High-precision high-speed rail switch cabinet digital on-line supervision system

Abstract

The invention discloses a high-precision high-speed rail switch cabinet digital online supervision system, and relates to the field of switch cabinet supervision, and the system comprises a switch cabinet online monitoring system which is electrically connected with a display terminal, a cloud platform, a data center, an expert system, PC terminal software, a mobile terminal APP, an industrial control computer, and an intelligent gateway. The high-precision digital online monitoring system for the high-speed rail switch cabinet can give an alarm in real time, increase, modify and delete basic information of a user, can define, increase, delete and modify various parameters and monitoring points by a user, and can automatically monitor communication related parameter patrol period, re-mining times, database access parameter information, monitoring frequency, monitoring time and the like. The data refreshing time and the like are self-defined, and a user defines and sets a terminal interface and a communication protocol, so that the rail transit safety operation can be effectively improved, the operation and maintenance cost is reduced, and the operation and maintenance efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of switch cabinet supervision, and in particular to a high-precision digital online supervision system for high-speed railway switch cabinets. Background Art

[0002] Switchgear is also called complete switch or complete distribution device. It is an electrical equipment mainly composed of circuit breakers. It means that the manufacturer assembles the relevant high and low voltage electrical appliances, busbars, current-carrying conductors, insulators, etc. in a closed or open metal cabinet according to the requirements of the primary electrical wiring diagram. It is a device for receiving and distributing electric energy in the power system. It is called a metal switchgear, or switchgear for short. Low-voltage switchgear refers to one or more low-voltage switchgear and related control, measurement, signal, protection, regulation and other equipment. The manufacturer is responsible for completing all internal electrical and mechanical connections. It is a combination of structural components that are completely assembled together. High-voltage switchgear: a general term for high-voltage switches mainly used for power generation, transmission, distribution and energy conversion, as well as control, measurement, protection devices, electrical connections, casings, supports, etc.

[0003] At present, the digital online monitoring system of the switch cabinet of the high-speed railway system is still in its initial stage. It cannot effectively monitor the electrical and mechanical characteristics of the circuit breaker in the switch cabinet, the full current of the lightning arrester, the number of lightning strikes, and the partial discharge of the switch cabinet, so the safety of the railway power supply system cannot be guaranteed.

[0004] Therefore, it is necessary to propose a high-precision digital online supervision system for high-speed railway switchgear to solve the above problems. Summary of the invention

[0005] The main purpose of the present invention is to provide a high-precision digital online monitoring system for high-speed railway switchgear, which can effectively solve the problems in the background technology.

[0006] To achieve the above object, the technical solution adopted by the present invention is:

[0007] A high-precision high-speed railway switch cabinet digital online supervision system, including a switch cabinet online monitoring system, wherein the switch cabinet online monitoring system is electrically connected to a display terminal, a cloud platform, a data center, an expert system, a PC terminal software and a mobile terminal APP, an industrial computer, and an intelligent gateway;

[0008] The industrial computer cloud service is electrically connected to the data collector, and the data collector is electrically connected to the sensor and the built-in local storage;

[0009] The industrial computer is used to integrate and synchronize different types of sensor data to realize the collection and management of multi-sensor data. Through the fusion analysis of the collected data, real-time monitoring of the overall status of the switch cabinet, abnormality detection and alarm are realized, including on-site sound and light alarm and remote terminal alarm through the system.

[0010] Preferably, the display terminal is used to display the status of the switch cabinet through the gateway. The switch cabinet includes monitoring equipment, each monitoring equipment including 1 mechanical property sensor, 1 energy storage motor current sensor, 1 opening current sensor, 1 closing current sensor, 1 vibration sensor, 2 secondary CT current sensors, 1 two-in-one partial discharge sensor, 2 high-frequency current partial discharge sensors, 2 MOA current monitors, 1 circuit breaker electrical characteristic data collector, 1 high-frequency current partial discharge data collector, 1 lightning arrester data collector, 1 temperature and humidity data collector, an industrial computer, and a matching 12v / 24v power adapter.

[0011] Preferably, the mechanical property sensor is used to monitor the mechanical movement state of the circuit breaker of the switch cabinet and evaluate the health state of the circuit breaker to determine whether the switch can be closed or opened normally;

[0012] The current sensors include energy storage motor current sensors, opening current sensors, and closing current sensors, which are used to measure the current size and change trend of the circuit breaker during operation, evaluate whether the current is within the normal range, and detect abnormal current conditions.

[0013] Preferably, the secondary ct current sensor is used to monitor the synchronization of the three-phase currents A, B, and C during the operation of the switch cabinet circuit breaker;

[0014] The vibration sensor is used to identify vibrations caused by mechanical failure or abnormal movement, so as to timely discover potential mechanical problems;

[0015] The two-in-one partial discharge sensor is used to detect discharge activities at the cable insulator of the circuit breaker, and monitor and evaluate the aging, defects and damage status of the circuit breaker insulation material in real time;

[0016] The high-frequency current partial discharge sensor is used to detect high-frequency current in the cable shielding layer and armor layer, and reflect the insulation state of the cable, so as to detect partial discharge phenomenon early, prevent insulation failure and prevent electrical accidents.

[0017] Preferably, the MOA current sensor is used to detect the leakage current of the zinc oxide arrester in real time and evaluate the insulation and health status of the arrester;

[0018] The temperature and humidity sensor is used to monitor the temperature and humidity environment inside the cabinet to prevent damage to electrical equipment caused by excessively high humidity or low temperature. At the same time, it monitors abnormal temperature and humidity data inside the switch cabinet room to assist in determining whether there is abnormal operation of the section switch cabinet.

