A method for constructing an information matrix of the state of electromechanical systems of highway bridges

Abstract

The present invention discloses a method for constructing an information matrix of the state of an electromechanical system of a highway bridge, including: determining the functional structure and physical structure of the electromechanical system of a highway bridge; collecting front-end electrical parameters through a highway data acquisition and monitoring system to construct an original electrical quantity data set; based on the structure of the electromechanical system of the highway bridge and the standardized format of data records, using regular expression matching and character segmentation clustering, and sorting the data according to the type of electrical quantity parameters, constructing an information matrix, and using an iterative method to standardize the data format; separating parameters and states into two matrices, and using grayscale images to express the parameters and state matrices. The present invention converts the electrical information collected by the electromechanical distribution group of the highway bridge into an information matrix method of a time series, so as to facilitate the classification and prediction of the faults of the electromechanical system of the highway bridge, and can provide technical support for the management and maintenance of the electromechanical system of the bridge in the highway scene.

Description

Technical Field

[0001] The patent of this invention relates to the research fields of intelligent transportation and smart highways, and specifically to a method for constructing an information matrix of the state of the electromechanical system of a highway bridge. Background Art

[0002] With the continuous development of intelligent assisted driving and ETC free-flow toll collection technology, more and more intelligent transportation electromechanical equipment are being used in highways, and the growing need to perceive and predict the health status of complex highway bridge electromechanical systems has brought new topics and challenges to intelligent transportation technology. There are two major difficulties in the state perception and operation and maintenance of traditional highway bridge electromechanical systems: 1) The acquisition system status information of the electromechanical equipment monitoring system is single and cannot identify the cause of the fault; 2) There is a lack of an integrated equipment management and monitoring system, the standard data format is chaotic, and even some inspection work requires manual recording and reporting, resulting in the data not having the formatting ability for big data research, which greatly reduces the overall efficiency of automated operation and maintenance and intelligent fault operation and maintenance.

[0003] The current research results in this field mainly focus on the fault diagnosis and prediction of electromechanical equipment or transformers, while a systematic method has not yet been formed for the monitoring of electromechanical systems of highway bridges. Therefore, this paper proposes a method for constructing an information matrix of the state of electromechanical systems of highway bridges. Summary of the invention

[0004] Purpose of the invention: In order to overcome the shortcomings in the monitoring technology of the electromechanical system of highway bridges, a method for constructing an information matrix of the state of the electromechanical system of highway bridges is provided. The method uses a regular expression traversal loop method to effectively convert the monitoring data of the electromechanical system of highway bridges into an information matrix, providing support for subsequent deep data mining and learning.

[0005] Technical solution: To achieve the above-mentioned purpose, the present invention provides a method for constructing an information matrix of the state of an electromechanical system of a highway bridge, comprising the following steps:

[0006] S1: Determine the functional and physical structure of the electromechanical system of highway bridges;

[0007] S2: The front-end electrical parameters are collected through the highway supervisory control and data acquisition system (SCADA) at a frequency of once an hour, 24 hours a day, 365 days a year to build the original electrical quantity data set;

[0008] S3: Based on the structure of the electromechanical system of highway bridges and the standardized format of data records, regular expression matching and character segmentation clustering are used, and the data are sorted according to the type of electrical quantity parameters to construct an information matrix, in which rows represent parameter values ​​and states, and columns represent time. The data format is standardized using an iterative method;

[0009] S4: Separate the parameters and states into two matrices, and use grayscale images to express the parameter and state matrices.

[0010] Furthermore, the method for determining the functional structure and physical structure of the electromechanical system of the highway bridge in step S1 is:

[0011] The functional structure of the electromechanical system of the highway bridge includes three major systems: traffic monitoring system, toll collection system and communication system. Its derivative systems also include: power supply and distribution system, lighting system, gantry system, toll collection system, navigation light system, dehumidification system, ventilation and heating system, energy system, etc. Each functional system collects and monitors the data of electromechanical equipment and line loops through the Supervisory Control And Data Acquisition (SCADA), ensures the connection and transmission of data through the communication system, and performs centralized control of software and hardware through the computer system.

