized in that: The method comprises the following steps: Step S100: combing each branch node in the bus duct line network, forming a bus duct line segment between any two adjacent branch nodes, and creating a maintenance object based on the bus duct line segment; allocating a service data cluster, recording and storing a maintenance log of the maintenance object; Step S200: Taking the branch node as the central node, extract all branch nodes adjacent to the central node in the bus duct line network to form a node-centralized feature set; taking the synchronous maintenance of the bus duct line network as a unified scale, extract the service label set of the maintenance object recorded during the different maintenance of the bus duct line network in the service data cluster, taking the node-centralized feature set as a discretization guide, and counting all the behavior features formed during the different maintenance of the bus duct line network to generate a maintenance behavior feature set; Step S300: Analyze the continuous behavior characteristics of the bus duct line network maintenance according to the maintenance behavior characteristic set, and calculate the first service characteristic value of the maintenance object; sort the first service characteristic values according to the continuous behavior sequence of the bus duct line network maintenance, and generate a first service characteristic value sequence; Step S400: construct a two-dimensional coordinate system, map the first service characteristic value sequence generated for each maintenance object to the two-dimensional coordinate system in the form of point coordinates, and generate a smooth curve function; analyze the second service characteristic value between any two smooth curve functions, and generate a synchronous secondary maintenance plan between the maintenance objects through function translation and iterative analysis; The specific implementation process of step S100 includes: Step S101: comb out each branch node in the bus duct line network through the connection tree model; form a bus duct line segment between any two adjacent branch nodes, and create a maintenance object based on the bus duct line segment, and record any maintenance object as I ij , where I ij Indicates the maintenance object corresponding to the bus duct line segment formed between the adjacent i-th node and j-th node in the bus duct line network; Step S102: Building a bus duct line maintenance cloud platform, recording and storing maintenance logs of maintenance objects, wherein the maintenance logs record different types of maintenance items; allocating service data clusters, and one maintenance object is allocated one service data cluster, and any service data cluster allocated to any maintenance object is recorded as SDC (I ij )={MR v |v∈[1,u]}, where SDC(I ij ) indicates maintenance object I ij Corresponding to the assigned service data cluster, MR v Indicates that during the vth maintenance of the bus duct line network, maintenance object I ij The service label set of MR v ={MP x |x∈[1,w]}, where MP x Indicates the xth type of maintenance item, u indicates the number of maintenance times for the bus duct line network, and w indicates the type code number of the maintenance item type; The specific implementation process of step S200 includes: Step S201: Taking the ith branch node as the central node, extract all the branch nodes adjacent to the ith branch node in the bus duct line network to form a node-centralized feature set, denoted as NC i ={j|j∈[1,J]}, where J represents the maximum code number of the branch nodes adjacent to the i-th branch node when the i-th branch node is the central node; Step S202: Taking the synchronous maintenance of the bus duct line network as a unified standard, in the service data cluster SDC (I ij ), extract the maintenance object I recorded during the vth maintenance of the bus duct line network ij Service label collection MR v , and constitute the behavioral characteristics, denoted as I ij :MR v , and I ij :MR v =MR v ; Node-centered feature set NC i For discretization guidance, all behavioral features formed during the vth maintenance of the bus duct line network are counted, and a maintenance behavior feature set is generated, denoted as MBF v (NC i )={I ij :MR v |j∈NC i }; The specific implementation process of step S300 includes: Step S301: According to the maintenance behavior feature set, calculate the maintenance object I when the maintenance flow from the vth maintenance of the bus duct line network to the v+1th maintenance of the bus duct line network ij The first service characteristic value is calculated as follows: Among them, FE[I ij (v)] indicates that when the maintenance flow from the vth maintenance of the bus duct line network to the v+1th maintenance of the bus duct line network, the maintenance object I ij The first service characteristic value, I ij :MR v+1 Indicates that when the bus duct line network is maintained for the V+1th time, the maintenance object I ij The corresponding behavior characteristics, NUM(I ij :MR v+1 -I ij :MR v ) indicates behavioral characteristics I ij :MR v+1 and behavioral characteristics I ij :MR v The number of maintenance item types included in the difference set, NUM(I ij :MR v+1 ) indicates behavioral characteristics I ij :MR v+1 The number of maintenance item types included; Step S302: Let v = v + 1, return to step S301, and iterate and calculate the maintenance object I ij When v = u, the iteration stops; obtain the maintenance object I in the process of bus duct line network maintenance flow ij All the first service characteristic values generated are arranged in sequence according to the flow order from first to last to generate a first service characteristic value sequence, which is recorded as AL(I ij ), where AL(I ij ) indicates maintenance