[0019] Preferably, the industrial computer includes local historical data storage and analysis. By storing sensor data in a local database, it provides data support for subsequent analysis of local equipment operating trends, aging assessment, fault analysis and equipment health management, helps local equipment perform predictive maintenance, optimizes maintenance plans, and thus improves system reliability and stability.

[0020] Preferably, the remote monitoring and system integration transmits data to the cloud through the network to achieve remote monitoring and remote management, supports integration with the substation's SCADA, EMS system, and user's own management system, improves the information and intelligent management capabilities of the power system, and includes time information and monitoring point ID information;

[0021] Each device in the system has a unique ID code, and can be customized. The coding must follow the following principles:

[0022] Each high-speed rail power supply line has a unique ID, which must be reflected in the coding;

[0023] Each switch cabinet room has a unique ID, which must be reflected in the coding;

[0024] Each switch cabinet has a unique ID, which must be reflected in the coding;

[0025] Each sensor in each switch cabinet is classified by function, and each sensor category must be reflected in the coding;

[0026] Each sensor has a unique ID, which must be reflected in the encoding;

[0027] In the same high-speed railway power supply line, switch cabinets of different voltage / power levels have unique IDs, which need to be reflected in the coding;

[0028] In the same high-speed railway power supply line, switch cabinets of the same voltage / power level, different brands, and different models have unique IDs, which need to be reflected in the coding;

[0029] Different high-speed rail power supply lines in the same power grid have unique IDs, which need to be reflected in the coding;

[0030] The sensor ID code must contain the corresponding ID information of the switch cabinet under test; the sensor ID code must be as simple and easy to understand as possible, and can be easily matched with the position of the switch cabinet under test during use;

[0031] The human-machine interface is equipped with a graphical HMI. The operator can locally view the data of each sensor and system status information. It also supports local parameter setting, historical data query, alarm log viewing, etc. to achieve local intuitive and convenient equipment management and maintenance.

[0032] Preferably, the industrial computer also includes a high-frequency current partial discharge data collector, a lightning arrester data collector and an environmental temperature and humidity data collector, which are installed by installing a standard rail at the double-row bracket on the right side of the secondary instrument room of the switch cabinet, and then fixing the collector housing on the rail with a rail buckle to achieve installation;

[0033] The data collector performs preprocessing and data storage on the sensor data; the memory adopts a power-off preservation design; the collector has its own clock system, and the uploaded monitoring results include: displacement, current, vibration, partial discharge, temperature and humidity data information, time information, and monitoring point ID information collected by the corresponding sensor;

[0034] The mobile terminal APP is used to view authorized monitoring data and accept the operation and maintenance management of the data center, and upload the operation and maintenance implementation information to the cloud platform.

[0035] Preferably, the intelligent gateway is used to upload the data collected by the data collector to the cloud platform in real time;

[0036] The cloud platform is used to store data collected by all sensors, analyze and display the data, and issue warnings for abnormal data;

[0037] The data center includes client PC software for real-time viewing of monitoring results, abnormal alarms, alarm processing progress, alarm point processing status and process records, and operation and maintenance management;

[0038] It includes an expert system, which collects and analyzes monitoring data and abnormal processing record data based on the expert system, optimizes the normal and abnormal data judgment and warning thresholds through self-learning, and provides more accurate accident prediction and operation and maintenance suggestions;

[0039] Includes computer host, monitor and supporting hardware equipment.

[0040] Preferably, when the data collector performs data transmission, the sensor first saves all the collected data locally in the corresponding data collector and uploads it to the corresponding industrial computer through the R485 line. The industrial computer uploads it to the cloud platform through the gateway, and the user can view and download the monitoring results in real time through the client software;

[0041] The data collector collects data from each sensor in the switch cabinet at a frequency of 1 minute per time and saves the data. The maximum storage capacity is 1 month.

[0042] The client software will automatically query the time information of the data collected by all sensors at an interval, where the time can be customized according to the actual situation. If data loss is found, it will automatically send a data re-send instruction to the corresponding industrial computer, and the corresponding industrial computer will re-upload the data of the missing time period to the cloud platform to ensure data integrity.

[0043] Compared with the prior art, the present invention provides a high-precision digital online monitoring system for high-speed railway switch cabinets, which has the following beneficial effects:

[0044] 1. This high-precision digital online supervision system for high-speed railway switchgear is based on the digital monitoring of 27.5kVGIS switchgear in the high-speed railway system. Through real-time monitoring of the electrical and mechanical characteristics of the circuit breaker, the full current of the lightning arrester, the number of lightning strikes, the partial discharge of the switchgear, and the operating parameters of the switchgear, a typical fault information database and a fault characteristic parameter system are established to effectively evaluate the operating status of the 27.5kVGIS switchgear (main specifications) of the substations along the railway and improve the safety of the railway power supply system.

[0045] 2. This high-precision digital online supervision system for high-speed railway switch cabinets uses electrical characteristic sensors to monitor the electrical characteristic parameters of the circuit breakers in the switch cabinet in real time and collect data. It uses mechanical characteristic sensors to monitor the opening / closing stroke, opening / closing time, and opening / closing coil current of the circuit breakers in the switch cabinet in real time and collect data. It uses lightning arrester sensors to monitor and collect lightning arrester operating parameters such as the total current and number of lightning strikes of the lightning arrester in real time; it uses partial discharge sensors to monitor and collect data on the partial discharge status in the cabinet in real time; and it uses temperature and humidity collectors to monitor and collect cabinet / environmental temperature and humidity data in real time; The collected data is comprehensively analyzed by the expert system, and a typical fault information database and fault characteristic parameter system are established to effectively evaluate the operating status of the 27.5kVGIS switchgear in the substations along the railway. The working characteristics of the main electrical components of the switchgear are monitored in real time under the operating status, combined with the expert system database fault model, to achieve early detection and early prevention of safety hazards. Through the user-side software and operation and maintenance APP, timely, efficient and controllable maintenance is achieved. Through the system and sensor devices, digital management of high-speed railway switchgear is realized, the inspection and maintenance efficiency of switchgear is improved, the operation and maintenance cost of the railway power supply system is reduced, and the safety of the railway power supply system is improved.

