Nuclear power medium-voltage grounding monitoring system based on artificial intelligence

Abstract

The invention relates to the technical field of nuclear power medium-voltage grounding monitoring, and discloses a nuclear power medium-voltage grounding monitoring system based on artificial intelligence, which comprises multi-level processing of data acquisition, processing, calculation and analysis, and realizes real-time monitoring and intelligent early warning of a grounding system. Compared with a traditional manual judgment or simple threshold alarm mode, the system optimizes an anomaly detection process through an artificial intelligence algorithm, can identify and position fault nodes more accurately, and remarkably improves the fault response speed and accuracy, in addition, a feedback module enhances the usability of the system through a visual report, and the fault detection accuracy is improved. According to the method, operation and maintenance personnel can efficiently process abnormal conditions, the abnormal conditions are divided into three levels, and threshold values suitable for different abnormal conditions are set, so that the system can identify changes of grounding electrical parameters, environmental parameters and power grid operation parameters in real time, and the severity of the abnormal conditions is judged according to the change degree; and a hierarchical response mechanism optimizes the processing capability of the monitoring system to the exception.

Description

Technical Field

[0001] The present invention relates to the technical field of nuclear power medium-voltage grounding monitoring, and in particular to a nuclear power medium-voltage grounding monitoring system based on artificial intelligence. Background Art

[0002] As a clean and efficient form of energy, nuclear energy has gradually become an indispensable part of the global energy structure. Nuclear power plants can provide long-term continuous power output through their stable power generation capacity, which is one of the key guarantees for power supply. In the electrical system of a nuclear power plant, the medium-voltage grounding system, as an important part of ensuring power safety, plays a vital role in the grounding safety of power equipment. Through real-time monitoring of key electrical parameters such as grounding current, grounding voltage and grounding resistance, the medium-voltage grounding system ensures the normal operation of nuclear power plants and equipment safety.

[0003] Traditional grounding monitoring methods usually rely on fixed thresholds and simple alarm mechanisms, lacking flexibility and intelligent processing capabilities. Especially when multiple factors are intertwined in the system, traditional monitoring methods are difficult to analyze and comprehensively judge this information in real time. Current technologies mostly rely on manual detection and simple outlier judgment, resulting in slow system response and inability to effectively diagnose complex fault modes. In addition, due to the high correlation between monitoring parameters, traditional methods are difficult to conduct comprehensive analysis and may ignore potential safety hazards, thus failing to provide accurate early warning information and effective fault location for nuclear power plants.

[0004] Therefore, we proposed an artificial intelligence-based nuclear power medium-voltage grounding monitoring system to solve the above-mentioned problems. Summary of the invention

[0005] The purpose of the present invention is to provide a nuclear power medium voltage grounding monitoring system based on artificial intelligence to solve the problem that the traditional grounding monitoring means proposed in the above background technology usually rely on fixed thresholds and simple alarm mechanisms, lacking flexibility and intelligent processing capabilities. In particular, when multiple factors are intertwined in the system, the traditional monitoring method is difficult to analyze and comprehensively judge this information in real time.

[0006] To achieve the above object, the present invention provides the following technical solutions: a nuclear power medium voltage grounding monitoring system based on artificial intelligence, comprising a data acquisition module, a data processing module, a first data calculation module, a first data analysis module, a second data calculation module, a second data analysis module and a feedback module;

[0007] The data acquisition module is used to collect nuclear power medium voltage grounding data and form a multi-source data set;

[0008] The data processing module is used to pre-process the multi-source data sets and reorganize them into a first data set, a second data set and a third data set;

[0009] The first data calculation module is used to integrate and calculate the first data set, the second data set and the third data set respectively, so as to generate a first reference coefficient S1, a second reference coefficient S2 and a third reference coefficient S3, and integrate and calculate the first reference coefficient S1, the second reference coefficient S2 and the third reference coefficient S3, so as to generate an abnormal alarm coefficient YCX;

[0010] The first data analysis module is used to compare the abnormal alarm coefficient YCX with the preset first threshold value Y, thereby generating a first comparison result, and judging whether the nuclear power medium voltage grounding data is abnormal according to the first comparison result;