object I ij A correspondingly generated first service characteristic value sequence; The specific implementation process of step S400 includes: Step S401: construct a two-dimensional coordinate system, and map the first service feature value sequence generated for each maintenance object to point coordinates (v, FE[I ij (v)]) in the form of, mapped to a two-dimensional coordinate system, and the first service characteristic value sequence is converted into a smooth curve function by Matlab software; the first service characteristic value sequence AL(I ij ) corresponds to the smooth curve function transformed, denoted as f ij (t), t represents the independent variable of the smooth curve function; Step S402: Obtain any two smooth curve functions, analyze the second service characteristic value between any two smooth curve functions, and calculate the second service characteristic value between any two smooth curve functions. The specific calculation formula is as follows: Among them, R[f ij (t)→f ab (t)] represents the smooth curve function f ij (t) and the smooth curve function f ab (t) the second service characteristic value between ab (t) represents the first service characteristic value sequence AL(I ab ) corresponds to the smooth curve function transformed, I ab Indicates the maintenance object corresponding to the bus duct line segment formed between the adjacent a-th node and b-th node in the bus duct line network, f ab (t+T) represents the smooth curve function f ab (t) The smooth curve function obtained after the independent variable is translated by step length T. The positive and negative values of T correspond to the smooth curve function f ab (t) translate to the left or right, and ij≠ab; Let T = T + 1, and iterate the calculation of the second service characteristic value between any two smooth curve functions until the iteration stops when the second service characteristic value is the maximum. The second service characteristic value corresponding to the maximum is recorded as R max [f ij (t)→f ab (t)]; Calculate the smooth curve function f ij (t) the maximum second service characteristic value between each smooth curve function; The second service characteristic value threshold is preset, and the maximum second service characteristic value R max [f ij (t)→f ab (t)] is greater than or equal to the second service characteristic value threshold, then the maintenance object I is screened out. ab , and maintain object I ab Maintenance Object I ij Perform synchronization maintenance.
2. A bus duct line management system based on artificial intelligence, characterized in that: The system comprises: a data source processing module, a data relationship analysis module, a feature sequence generation module and an iteration analysis module; The data source processing module is used to sort out each branch node in the bus duct line network, form a bus duct line segment between any two adjacent branch nodes, and create a maintenance object based on the bus duct line segment; allocate a service data cluster, record and store a maintenance log of the maintenance object; The data relationship analysis module is used to extract all branch nodes adjacent to the central node in the bus duct line network with the branch node as the central node to form a node-centric feature set; take the synchronous maintenance of the bus duct line network as a unified scale, extract the service label set of the maintenance object recorded during the different maintenance of the bus duct line network in the service data cluster, take the node-centric feature set as a discretization guide, count all the behavior features formed during the different maintenance of the bus duct line network, and generate a maintenance behavior feature set; The feature sequence generation module analyzes the continuous behavior characteristics of the bus duct line network maintenance according to the maintenance behavior feature set, calculates the first service feature value of the maintenance object; sorts the first service feature values according to the continuous behavior sequence of the bus duct line network maintenance, and generates a first service feature value sequence; The iterative analysis module is used to construct a two-dimensional coordinate system, map the first service characteristic value sequence generated corresponding to each maintenance object to the two-dimensional coordinate system in the form of point coordinates, and generate a smooth curve function; analyze the second service characteristic value between any two smooth curve functions, and generate a synchronous secondary maintenance plan between the maintenance objects through function translation and iterative analysis; The data source processing module also includes a data object construction unit and a data cluster allocation unit; The data object construction unit is used to sort out each branch node in the bus duct line network through the connection tree model; form a bus duct line segment between any two adjacent branch nodes, and create a maintenance object based on the bus duct line segment, and record any maintenance object as I ij , where I ij Indicates the maintenance object corresponding to the bus duct line segment formed between the adjacent i-th node and j-th node in the bus duct line network; The data cluster allocation unit is used to build a bus duct line maintenance cloud platform, record and store the maintenance log of the maintenance object, and the maintenance log records different types of maintenance items; allocate service data clusters, and one maintenance object is correspondingly allocated a service data cluster, and any service data cluster corresponding to a maintenance object is recorded as SDC (I ij )={MR v |v∈[1,u]}, where SDC(I ij ) indicates maintenance object I ij Corresponding to the assigned service data