[0046] 3. The high-precision digital online supervision system for high-speed railway switchgear can realize real-time monitoring of various operating data of high-speed railway switchgear and realize real-time warning of abnormal status of switchgear; through the digital monitoring system of switchgear, the efficiency and effectiveness of railway power supply operation and maintenance management can be improved; while saving operation and maintenance costs, the quality of operation and maintenance can be improved; through real-time monitoring of various operating data of switchgear, uploading of monitoring data, and using expert system to collect and cross-compare the operating data in the switchgear / the same switchgear on the same trunk line / different switchgears on the same trunk line in different states and climate environments, the expert library data and algorithms in the digital monitoring system can be automatically improved to realize intelligent prediction of the switchgear status, thereby improving the safety of high-speed railway power grid operation and reducing the pressure of power grid operation and maintenance.

[0047] 4. The high-precision digital online supervision system for high-speed railway switchgear can give real-time alarms, add, modify and delete basic user information, define, add, delete and modify various parameters and monitoring points, and dynamically configure monitoring indicators. For designated terminals, various parameter default values ​​and alarm upper and lower limits can be set. The communication-related parameter inspection cycle, re-sampling times, database access parameter information, monitoring frequency, data refresh time, etc. can be defined by the user. The user defines the terminal interface and communication protocol: the terminal data collection cycle and storage cycle can be set. The system can autonomously learn according to the collected data, calibrate the standard comparison value, and the alarm threshold to achieve safer and more effective protection. The system can automatically generate operation and maintenance work orders, remotely dispatch operation and maintenance tasks, and automatically generate operation and maintenance record sheets, which can effectively improve the safe operation of rail transit, reduce operation and maintenance costs, and improve operation and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a system block diagram of the digital management of high-speed railway switch cabinet of the present invention;

[0049] Figure 2 It is a system architecture diagram of a single industrial computer of the present invention;

[0050] Figure 3 It is a system architecture diagram of a single switch cabinet of the present invention;

[0051] Figure 4 It is a structural schematic diagram of data transmission of the present invention;

[0052] Figure 5 It is a calculation formula diagram of the average value of the present invention;

[0053] Figure 6 It is a calculation formula diagram of the standard deviation of the present invention;

[0054] Figure 7 It is an algorithm formula diagram of anomaly detection of the present invention;

[0055] Figure 8It is the algorithm formula diagram of the mean value of the present invention;

[0056] Fig. 9 It is an algorithm formula diagram of the overall health status assessment model of the switch cabinet of the present invention;

[0057] Fig.10 It is an algorithm formula diagram of the expert system weight allocation of the present invention;

[0058] Fig.11 It is a comprehensive scoring algorithm formula diagram of the health status of the switch cabinet of the present invention;

[0059] Fig.12 It is a diagram of the algorithm formula predicted by the present invention;

[0060] Fig.13 It is a diagram of the algorithm formula for representing the proportion of samples correctly predicted by the model to the total samples in the present invention;

[0061] Fig.14 It is an algorithm formula diagram of precision, recall and F1-score in the present invention.

[0062] Figure 6 Where σ is the overall standard value; I i is each data point; μ is the population mean; N is the number of population data. DETAILED DESCRIPTION

[0063] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0064] Embodiment 1:

[0065] like Figure 1 As shown in the figure, a high-precision digital online supervision system for high-speed railway switchgear includes a switchgear online monitoring system, and the switchgear online monitoring system is electrically connected with a display terminal, a cloud platform, a data center, an expert system, a PC terminal software and a mobile terminal APP, an industrial computer, and an intelligent gateway;

[0066] The industrial computer cloud service is electrically connected to the data collector, and the data collector is electrically connected to the sensor and the built-in local storage;

[0067] The industrial computer is used to integrate and synchronize different types of sensor data to realize the collection and management of multi-sensor data. Through the fusion analysis of the collected data, the real-time monitoring of the overall status of the switch cabinet, abnormality detection and alarm are realized, including on-site sound and light alarm and remote terminal alarm through the system.

[0068] Embodiment 2:

[0069] A high-precision high-speed railway switch cabinet digital online supervision system, the display terminal is used to display the status of the switch cabinet through the gateway, the switch cabinet includes monitoring equipment, each monitoring equipment includes 1 mechanical characteristic sensor, 1 energy storage motor current sensor, 1 opening current sensor, 1 closing current sensor, 1 vibration sensor, 2 secondary CT current sensors, 1 two-in-one partial discharge sensor, 2 high-frequency current partial discharge sensors, 2 MOA current monitors, 1 circuit breaker electrical characteristic data collector, 1 high-frequency current partial discharge data collector, 1 lightning arrester data collector, 1 temperature and humidity data collector, an industrial computer, and a matching 12v / 24v power adapter. The mechanical characteristic sensor is used to monitor the mechanical movement state of the circuit breaker of the switch cabinet and evaluate the health state of the circuit breaker to determine whether the switch can be closed or opened normally;

[0070] The current sensors include energy storage motor current sensors, opening current sensors, and closing current sensors, which are used to measure the current size and change trend of the circuit breaker during operation, evaluate whether the current is within the normal range, and detect abnormal current conditions.