[0012] The highway bridge electromechanical centralized control system consists of a road section monitoring substation and a bridge section monitoring substation. All data will be transmitted to the highway data acquisition and monitoring system through the optical fiber network. The structure of the road section monitoring substation is mainly based on the substation. The system is mainly composed of several feeder cabinets, high-voltage measurement and control units, UPS, transformer temperature controllers, generators, fans and lighting, and is equipped with video monitoring equipment; the industrial Ethernet switch uploads the equipment parameters and status to the database, among which, each toll station, service area substation, and switch station is respectively equipped with a road section monitoring substation. The bridge section monitoring substation is composed of a substation, a medium-voltage power supply system and several buried transformers, dehumidification facilities, elevators, lighting distribution boxes, aviation, navigation lights, etc., and is also equipped with several feeder cabinets, transformer temperature controllers, and UPS equipment; the industrial Ethernet switch uploads the equipment parameters and status to the database, among which, the middle tower, south tower, north anchor, north anchor, south anchor and Jiajiang bridge section are each equipped with a bridge section monitoring substation.

[0013] Further, in step S2, the front-end electrical parameters are collected by the highway data acquisition and monitoring system (SCADA) at a frequency of 365 days a year, 24 hours a day, and once an hour, and the method for constructing the original electrical quantity data set is: according to the arrangement rules of the data record items, the process of establishing a sparse two-dimensional information matrix can be regarded as the decomposition of the main branch structure. The structure of the highway bridge electromechanical centralized control system can be roughly expressed as a tree structure of "SCADA data acquisition and control system"-"monitoring substation"-"distribution or equipment line loop"-"sub-line electrical quantity parameters". The original electrical quantity data set is constructed by collecting the front-end electrical parameters through the SCADA system, as shown in Table 1. Among them, the electrical quantity collection interval is 1 hour, and the collection is carried out from 0 o'clock to 24 o'clock every day, with a total of 25 sampling points. The main equipment and lines collected are comprehensive relay protection equipment, 6kV incoming line, low-voltage main switch loop, 0.4kV feeder, generator output, field lighting loop and UPS, and the main technical parameters are three-phase voltage, three-phase current, active power, reactive power, power factor, frequency, etc.

[0014] Among them, the main parameters and technical indicators collected are as follows:

[0015] Comprehensive relay protection equipment: current, voltage, active power, reactive power, power factor, frequency;

[0016] 6kV incoming line: three-phase voltage, current, active power, reactive power, power factor, frequency;

[0017] Low-voltage main switch circuit: three-phase voltage, current, active power, reactive power, power factor, frequency;

[0018] 0.4kV feeder: three-phase voltage, three-phase current, power factor, active power, reactive power, frequency;

[0019] Generator output: three-phase voltage, current, active power, reactive power, power factor, frequency;

[0020] ·Outdoor lighting circuit: three-phase voltage, current, active power, reactive power, power factor;

[0021] UPS: input / output voltage, frequency, battery operating parameters, load.

[0022] Table 1. Original data of electrical quantities of highway bridges and expressway bridge electromechanical systems 20170101 (partial)

[0023]

[0024] Furthermore, the specific steps of using regular expression matching and character segmentation in step S3 to cluster and sort the data and construct an information matrix are as follows:

[0025] ① Use regular expressions to find Chinese matches for data record items, i.e., site names, and filter using the encoding interval of Unicode characters [\u4e00-\u9fa5].

[0026] ② If Chinese is found, the output is site, otherwise the output is <error>;

[0027] ③ Split into three substrings based on the symbols ".." and "_", str_subname[0], str_subname[1] and str_subname[2];

[0028] ④str_subname[0] indicates the monitoring substation name. Str_subname[0] is checked. If it does not match the current monitoring substation name, the data record line is deleted.