cluster, MR v Indicates that during the vth maintenance of the bus duct line network, maintenance object I ij The service label set of MR v ={MP x |x∈[1,w]}, where MP x Indicates the xth type of maintenance item, u indicates the number of maintenance times for the bus duct line network, and w indicates the type code number of the maintenance item type; The data relationship analysis module also includes a node centralization feature unit and a behavior integration unit; The node-centric feature unit is used to extract all branch nodes adjacent to the i-th branch node in the bus duct line network with the i-th branch node as the central node to form a node-centric feature set, which is recorded as NC i ={j|j∈[1,J]}, where J represents the maximum code number of the branch nodes adjacent to the i-th branch node when the i-th branch node is the central node; The behavior integration unit is used to use the synchronous secondary maintenance of the bus duct line network as a unified standard, in the service data cluster SDC (I ij ), extract the maintenance object I recorded during the vth maintenance of the bus duct line network ij Service label collection MR v , and constitute the behavioral characteristics, denoted as I ij :MR v , and I ij :MR v =MR v ; Node-centered feature set NC i For discretization guidance, all behavioral features formed during the vth maintenance of the bus duct line network are counted, and a maintenance behavior feature set is generated, denoted as MBF v (NC i )={I ij :MR v |j∈NC i }; The feature sequence generation module also includes a maintenance flow feature analysis unit and a feature sequence generation unit; The maintenance flow feature analysis unit calculates, based on the maintenance behavior feature set, the maintenance object I when the maintenance flow from the vth maintenance of the bus duct line network to the v+1th maintenance of the bus duct line network. ij The first service characteristic value is calculated as follows: Among them, FE[I ij (v)] indicates that when the maintenance flow from the vth maintenance of the bus duct line network to the v+1th maintenance of the bus duct line network, the maintenance object I ij The first service characteristic value, I ij :MR v+1 Indicates that when the bus duct line network is maintained for the v+1th time, the maintenance object I ij The corresponding behavior characteristics, NUM(I ij :MR v+1 -I ij :MR v ) indicates behavioral characteristics I ij :MR v+1 and behavioral characteristics I ij :MR v The number of maintenance item types included in the difference set, NUM(I ij :MR v+1 ) indicates behavioral characteristics I ij :MR v+1 The number of maintenance item types included in the The feature sequence generation unit is used to set v=v+1, return to step S301, and iterate and calculate the maintenance object I in turn. ij When v = u, the iteration stops; obtain the maintenance object I in the process of bus duct line network maintenance flow ij All the first service characteristic values generated are arranged in sequence according to the flow order from first to last to generate a first service characteristic value sequence, which is recorded as AL(I ij ), where AL(I ij ) indicates maintenance object I ij A correspondingly generated first service characteristic value sequence; The iterative analysis module also includes a function generation unit and a solution decision unit; The function generation unit is used to construct a two-dimensional coordinate system, and the first service feature value sequence generated corresponding to each maintenance object is converted into a point coordinate (v, FE[I ij (v)]) in the form of, mapped to a two-dimensional coordinate system, and the first service characteristic value sequence is converted into a smooth curve function by Matlab software; the first service characteristic value sequence AL(I ij ) corresponds to the smooth curve function transformed, denoted as f ij (t), t represents the independent variable of the smooth curve function; The solution decision unit is used to obtain any two smooth curve functions, analyze the second service characteristic value between any two smooth curve functions, and calculate the second service characteristic value between any two smooth curve functions. The specific calculation formula is as follows: Among them, R[f ij (t)→f ab (t)] represents the smooth curve function f ij (t) and the smooth curve function f ab (t) the second service characteristic value between ab (t) represents the first service characteristic value sequence AL(I ab ) corresponds to the smooth curve function transformed, I ab Indicates the maintenance object corresponding to the bus duct line segment formed between the adjacent a-th node and b-th node in the bus duct line network, f ab (t+T) represents the smooth curve function f ab (t) The smooth curve function obtained after the independent variable is translated by step length T. The positive and negative values of T correspond to the smooth curve function f ab (t) translate to the left or right, and ij≠ab; Let T = T + 1, and iterate the calculation of the second service characteristic value between any two smooth curve functions until the iteration stops when the second service characteristic value is the maximum. The second service characteristic value corresponding to the maximum is recorded as R max [f ij (t)→f ab (t)]; Calculate the smooth curve function f ij (t) the maximum second service characteristic value between each smooth curve function; The second service characteristic value threshold is preset, and the maximum second service characteristic value R max [f ij (t)→f ab (t)] is greater than or equal to the second service characteristic value threshold, then the maintenance object I is screened out. ab , and maintain object I ab Maintenance Object I ij Perform synchronization maintenance.