[0071] The secondary ct current sensor is used to monitor the synchronization of the three-phase currents A, B, and C during the operation of the switch cabinet circuit breaker;

[0072] Vibration sensors are used to identify vibrations caused by mechanical failures or abnormal movements, so as to detect potential mechanical problems in a timely manner;

[0073] The two-in-one partial discharge sensor is used to detect the discharge activity at the cable insulator of the circuit breaker, and monitor and evaluate the aging, defects and damage status of the circuit breaker insulation material in real time;

[0074] High-frequency current partial discharge sensors are used to detect high-frequency currents in the cable shielding and armoring layers, reflecting the insulation status of the cable so as to detect partial discharge phenomena early, prevent insulation failures and prevent electrical accidents.

[0075] MOA current sensor is used to detect the leakage current of zinc oxide arrester in real time and evaluate the insulation and health status of arrester;

[0076] The temperature and humidity sensor is used to monitor the temperature and humidity environment inside the cabinet to prevent damage to electrical equipment caused by excessive humidity or low temperature. At the same time, it monitors abnormal temperature and humidity data in the switch cabinet room to assist in determining whether there is abnormal operation of the section switch cabinet.

[0077] Embodiment three:

[0078] A high-precision digital online supervision system for high-speed railway switchgear. The industrial computer includes local historical data storage and analysis. By storing sensor data in the local database, it provides data support for subsequent analysis of local equipment operation trends, aging assessment, fault analysis and equipment health management, helps local equipment to perform predictive maintenance and optimize maintenance plans, thereby improving system reliability and stability.

[0079] Remote monitoring and system integration: data is transmitted to the cloud through the network to achieve remote monitoring and remote management. It supports integration with the substation's SCADA, EMS system, and the user's own management system, improves the information and intelligent management capabilities of the power system, and includes time information and monitoring point ID information.

[0080] Each device in the system has a unique ID code, and can be customized. The coding must follow the following principles:

[0081] Each high-speed rail power supply line has a unique ID, which must be reflected in the coding;

[0082] Each switch cabinet room has a unique ID, which must be reflected in the coding;

[0083] Each switch cabinet has a unique ID, which must be reflected in the coding;

[0084] Each sensor in each switch cabinet is classified by function, and each sensor category must be reflected in the coding;

[0085] Each sensor has a unique ID, which must be reflected in the encoding;

[0086] In the same high-speed railway power supply line, switch cabinets of different voltage / power levels have unique IDs, which need to be reflected in the coding;

[0087] In the same high-speed railway power supply line, switch cabinets of the same voltage / power level, different brands, and different models have unique IDs, which need to be reflected in the coding;

[0088] Different high-speed rail power supply lines in the same power grid have unique IDs, which need to be reflected in the coding;

[0089] The sensor ID code must contain the corresponding ID information of the switch cabinet under test; the sensor ID code must be as simple and easy to understand as possible, and can be easily matched with the position of the switch cabinet under test during use;

[0090] The human-machine interface is equipped with a graphical HMI. The operator can locally view the data of each sensor and system status information. It also supports local parameter setting, historical data query, alarm log viewing, etc. to achieve local intuitive and convenient equipment management and maintenance.

[0091] The industrial computer also includes a high-frequency current partial discharge data collector, a lightning arrester data collector, and an environmental temperature and humidity data collector. The installation is achieved by installing a standard rail at the double-row bracket on the right side of the secondary instrument room of the switch cabinet, and then fixing the collector housing on the rail with a rail buckle.

[0092] The data collector pre-processes and stores sensor data; the memory adopts a power-off storage design; the collector has its own clock system, and the uploaded monitoring results include: displacement, current, vibration, partial discharge, temperature and humidity data information, time information, and monitoring point ID information collected by the corresponding sensor;

[0093] The mobile terminal APP is used to view authorized monitoring data and accept the operation and maintenance management of the data center, and upload the operation and maintenance implementation information to the cloud platform.

[0094] The intelligent gateway is used to upload the data collected by the data collector to the cloud platform in real time;

[0095] The cloud platform is used to store data collected by all sensors, analyze and display the data, and issue warnings for abnormal data;

[0096] The data center includes client PC software for real-time viewing of monitoring results, abnormal alarms, alarm processing progress, alarm point processing status and process records, and operation and maintenance management;

[0097] It includes an expert system, which collects and analyzes monitoring data and abnormal processing record data based on the expert system, optimizes the normal and abnormal data judgment and warning thresholds through self-learning, and provides more accurate accident prediction and operation and maintenance suggestions;

[0098] Includes computer host, monitor and supporting hardware equipment.

[0099] When the data collector is transmitting data, the sensor will first save all the collected data locally in the corresponding data collector and upload it to the corresponding industrial computer through the R485 line. The industrial computer will upload it to the cloud platform through the gateway. Users can view and download the monitoring results in real time through the client software.

[0100] The data collector collects data from each sensor in the switch cabinet at a frequency of 1 minute / time and saves the data. The maximum storage capacity is 1 month.

[0101] The client software will automatically query the time information of the data collected by all sensors at an interval, where the time can be customized according to the actual situation. If data loss is found, it will automatically send a data re-send instruction to the corresponding industrial computer, and the corresponding industrial computer will re-upload the data of the missing time period to the cloud platform to ensure data integrity. Specific embodiment four:

[0103] Early warning strategy:

[0104] For each sensor, set key monitoring indicators and set alarm thresholds for each indicator.