[0029] ⑤str_subname[1] indicates the line loop or device name, and clustering is performed based on string matching;

[0030] ⑥str_subname[2] indicates the name of the electrical quantity parameter, which is arranged in the order of current, voltage, active power, reactive power, power factor, frequency and others.

[0031] The data records are arranged in an orderly structure as shown in Table 2.

[0032] Table 2 Original data of electrical quantities of highway bridge and expressway bridge electromechanical system at a certain time (ordered)

[0033]

[0034] Furthermore, the specific steps of dividing the ordered electrical quantity matrix into a parameter matrix and a state matrix and forming a grayscale image in step S4 are as follows:

[0035] S4-1: According to the characteristics of the interval arrangement of parameter and state data, a circular method is used to extract the parameter matrix and state matrix from the information matrix obtained in S3;

[0036] S4-2: Express the parameter vector of each substation at each moment using a grayscale image;

[0037] S4-3: The state matrix of each substation at a certain period of time is expressed as a grayscale image.

[0038] Furthermore, in step S4-2, the step of using a grayscale image to express the parameter vector of each substation at each moment is as follows:

[0039] ① Through the analysis of the example of the electromechanical system of the highway bridge, each substation has an average of about 800 line parameters. Based on this parameter quantity, a 60×15 matrix format can be designed;

[0040] ②Convert the data obtained at each substation at each moment into a 60×15 matrix. Since the number of lines at different stations is inconsistent, the empty parts are filled with -1;

[0041] ③ Use min-max standardization to normalize the data;

[0042] ④Use Z-score Normalization to standardize the data;

[0043] ⑤Map the numerical value to [0,255] and output it as a grayscale image.

[0044] Furthermore, in step S4-3, each substation is expressed as a grayscale image.

[0045] The steps of the state matrix for a period of time are as follows:

[0046] ① The acquisition frequency of the electromechanical system of highway bridges is once per hour, so the time dimension of the annual status information matrix is ​​365×24;

[0047] ②The line number dimension is the line parameters actually owned by each substation, and the parameters at each moment are expressed as each column of the substation;

[0048] ③ Since the state matrix contains only values ​​0 and 1, 1 is mapped to 255 and the output is a grayscale image;

[0049] The present invention converts the electrical information collected by the electromechanical distribution group of the highway bridge into an information matrix method of time series, so as to facilitate the subsequent classification and prediction of system faults, and can provide technical support for the management and maintenance of the bridge electromechanical system in the highway scene.

[0050] Beneficial effects: The data processing method of the information matrix proposed in the present invention has an efficiency optimization effect on the identification of regional faults. At the same time, the construction based on the information matrix improves the scalability of the big data maintenance of the electromechanical system. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a functional structure diagram of the electromechanical system of the highway bridge in the present invention.

[0052] Figure 2 It is a physical structure diagram of the electromechanical system of the highway bridge in the present invention.

[0053] Figure 3 It is a process diagram of converting data into an information matrix in the present invention.

[0054] Figure 4 It is the parameter information matrix at a certain moment in the present invention.

[0055] Figure 5 It is the annual status information matrix in the present invention. DETAILED DESCRIPTION

[0056] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0057] A method for constructing an information matrix of the state of an electromechanical system of a highway bridge comprises the following steps:

[0058] S1: Determine the functional and physical structure of the electromechanical system of highway bridges;

[0059] S2: The front-end electrical parameters are collected through the highway supervisory control and data acquisition system (SCADA) at a frequency of once an hour, 24 hours a day, 365 days a year to build the original electrical quantity data set;

[0060] S3: Based on the structure of the electromechanical system of highway bridges and the standardized format of data records, regular expression matching and character segmentation clustering are used, and the data are sorted according to the type of electrical quantity parameters to construct an information matrix, in which rows represent parameter values ​​and states, and columns represent time. The data format is standardized using an iterative method;

[0061] S4: Separate the parameters and states into two matrices, and use grayscale images to express the parameter and state matrices.