[0011] If the first comparison result shows that the data is abnormal, the second data calculation module calculates the first data set, the second data set and the third data set again to generate the first node alarm coefficient BJ1, the second node alarm coefficient BJ2 ​​and the third node alarm coefficient BJ3;

[0012] The second data analysis module is used to compare the first node alarm coefficient BJ1, the second node alarm coefficient BJ2 ​​and the third node alarm coefficient BJ3 with the preset first node alarm threshold R1, the second node alarm threshold R2 and the third node alarm threshold R3 respectively, identify them through artificial intelligence, thereby generating a second comparison result, and judging the specific abnormal node according to the second comparison result;

[0013] The feedback module is used to send the first comparison result and the second comparison result to the terminal and generate a visual report.

[0014] Preferably, the data acquisition module includes a first acquisition unit, a second acquisition unit and a third acquisition unit;

[0015] The first acquisition unit obtains grounding electrical parameters including grounding current, grounding voltage and grounding resistance through sensors, and obtains standard values ​​of grounding current, grounding voltage and grounding resistance through a cloud platform;

[0016] The second acquisition unit obtains environmental parameters including soil temperature, soil moisture and soil salinity through sensors, and obtains standard values ​​of soil temperature, soil moisture and soil salinity of local soil through the cloud platform;

[0017] The third acquisition unit obtains grid operation parameters including grid frequency, power load and grid harmonics through sensors, and obtains standard values ​​of grid frequency, power load and grid harmonics through the cloud platform;

[0018] The ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load and grid harmonics constitute the multi-source data set.

[0019] Preferably, the data processing module includes a data preprocessing unit and a data integration unit, wherein the data preprocessing unit is used to preprocess and dimensionlessly transform the multi-source data set, and the data integration unit is used to rearrange the processed multi-source data set to generate the first data set, the second data set, and the third data set;

[0020] The first data set includes ground current, ground voltage, and ground resistance;

[0021] The ground current includes a ground current standard value BZA and a ground current real-time value, wherein the ground current real-time value is recorded as A1, A2, A3, ..., An according to the timestamp;

[0022] The ground voltage includes a ground voltage standard value BZB and a ground voltage real-time value, wherein the ground voltage real-time values ​​are recorded as B1, B2, B3, ..., Bn according to timestamps;

[0023] The grounding resistance includes a grounding resistance standard value BZC and a grounding resistance real-time value, wherein the grounding resistance real-time values ​​are recorded as C1, C2, C3, ..., Cn according to timestamps;

[0024] The second data set includes soil temperature, soil moisture, and soil salinity;

[0025] The soil temperature includes a soil temperature standard value BZD and a soil temperature real-time value, wherein the soil temperature real-time values ​​are recorded as D1, D2, D3, ..., Dn according to timestamps;

[0026] The soil moisture includes a soil moisture standard value BZE and a soil moisture real-time value, wherein the soil moisture real-time value is recorded as E1, E2, E3, ..., En according to the timestamp;

[0027] The soil salinity includes a soil salinity standard value BZF and a soil salinity real-time value, wherein the soil salinity real-time value is recorded as F1, F2, F3, ..., Fn according to the timestamp;

[0028] The third data set includes grid frequency, power load, and grid harmonics;

[0029] The grid frequency includes a grid frequency standard value BZG and a grid frequency real-time value, wherein the grid frequency real-time values ​​are recorded as G1, G2, G3, ..., Gn according to timestamps;

[0030] The electric load includes an electric load standard value BZH and an electric load real-time value, wherein the electric load real-time value is recorded as H1, H2, H3, ..., Hn according to the timestamp;

[0031] The power grid harmonics include power grid harmonic standard values ​​BZI and power grid harmonic real-time values, wherein the power grid harmonic real-time values ​​are recorded as K1, K2, K3, . . . , Kn according to timestamps.