[0105] 1) Displacement sensor:

[0106] The data collector compares the collected opening and closing waveforms with the normal working waveforms provided by the manufacturer, and alarms when the error exceeds a certain range (the error range is defined according to the actual working conditions and circuit breaker model).

[0107] 2) Hall current sensor:

[0108] The maximum current threshold and the minimum current threshold are set according to the circuit breaker model and the actual working conditions. When the collected current data exceeds the set threshold, an overload or load loss alarm is generated.

[0109] set up:

[0110] Amax: maximum allowed current;

[0111] Amin: maximum allowed current;

[0112] A: the current value collected;

[0113] Overload alarm when A≥Amax;

[0114] When A≤Amin, a load loss alarm is triggered;

[0115] The system software generates a current waveform of the opening and closing process, which is compared and analyzed with the original operating waveform provided by the circuit breaker manufacturer. When the waveform error value exceeds a certain range, an alarm is triggered.

[0116] When the manufacturer of a certain type of switch cabinet circuit breaker does not provide the original displacement waveform of the corresponding circuit breaker opening and closing, and the allowable working current range or waveform, the expert system averages the collected working displacement / current data of the same type of short circuit breaker, and adds a certain margin as the alarm threshold. The displacement / current waveform comparison chart is made based on the average value collected within a certain period of time.

[0117] Take the current as an example, that is: list all the working currents A1, A2, A3, ..., An, assuming there are n current values ​​in total;

[0118] Calculate the sum of all current values: Total = A1 + A2 + A3 + ... + An;

[0119] Calculate the mean: mean = total / N (number);

[0120] In the process of averaging, when the collected single circuit breaker operating current value exceeds a certain upper / lower limit, the value needs to be eliminated in the calculation. The reference value is corrected in real time according to the expert system.

[0121] 3) Vibration sensor: Set the normal vibration range and trigger an alarm if it exceeds the range. Since there is no original vibration data, baseline vibration data collection is performed on circuit breakers of the same model when the equipment is operating normally. These data are collected over multiple time periods to obtain vibration levels under different working conditions. The collected vibration data include vibration amplitude, frequency and acceleration. Statistical analysis is performed on the collected data to calculate the average vibration level and standard deviation under normal working conditions, identify the normal vibration range, and define the alarm threshold by calculating the upper and lower limits of the normal value.

[0122] Static threshold: set to the average value of normal vibration plus a multiple of standard deviation (determined according to actual working conditions); static threshold = average value + 1 × standard deviation; dynamic threshold: when the operating status of the equipment changes, the dynamic threshold is used;

[0123] For example, under high load or high temperature conditions, vibration may increase, and the alarm threshold = static threshold + 2x standard deviation. The standard deviation is calculated by the expert system;

[0124] Example:

[0125] Assume that under normal working conditions, the average value of vibration data of a certain device is 5mm / s and the standard deviation is 1mm / s. The following thresholds can be set: static threshold: vibration>6mm / s (average value + 1×standard deviation); dynamic threshold: vibration>7mm / s (average value + 2×standard deviation)

[0126] 4) Two-in-one partial discharge sensor: Set the circuit breaker discharge intensity threshold, and trigger the alarm when the threshold is exceeded. The initial threshold is set to the average value of normal partial discharge plus a multiple of the standard deviation (determined according to actual working conditions), and is adjusted later according to expert system data;

[0127] Assume normal partial discharge value: <100pC; when the collected data >100pC, an alarm is triggered.

[0128] 5) High-frequency current partial discharge sensor: Set the discharge intensity threshold of the cable shielding layer and armor layer. If the threshold is exceeded, an alarm will be triggered. The initial threshold is set to the average value of normal partial discharge plus a multiple of the standard deviation. It will be adjusted later based on the data obtained from expert system analysis.

[0129] Assume that the calculated normal PD average value + standard deviation is: <500pC. When the collected data is >500pC, an alarm is triggered.

[0130] 6) MOA current monitoring sensor: set a maximum allowable leakage current value Amax and a maximum allowable leakage current fluctuation range Aa (Aa is an interval value, for example: 2≤Aa≤10). When the collected current data A≥Amax, an excessive leakage alarm is triggered; when the collected data A≥Aa, an abnormal leakage alarm is triggered. The initial thresholds of Amax and Aa are set to the normal leakage current average value plus a multiple of the standard deviation (determined according to the actual working conditions), and are adjusted later based on the data analyzed by the expert system.

[0131] 7) Temperature and humidity sensor: Set high temperature, high humidity and low temperature, low humidity thresholds. Alarm when the temperature exceeds the threshold. The initial threshold is set to the average value of the temperature under normal working conditions plus a multiple of the standard deviation, and then adjusted according to the data calculated by the expert system.

[0132] For example: In spring / autumn (when the average temperature for 30 days is 20-25℃), the normal temperature range during normal operation is: -10℃ to 60℃; the high temperature alarm threshold: >60℃, triggering the high temperature alarm; the low temperature alarm threshold: <-10℃, triggering the low temperature alarm.

[0133] 8) Prediction of the overall health status of the switch cabinet: After comprehensive comparison and analysis of the collected data through the expert system, a comprehensive assessment of the overall status of the switch cabinet is conducted. When the comprehensive assessment value is lower than the set value, an alarm for poor health status is issued. Specific embodiment five:

[0135] Gradual alarm: set different levels of alarm thresholds according to the vibration amplitude;

[0136] Includes informational alarms: prompting changes in motion status, determined by the mean and standard deviation of alarm data measured at different motion states;

[0137] Warning level: slightly beyond the normal range, requiring attention (such as mean + 1 × standard deviation);

[0138] Severe level: Clearly beyond the normal range and requires immediate attention (such as mean + 2 × standard deviation).