[0062] Furthermore, the method for determining the functional structure and physical structure of the electromechanical system of the highway bridge in step S1 is:

[0063] The functional structure of the electromechanical system of the highway bridge includes three major systems: traffic monitoring system, toll collection system and communication system. Its derivative systems also include: power supply and distribution system, lighting system, gantry system, toll collection system, navigation light system, dehumidification system, ventilation and heating system, energy system, etc. Each functional system collects and monitors the data of electromechanical equipment and line loops through the Supervisory Control And Data Acquisition (SCADA), ensures the connection and transmission of data through the communication system, and performs centralized control of software and hardware through the computer system. The system logic structure is as follows: Figure 1 shown.

[0064] The physical structure of the highway bridge electromechanical system consists of a road section monitoring substation and a bridge section monitoring substation. All data will be transmitted to the highway data acquisition and monitoring system through the optical fiber network. The structure of the road section monitoring substation is mainly based on the substation. The system is mainly composed of several feeder cabinets, high-voltage measurement and control units, UPS, transformer temperature controllers, generators, fans and lighting, and is equipped with video monitoring equipment; the industrial Ethernet switch uploads the equipment parameters and status to the database, among which, each toll station, service area substation, and switch station is respectively equipped with a road section monitoring substation. The bridge section monitoring substation is composed of a substation, a medium-voltage power supply system and several buried transformers, dehumidification facilities, elevators, lighting distribution boxes, aviation, navigation lights, etc., and is also equipped with several feeder cabinets, transformer temperature controllers, and UPS equipment; the industrial Ethernet switch uploads the equipment parameters and status to the database, among which, the middle tower, south tower, north anchor, north anchor, south anchor and Jiajiang bridge section are each equipped with a bridge section monitoring substation.

[0065] Furthermore, in step S2, the front-end electrical parameters are collected by the highway data acquisition and monitoring system (SCADA) at a frequency of 365 days a year, 24 hours a day, and once an hour, and the method for constructing the original electrical quantity data set is: according to the arrangement rules of the data record items, the process of establishing a sparse two-dimensional information matrix can be regarded as the decomposition of the main branch structure. The structure of the highway bridge electromechanical centralized control system can be roughly expressed as a tree structure of "SCADA data acquisition and control system"-"monitoring substation"-"distribution or equipment line loop"-"sub-line electrical quantity parameters", such as Figure 2 As shown. The SCADA system collects the front-end electrical parameters and constructs the original electrical quantity data set, as shown in Table 1. Among them, the electrical quantity collection interval is 1 hour, from 0 to 24 every day, with a total of 25 sampling points. The main equipment and lines collected are comprehensive relay protection equipment, 6kV incoming line, low-voltage main switch circuit, 0.4kV feeder, generator output, field lighting circuit and UPS. The main technical parameters are three-phase voltage, three-phase current, active power, reactive power, power factor, frequency, etc.

[0066] Among them, the main parameters and technical indicators collected are as follows:

[0067] Comprehensive relay protection equipment: current, voltage, active power, reactive power, power factor, frequency;

[0068] 6kV incoming line: three-phase voltage, current, active power, reactive power, power factor, frequency;

[0069] Low-voltage main switch circuit: three-phase voltage, current, active power, reactive power, power factor, frequency;

[0070] 0.4kV feeder: three-phase voltage, three-phase current, power factor, active power, reactive power, frequency;

[0071] Generator output: three-phase voltage, current, active power, reactive power, power factor, frequency;

[0072] ·Outdoor lighting circuit: three-phase voltage, current, active power, reactive power, power factor;

[0073] UPS: input / output voltage, frequency, battery operating parameters, load.

[0074] Table 1. Original data of electrical quantities of highway bridges and expressway bridge electromechanical systems 20170101 (partial)

[0075]

[0076] Furthermore, the specific steps of using regular expression matching and character segmentation in step S3 to cluster and sort the data and construct an information matrix are as follows:

[0077] ① Use regular expressions to find Chinese matches for data record items, i.e., site names, and filter using the encoding interval of Unicode characters [\u4e00-\u9fa5].