[0032] Preferably, the first data calculation module calculates and obtains the first reference coefficient S1, the second reference coefficient S2 and the third reference coefficient S3 respectively by the following formulas:

[0033] S1=a1×PA+a2×PB+a3×PC;

[0034] S2=b1×PD+b2×PE+b3×PF;

[0035] S3 = c1 × PG + c2 × PH + c3 × PK;

[0036]

[0037]

[0038] Where: BZA is the standard value of ground current, BZB is the standard value of ground voltage, BZC is the standard value of ground resistance, BZD is the standard value of soil temperature, BZE is the standard value of soil moisture, BZF is the standard value of soil salinity, BZG is the standard value of grid frequency, BZH is the standard value of power load, and BZK is the standard value of grid harmonics;

[0039] An, Bn, Cn, Dn, En, Fn, Gn, Hn, and Kn are the real-time values ​​of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load, and grid harmonics at the nth timestamp, respectively;

[0040] Ai, Bi, Ci, Di, Ei, Fi, Gi, Hi, and Ki are the real-time values ​​of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load, and grid harmonics at the i-th timestamp, respectively;

[0041] i=n-m+1, m is the window value, n is the total number of timestamps;

[0042] PA is the ground current intervention coefficient, PB is the ground voltage intervention coefficient, PC is the ground resistance intervention coefficient, PD is the soil temperature intervention coefficient, PE is the soil moisture intervention coefficient, PF is the soil salinity intervention coefficient, PG is the grid frequency intervention coefficient, PH is the power load intervention coefficient, and PK is the grid harmonic intervention coefficient.

[0043] Preferably, the first data calculation module calculates and obtains the abnormal alarm coefficient YCX by the following formula:

[0044] YCX=d1×S1+d2×S2+d3×S3;

[0045] Where: S1 is the first reference coefficient, S2 is the second reference coefficient, and S3 is the third reference coefficient;

[0046] d1, d2 and d3 are weight values, and the values ​​of d1, d2 and d3 are adjusted and set by the user.

[0047] Preferably, the first comparison result generated by the first data analysis module is specifically as follows:

[0048] When YCX≤Y, it means that there is no abnormality in the current nuclear power medium-voltage grounding. When YCX≤Y, it means that there is an abnormality in the current nuclear power medium-voltage grounding.

[0049] Preferably, the second data calculation module calculates and obtains the first node alarm coefficient BJ1, the second node alarm coefficient BJ2 ​​and the third node alarm coefficient BJ3 respectively through the following formulas:

[0050]

[0051] Where: An, Bn, Cn, Dn, En, Fn, Gn, Hn, Kn are the real-time values ​​of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load and grid harmonic at the nth timestamp respectively;

[0052] JA is the mean value of ground current, JB is the mean value of ground voltage, JC is the mean value of ground resistance, JD is the mean value of soil temperature, JE is the mean value of soil moisture, JF is the mean value of soil salinity, JG is the mean value of grid frequency, JH is the mean value of power load, and JK is the mean value of grid harmonics;

[0053] BA is the standard deviation of ground current, BB is the standard deviation of ground voltage, BC is the standard deviation of ground resistance, BD is the standard deviation of soil temperature, BE is the standard deviation of soil moisture, BF is the standard deviation of soil salinity, BG is the standard deviation of grid frequency, BH is the standard deviation of power load, and BK is the standard deviation of grid harmonics.

[0054] Preferably, the second data calculation module calculates and obtains the ground current mean JA, ground voltage mean JB, ground resistance mean JC, soil temperature mean JD, soil moisture mean JE, soil salinity mean JF, grid frequency mean JG, power load mean JH, and grid harmonic mean JK through the following formulas:

[0055]

[0056]

[0057] Where: Ao, Bo, Co, Do, Eo, Fo, Go, Ho and Ko are the real-time value of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load and grid harmonic of the oth node in the timestamp respectively, and n is the total number of timestamps.