[0139] The average value calculation formula is as follows Figure 5 As shown;

[0140] The standard deviation calculation formula is as follows Figure 6 As shown, when calculating the mean and standard deviation, the same type of switch cabinet in different positions in the same functional circuit loop, the same measured position, and the same sensor are used. Within a unit time, the time can be defined according to the actual working conditions, and the collected data is used as the alarm threshold in the corresponding early warning strategy. The later alarm data threshold is automatically corrected by the expert system after comparing and analyzing all the collected data.

[0141] Verification and adjustment of alarm thresholds: During the implementation process, monitor the actual alarm situation, conduct cross-comparison analysis based on the data collected by the expert system, and regularly revise the alarm thresholds to ensure the rationality, accuracy, and safety of the alarm thresholds.

[0142] Alarm method:

[0143] Sound and light alarm: alarm through virtual buzzer and indicator light designed by system software; realized on site through industrial computer and in data center through central PC;

[0144] Phone / SMS / Email Notification: In case of serious alarm, you can directly call the person in charge, or send a text message or email to the person in charge to ensure timely response (the customer sets the relevant notification form, authority, and designated person in charge according to the actual working conditions and operation and maintenance management status);

[0145] HMI interface prompt: An alarm window pops up on the operation interface. If the specific fault information can be clearly displayed, the specific fault information will be displayed. If the specific fault information cannot be clearly displayed, the expert system will predict the fault and give the possible cause, weight, recommended treatment method and other information of the evaluated fault information.

[0146] Alarm (abnormal) processing flow:

[0147] 1. Alarm triggering: When the system detects an abnormal situation, it immediately triggers the corresponding alarm mechanism;

[0148] 2. Information recording: record the alarm information in the system log for subsequent analysis and tracing;

[0149] 3. Response measures: According to the alarm level, the system automatically executes response measures (customized by the customer based on actual conditions).

[0150] For example: severe alarms - remotely shut down equipment to prevent accidental damage; warning alarms - prompt operators to conduct timely re-inspections and take manual intervention when necessary; the system's built-in map can accurately provide fault location information to assist maintenance personnel in quickly troubleshooting.

[0151] Regular evaluation and optimization of alarm strategies: The expert system automatically and regularly (user-defined) evaluates the effectiveness of the alarm strategy based on the collected sensor data, alarm information, and alarm processing feedback information, adjusts the alarm thresholds, and optimizes the alarm strategy (the expert system makes optimization strategy recommendations based on data analysis of all collected operation and maintenance information, and authorized operation and maintenance personnel confirm the alarm strategy changes) to adapt to the actual operation of the equipment.

[0152] Equipment maintenance strategy optimization: collect and analyze alarm data, identify common failure modes, and optimize equipment maintenance strategies. Operation and maintenance data include: 1. Alarm processing records (including alarm level, alarm processing process, and alarm processing result records); 2. Routine maintenance data records (including time, location, personnel, maintenance items, and maintenance process); 3. Abnormal discovery process, abnormal operation and maintenance progress, and abnormal occurrence time data records; 4. Abnormal operation and maintenance accident determination results, maintenance person in charge records; 5. Abnormal operation and maintenance processing process (processing method) records.

[0153] Expert system: All data are uploaded to the expert system for cross-comparison processing, which automatically corrects the relevant warning data in the warning strategy, and makes accident predictions based on the accumulated analysis of maintenance data.

[0154] The data analysis includes:

[0155] Analysis of operating data collected by the same sensor; cross-comparison analysis of data collected by different sensors of the same switchgear; comparison and analysis of data collected by the same sensors corresponding to the same model switchgear of the same functional circuit in the same high-speed railway power supply line; comparison and analysis of data collected by the same sensors corresponding to the same model switchgear of the same functional circuit in different high-speed railway power supply lines; comparison and analysis of data collected by the same sensors corresponding to the same model switchgear of different functional circuits in the same high-speed railway power supply line; comparison and analysis of data collected by the same sensors corresponding to the same model switchgear of different functional circuits in different high-speed railway power supply lines; comparison and analysis of data of switchgear of different brands with the same function in the same functional circuit of the same high-speed railway power supply line; comparison and analysis of data of switchgear of different brands with the same function in the same functional circuit of different high-speed railway power supply lines; comparison and analysis of data of switchgear of different brands with the same function in different functional circuits of the same high-speed railway line; comparison and analysis of data of switchgear of different brands with the same function in different functional circuits of different high-speed railway lines; comparison and analysis of data of different service years of the same switchgear.

[0156] Data comparison and analysis of switchgears of the same brand and model but different service years:

[0157] The implementation steps include:

[0158] Data collection: record the mechanical characteristics of the circuit breaker, the electrical characteristics of the circuit breaker, the partial discharge data of the circuit breaker, the vibration data, the partial discharge data of the insulation layer and armor layer of the switchgear, and the ambient temperature and humidity data of each switchgear in different environments and motion states.