[0078] ② If Chinese is found, the output is site, otherwise the output is <error>;

[0079] ③ Split into three substrings based on the symbols ".." and "_", str_subname[0], str_subname[1] and str_subname[2];

[0080] ④str_subname[0] indicates the monitoring substation name. Str_subname[0] is checked. If it does not match the current monitoring substation name, the data record line is deleted.

[0081] ⑤str_subname[1] indicates the line loop or device name, and clustering is performed based on string matching;

[0082] ⑥str_subname[2] indicates the name of the electrical quantity parameter, which is arranged in the order of current, voltage, active power, reactive power, power factor, frequency and others.

[0083] The process of constructing the data record table into a two-dimensional information matrix is ​​as follows Figure 3 The data records are arranged in an orderly structure as shown in Table 2.

[0084] Table 2 Original data of electrical quantities of highway bridge and expressway bridge electromechanical system at a certain time (ordered)

[0085]

[0086] Furthermore, the specific steps of dividing the ordered electrical quantity matrix into a parameter matrix and a state matrix and forming a grayscale image in step S4 are as follows:

[0087] S4-1: According to the characteristics of the interval arrangement of parameter and state data, a circular method is used to extract the parameter matrix and state matrix from the information matrix obtained in S3;

[0088] S4-2: Express the parameter vector of each substation at each moment using a grayscale image;

[0089] S4-3: The state matrix of each substation at a certain period of time is expressed as a grayscale image.

[0090] Furthermore, in step S4-2, the step of using a grayscale image to express the parameter vector of each substation at each moment is as follows:

[0091] ① Through the analysis of the example of the electromechanical system of the highway bridge, each substation has an average of about 800 line parameters. Based on this parameter quantity, a 60×15 matrix format can be designed;

[0092] ②Convert the data obtained at each substation at each moment into a 60×15 matrix. Since the number of lines at different stations is inconsistent, the empty parts are filled with -1;

[0093] ③ Use min-max standardization to normalize the data;

[0094] ④Use Z-score Normalization to standardize the data;

[0095] ⑤Map the numerical value to [0,255] and output it as a grayscale image.

[0096] The parameter information matrix of a substation at a certain time is as follows: Figure 4 shown.

[0097] Furthermore, in step S4-3, each substation is expressed as a grayscale image.

[0098] The steps of the state matrix for a period of time are as follows:

[0099] ① The acquisition frequency of the electromechanical system of highway bridges is once per hour, so the time dimension of the annual status information matrix is ​​365×24;

[0100] ②The line number dimension is the line parameters actually owned by each substation, and the parameters at each moment are expressed as each column of the substation;

[0101] ③ Since the state matrix contains only values ​​0 and 1, 1 is mapped to 255 and the output is a grayscale image;

[0102] Status information matrix of a substation in a certain year Figure 5 shown.

[0103] Example

[0104] A method for constructing an information matrix of a highway bridge electromechanical system state comprises the following steps:

[0105] S1: Determine the functional and physical structure of the electromechanical system of highway bridges;

[0106] Based on the research on the functional structure and physical structure of the electromechanical system of the highway bridge in the present invention, the number, type, name, etc. of monitoring substations of the electromechanical system of the highway bridge are clarified according to the actual situation of the specific project.

[0107] S2: The front-end electrical parameters are collected through the highway supervisory control and data acquisition system (SCADA) at a frequency of once an hour, 24 hours a day, 365 days a year to build the original electrical quantity data set;

[0108] The electrical quantity collection interval is 1 hour, from 0 o'clock to 24 o'clock every day, with a total of 25 sampling points. The main equipment and lines collected are comprehensive relay protection equipment, 6kV incoming line, low-voltage main switch circuit, 0.4kV feeder, generator output, field lighting circuit and UPS. The main technical parameters are three-phase voltage, three-phase current, active power, reactive power, power factor, frequency, etc. After completing the data collection, the data structure can be represented as a tree structure of "SCADA data acquisition and control system"-"monitoring substation"-"distribution or equipment line circuit"-"sub-line electrical quantity parameters" to form the original electrical quantity data set.