[0159] Data comparison and analysis: The expert system compares and analyzes the collected relevant data to obtain the working data of the same switchgear at different working times, in different environments, and under different working conditions, as well as the effects of these factors on the mechanical structure stability, electrical performance stability, mechanical fatigue, material aging, component damage, electrical insulation changes, mechanical failure factors, electrical failure factors, leakage, etc. during the use of the switchgear, change trends, and abnormal situation data. Specific embodiment 6:

[0161] In data cleaning, outlier detection uses Z-score to identify and remove outliers that are beyond the normal range. The algorithm formula is as follows: Figure 7 ;

[0162] Missing value processing: Use the mean to fill the rows with missing values. The mean algorithm formula is as follows Figure 8 ;

[0163] Data standardization: Normalization is used to normalize the data to the range of [0,1] to facilitate comparison between different data types. The formula is:

[0164] Displacement data: directly use the maximum unique value and calculate the displacement change rate, formula:

[0165] Current data:

[0166] Statistical characteristics: including maximum value, minimum value, average value, and standard deviation;

[0167] Fluctuation amplitude: Calculate the range of current change, reflecting the stability of current.

[0168] Vibration data:

[0169] Vibration frequency and amplitude: Extract frequency and amplitude features to assess mechanical condition.

[0170] Partial discharge data:

[0171] Partial discharge amplitude: record the maximum value and frequency of partial discharge;

[0172] Partial Discharge Frequency: Monitors the frequency of partial discharge events.

[0173] Temperature and humidity data:

[0174] Temperature change rate: record the temperature change trend;

[0175] Humidity change rate: monitors changes in humidity.

[0176] Switchgear overall health status assessment model:

[0177] State scoring: For each data type, define a scoring function. Taking a current data as an example, the scoring formula is: Fig. 9 As shown, similarly, scoring formulas are defined for displacement, vibration, partial discharge, temperature and humidity data.

[0178] Weight allocation (Expert system weight allocation algorithm)

[0179] Based on the analytic hierarchy process (AHP), the implementation and algorithm formula are as follows Fig.10 shown.

[0180] The comprehensive score of switchgear health status is as follows: Fig.11 As shown;

[0181] Expert system for switch cabinet fault prediction, implementation method and some algorithm formulas

[0182] Fault prediction using random forest algorithm

[0183] Build a random forest model:

[0184] Determine parameters: select the number of decision trees (usually 100-500), the maximum depth of each tree, and the ratio of feature selection.

[0185] Constructing decision trees: Each tree is trained using different Bootstrap samples. At each node split, m features (the square root of the total number of features) are randomly selected from the feature set for splitting.

[0186] Prediction and evaluation:

[0187] Prediction: Input the new sample data (data collected by different sensors in the same switch cabinet) into the random forest model, and after voting by all decision trees, the final prediction result is obtained, such as Fig.12 shown.

[0188] Evaluation: The expert system prediction model is evaluated using the test set. Evaluation indicators include accuracy, precision, recall, and F1-score.

[0189] Training set: used to train the model (historical data set collected by sensors corresponding to the same type of switch cabinet in the same functional circuit of the same high-speed railway line).

[0190] Validation set: used for parameter adjustment and model selection (maintenance data, alarm processing records).

[0191] Test set: used to evaluate the performance of the model and ensure that the model can perform well on unseen data. (Test data is randomly selected from the same model of switchgear with the same functional circuit on different high-speed rail lines)

[0192] The division ratio is 70% training set, 15% validation set, and 15% test set.

[0193] The model evaluation algorithm formula is as follows Figure 13-14 shown.

[0194] Data application: After analyzing and comparing the collected information, the expert system automatically corrects the corresponding parameter warning data of the single data in the warning strategy based on regression analysis calculation; the overall health status of the switch cabinet is evaluated according to the expert system weight distribution method; the switch cabinet is fault-predicted through the random forest algorithm; the health and fault assessment model in the expert system is verified and optimized using the test set; the potential problems of the switch cabinet system in the high-speed rail power supply system are discovered in time through the whole system, and the safety and reliability of the switch cabinet equipment operation are improved. Improve the system operation and maintenance efficiency and reduce the operation and maintenance cost, and gradually realize precise operation and maintenance.

[0195] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A high-precision high-speed railway switchgear digital online supervision system, including a switchgear online monitoring system, characterized in that: The switch cabinet online monitoring system is electrically connected to a display terminal, a cloud platform, a data center, an expert system, a PC terminal software and a mobile terminal APP, an industrial computer, and an intelligent gateway; The industrial computer cloud service is electrically connected to the data collector, and the data collector is electrically connected to the sensor and the built-in local storage; The industrial computer is used to integrate and synchronize different types of sensor data to realize the collection and management of multi-sensor data. Through the fusion analysis of the collected data, real-time monitoring of the overall status of the switch cabinet, abnormality detection and alarm are realized, including on-site sound and light alarm and remote terminal alarm through the system.

2. According to claim 1, a high-precision high-speed railway switch cabinet digital online monitoring system is characterized by: The display terminal is used to display the status of the switch cabinet through the gateway. The switch cabinet includes monitoring equipment, each of which includes 1 mechanical property sensor, 1 energy storage motor current sensor, 1 opening current sensor, 1 closing current sensor, 1 vibration sensor, 2 secondary CT current sensors, 1 two-in-one partial discharge sensor, 2 high-frequency current partial discharge sensors, 2 MOA current monitors, 1 circuit breaker electrical characteristic data collector, 1 high-frequency current partial discharge data collector, 1 lightning arrester data collector, 1 temperature and humidity data collector, an industrial computer, and a matching 12v / 24v power adapter.

3. According to claim 2, a high-precision high-speed railway switch cabinet digital online monitoring system is characterized by: The mechanical characteristic sensor is used to monitor the mechanical motion state of the circuit breaker of the switch cabinet and evaluate the health state of the circuit breaker to determine whether the switch can be closed or opened normally; The current sensors include energy storage motor current sensors, opening current sensors, and closing current sensors, which are used to measure the current size and change trend of the circuit breaker during operation, evaluate whether the current is within the normal range, and detect abnormal current conditions.