[0109] S3: Based on the structure of the electromechanical system of highway bridges and the standardized format of data records, regular expression matching and character segmentation clustering are used, and the data are sorted according to the type of electrical quantity parameters to construct an information matrix, in which rows represent parameter values ​​and states, and columns represent time. The data format is standardized using an iterative method;

[0110] Specifically:

[0111] ① Use regular expressions to find Chinese matches for data record items, i.e., site names, and filter using the encoding interval of Unicode characters [\u4e00-\u9fa5].

[0112] ② If Chinese is found, the output is site, otherwise the output is <error>;

[0113] ③ Split into three substrings based on the symbols ".." and "_", str_subname[0], str_subname[1] and str_subname[2];

[0114] ④str_subname[0] indicates the monitoring substation name. Str_subname[0] is checked. If it does not match the current monitoring substation name, the data record line is deleted.

[0115] ⑤str_subname[1] indicates the line loop or device name, and clustering is performed based on string matching;

[0116] ⑥str_subname[2] indicates the name of the electrical quantity parameter, which is arranged in the order of current, voltage, active power, reactive power, power factor, frequency and others.

[0117] S4: Separate the parameters and states into two matrices, and use grayscale images to express the parameters and state matrices, which are specifically:

[0118] S4-1: According to the characteristics of the interval arrangement of parameter and state data, a circular method is used to extract the parameter matrix and state matrix from the information matrix obtained in S3;

[0119] S4-2: Express the parameter vector of each substation at each moment using a grayscale image;

[0120] ① Through the analysis of the example of the electromechanical system of the highway bridge, each substation has an average of about 800 line parameters. Based on this parameter quantity, a 60×15 matrix format can be designed;

[0121] ②Convert the data obtained at each substation at each moment into a 60×15 matrix. Since the number of lines at different stations is inconsistent, the empty parts are filled with -1;

[0122] ③ Use min-max standardization to normalize the data;

[0123] ④Use Z-score Normalization to standardize the data;

[0124] ⑤Map the numerical value to [0,255] and output it as a grayscale image.

[0125] S4-3: The state matrix of each substation in a certain period of time is expressed as a grayscale image, which is specifically:

[0126] ① The acquisition frequency of the electromechanical system of highway bridges is once per hour, so the time dimension of the annual status information matrix is ​​365×24;

[0127] ②The line number dimension is the line parameters actually possessed by each substation, and the parameters at each moment are represented by each column of the substation; ③Since the state matrix contains only values ​​0 and 1, 1 is mapped to 255 and the output is a grayscale image.< / error> < / error> < / error>

Claims

1. A method for constructing an information matrix of the state of an electromechanical system of a highway bridge, characterized by: The following steps are involved: S1: Determine the functional and physical structure of the electromechanical system of highway bridges; S2: The highway data acquisition and monitoring system, i.e., the SCADA system, collects front-end electrical parameters once an hour, 24 hours a day, 365 days a year to build a raw electrical quantity data set; S3: Based on the standardized format of the electromechanical system structure and electrical quantity data set records of highway bridges, regular expression matching and character segmentation clustering are used, and the data are sorted according to the type of electrical quantity parameters to construct an information matrix, in which rows represent parameter values ​​and states, columns represent time, and an iterative method is used to standardize the data format; S4: The ordered electrical quantity matrix is ​​divided into a parameter matrix and a state matrix, the values ​​are mapped to [0, 255], and the parameter and state matrices are expressed using grayscale images; The ordered electrical quantity matrix is ​​divided into a parameter matrix and a state matrix, and a grayscale image is formed. In the step S1, the functional structure and physical structure of the electromechanical system of the highway bridge are determined; The functional structure of the electromechanical system of the highway bridge includes three major systems: traffic monitoring system, toll collection system and communication system; each system collects and monitors data of electromechanical equipment and line loops through the data acquisition and monitoring control system, ensures the connection and transmission of data through the communication system, and performs centralized control of software and hardware through the computer system; The physical structure of the electromechanical system of the highway bridge consists of a road section monitoring substation and a bridge section monitoring substation. Each toll station, service area substation, and switch station is equipped with a road section monitoring substation, and the middle tower, south tower, north anchor, south anchor and Jiajiang Bridge section are equipped with a bridge section monitoring substation; The step S2 constructs the original electrical quantity data set, which specifically includes the following steps: S2-1: From the perspective of electrical quantity acquisition data structure, the structure of the electromechanical system of highway bridges can be expressed as a tree structure of "data acquisition and monitoring system, i.e. SCADA system" - "monitoring substation" - "power distribution or equipment line loop" - "sub-line electrical quantity parameters"; S2-2: The SCADA system collects the front-end electrical parameters and constructs the original electrical quantity data set. The electrical quantity collection interval is 1 hour, from 0:00 to 24:00 every day, with a total of 25 sampling points. The main equipment and lines collected include comprehensive relay protection equipment, 6kV incoming line, low-voltage main switch circuit, 0.4kV feeder, generator output, outdoor lighting circuit and UPS. The main technical parameters include three-phase voltage, three-phase current, active power, reactive power, power factor and frequency. In step S3, regular expression matching and character segmentation clustering are used to sort the data according to the type of electrical quantity parameters to construct an information matrix. The specific steps are as follows: S3-1: Use regular expressions to find Chinese matches for data record items, i.e., site names, and filter using the encoding interval of Unicode characters [\u4e00-\u9fa5]; S3-2: If Chinese is found, the output is site, otherwise the output is <error> ；< / error> S3-3: Split into three substrings based on the symbols ".." and "_", str_subname[0], str_subname[1] and str_subname[2]; S3-4: str_subname[0] indicates the monitoring substation name. Str_subname[0] is checked. If it does not match the current monitoring substation name, the data record row is deleted. S3-5: str_subname[1] represents the line loop or device name, and clustering is performed based on string matching; S3-6: str_subname[2] indicates the name of the electrical quantity parameter, which is arranged in the order of current, voltage, active power, reactive power, power factor, and frequency; In step S4, the ordered electrical quantity matrix is ​​divided into a parameter matrix and a state matrix, the values ​​are mapped to between [0, 255], and the parameters and state matrix are expressed using grayscale images. The specific steps are as follows: S4-1: According to the characteristics of the interval arrangement of parameter and state data, a circular method is used to extract the parameter matrix and state matrix from the information matrix obtained in S3; S4-2: The parameter vector of each substation at each moment is expressed as a grayscale image; S4-3: The state matrix of each substation in a certain period of time is expressed as a grayscale image; In the step S4-2, expressing the parameter vector of each substation at each moment using a grayscale image specifically includes: ① Through the example analysis of the electromechanical system of highway bridges, each substation has an average of about 800 line parameters, and a 60×15 matrix format is designed based on this parameter quantity; ②Convert the data obtained at each substation at each moment into a 60×15 matrix. Since the number of lines at different stations is inconsistent, the empty parts are filled with -1; ③ Use min-max standardization to normalize the data; ④Use Z-score Normalization to standardize the data; ⑤ Map the value to [0,255] and output it as a grayscale image; In step S4-3, the state matrix of each substation for a period of time is expressed using a grayscale image, specifically including: ① The acquisition frequency of the electromechanical system of highway bridges is once per hour, so the time dimension of the annual status information matrix is ​​365×24; ②The line number dimension is the line parameters actually owned by each substation, and the parameters at each moment are expressed as each column of the substation; ③ Since the state matrix contains only values ​​0 and 1, 1 is mapped to 255 and the output is a grayscale image.

Images