4. According to claim 2, a high-precision high-speed railway switch cabinet digital online monitoring system is characterized by: The secondary ct current sensor is used to monitor the synchronization of the three-phase currents A, B, and C during the operation of the switch cabinet circuit breaker; The vibration sensor is used to identify vibrations caused by mechanical failure or abnormal movement, so as to timely discover potential mechanical problems; The two-in-one partial discharge sensor is used to detect discharge activities at the cable insulator of the circuit breaker, and monitor and evaluate the aging, defects and damage status of the circuit breaker insulation material in real time; The high-frequency current partial discharge sensor is used to detect high-frequency current in the cable shielding layer and armor layer, and reflect the insulation state of the cable, so as to detect partial discharge phenomenon early, prevent insulation failure and prevent electrical accidents.

5. According to claim 2, a high-precision high-speed railway switch cabinet digital online monitoring system is characterized by: The MOA current sensor is used to detect the leakage current of the zinc oxide arrester in real time and evaluate the insulation and health status of the arrester; The temperature and humidity sensor is used to monitor the temperature and humidity environment inside the cabinet to prevent damage to electrical equipment caused by excessively high humidity or low temperature. At the same time, it monitors abnormal temperature and humidity data inside the switch cabinet room to assist in determining whether there is abnormal operation of the section switch cabinet.

6. According to claim 1, a high-precision high-speed railway switch cabinet digital online monitoring system is characterized by: The industrial computer includes local historical data storage and analysis. By storing sensor data in a local database, it provides data support for subsequent analysis of local equipment operation trends, aging assessment, fault analysis and equipment health management, helps local equipment perform predictive maintenance, optimizes maintenance plans, and thus improves system reliability and stability.

7. According to claim 1, a high-precision high-speed railway switch cabinet digital online monitoring system is characterized by: The remote monitoring and system integration transmits data to the cloud through the network to achieve remote monitoring and remote management, support integration with the substation's SCADA, EMS system, and the user's own management system, improve the information and intelligent management capabilities of the power system, and include time information and monitoring point ID information; Each device in the system has a unique ID code, and can be customized. The coding must follow the following principles: Each high-speed rail power supply line has a unique ID, which must be reflected in the coding; Each switch cabinet room has a unique ID, which must be reflected in the coding; Each switch cabinet has a unique ID, which must be reflected in the coding; Each sensor in each switch cabinet is classified by function, and each sensor category must be reflected in the coding; Each sensor has a unique ID, which must be reflected in the encoding; In the same high-speed railway power supply line, switch cabinets of different voltage / power levels have unique IDs, which need to be reflected in the coding; In the same high-speed railway power supply line, switch cabinets of the same voltage / power level, different brands, and different models have unique IDs, which need to be reflected in the coding; Different high-speed rail power supply lines in the same power grid have unique IDs, which need to be reflected in the coding; The sensor ID code must contain the corresponding ID information of the switch cabinet under test; the sensor ID code must be as simple and easy to understand as possible, and can be easily matched with the position of the switch cabinet under test during use; The human-machine interface is equipped with a graphical HMI. The operator can locally view the data of each sensor and system status information. It also supports local parameter setting, historical data query, alarm log viewing, etc. to achieve local intuitive and convenient equipment management and maintenance.

8. According to claim 1, a high-precision high-speed railway switch cabinet digital online monitoring system is characterized by: The industrial computer also includes a high-frequency current partial discharge data collector, a lightning arrester data collector, and an environmental temperature and humidity data collector, which are installed by installing a standard rail at the double-row bracket on the right side of the secondary instrument room of the switch cabinet, and then fixing the collector housing on the rail with a rail buckle to achieve installation; The data collector performs preprocessing and data storage on the sensor data; the memory adopts a power-off storage design; The collector has its own clock system. The uploaded monitoring results include: displacement, current, vibration, partial discharge, temperature and humidity data information, time information, and monitoring point ID information collected by the corresponding sensor; The mobile terminal APP is used to view authorized monitoring data and accept the operation and maintenance management of the data center, and upload the operation and maintenance implementation information to the cloud platform.

9. A high-precision high-speed railway switch cabinet digital online monitoring system according to claim 1, characterized in that: The intelligent gateway is used to upload the data collected by the data collector to the cloud platform in real time; The cloud platform is used to store data collected by all sensors, analyze and display the data, and issue warnings for abnormal data; The data center includes client PC software for real-time viewing of monitoring results, abnormal alarms, alarm processing progress, alarm point processing status and process records, and operation and maintenance management; It includes an expert system, which collects and analyzes monitoring data and abnormal processing record data based on the expert system, optimizes the normal and abnormal data judgment and warning thresholds through self-learning, and provides more accurate accident prediction and operation and maintenance suggestions; Includes computer host, monitor and supporting hardware equipment.

10. A high-precision high-speed railway switch cabinet digital online monitoring system according to claim 1, characterized in that: When the data collector transmits data, the sensor will first save all the collected data locally in the corresponding data collector and upload it to the corresponding industrial computer through the R485 line. The industrial computer uploads it to the cloud platform through the gateway, and the user can view and download the monitoring results in real time through the client software; The data collector collects data from each sensor in the switch cabinet at a frequency of 1 minute per time and saves the data. The maximum storage capacity is 1 month. The client software will automatically query the time information of the data collected by all sensors at an interval, where the time can be customized according to the actual situation. If data loss is found, it will automatically send a data re-send instruction to the corresponding industrial computer, and the corresponding industrial computer will re-upload the data of the missing time period to the cloud platform to ensure data integrity.

Images