[0058] Preferably, the second data calculation module calculates and obtains the ground current standard deviation BA, ground voltage standard deviation BB, ground resistance standard deviation BC, soil temperature standard deviation BD, soil moisture standard deviation BE, soil salinity standard deviation BF, grid frequency standard deviation BG, power load standard deviation BH, and grid harmonic standard deviation BK through the following formulas:

[0059]

[0060]

[0061] Where: Ao, Bo, Co, Do, Eo, Fo, Go, Ho and Ko are the real-time value of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load and grid harmonic of the oth node in the timestamp respectively, and n is the total number of timestamps;

[0062] JA is the mean value of ground current, JB is the mean value of ground voltage, JC is the mean value of ground resistance, JD is the mean value of soil temperature, JE is the mean value of soil moisture, JF is the mean value of soil salinity, JG is the mean value of grid frequency, JH is the mean value of power load, and JK is the mean value of grid harmonics;

[0063] n is the total number of timestamps.

[0064] Preferably, the second comparison result generated by the second data analysis module is specifically as follows:

[0065] When BJ1 ≤ R1, it represents a first-level abnormality in the grounding electrical parameters; when R1 < BJ1 ≤ R1 × 105%, it represents a second-level abnormality in the grounding electrical parameters; when BJ1 > R1 × 105%, it represents a third-level abnormality in the grounding electrical parameters.

[0066] When BJ2 ≤ R2, it represents a first-level abnormality in the environmental parameters; when R2 < BJ2 ≤ R2 × 105%, it represents a second-level abnormality in the environmental parameters; when BJ2 > R2 × 105%, it represents a third-level abnormality in the environmental parameters.

[0067] When BJ3 ≤ R3, it represents a first-level abnormality in the power grid operation parameters; when R3 < BJ3 ≤ R3 × 105%, it represents a second-level abnormality in the power grid operation parameters; when BJ3 > R3 × 105%, it represents a third-level abnormality in the power grid operation parameters.

[0068] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0069] 1. Through multi-level processing of data collection, processing, calculation, and analysis, this system realizes real-time monitoring and intelligent early warning of the grounding system. Compared with traditional manual judgment or simple threshold alarm methods, this system optimizes the abnormal detection process through artificial intelligence algorithms, can more accurately identify and locate fault nodes, significantly improves the fault response speed and accuracy. In addition, the feedback module enhances the usability of the system through visual reports, enabling maintenance personnel to efficiently handle abnormal situations.

[0070] 2. The real-time calculation of the abnormal alarm coefficient YCX and threshold comparison, combined with intelligent algorithms, enable the system to issue alarms in a timely manner when potential risks are detected. This intelligent abnormal detection can help operators identify possible faults or risks in the grounding system earlier, take measures in advance, and improve the safety and stability of nuclear power plants.

[0071] 3. By calculating YCX in real time and comparing it with the threshold Y, the system can continuously monitor the status of the grounding system. When an abnormality occurs, the system can quickly give feedback to ensure that the fault can be detected and processed in the first time. Based on this abnormal determination mechanism, the medium-voltage grounding monitoring system for nuclear power can automatically trigger an alarm when an abnormality occurs in the grounding system without manual intervention. This automatic function greatly improves the response speed, especially suitable for high-risk environments such as nuclear power. The automatic alarm system reduces the complexity of manual operations, ensures that corresponding measures can be taken in a timely manner when an abnormality occurs, and reduces potential safety hazards.

[0072] 4. By classifying abnormal situations into three levels and setting thresholds that adapt to different abnormal situations, the system can identify changes in grounding electrical parameters, environmental parameters, and grid operation parameters in real time, and judge the severity of the abnormality based on the degree of change. The hierarchical response mechanism optimizes the monitoring system's ability to handle abnormalities, and can take different measures based on the level of abnormality, thereby improving the flexibility and fault tolerance of the system. For minor abnormalities, the system can provide warnings without immediate intervention, while for serious abnormalities, the system can trigger an emergency response to avoid the expansion of risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a system flow chart of the present invention.

[0074] In the figure: 1. data acquisition module; 11. first acquisition unit; 12. second acquisition unit; 13. third acquisition unit; 2. data processing module; 21. data preprocessing unit; 22. data integration unit; 3. first data calculation module; 4. first data analysis module; 5. second data calculation module; 6. second data analysis module; 7. feedback module. DETAILED DESCRIPTION

[0075] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0076] Example 1: Please refer to Figure 1 , a nuclear power medium voltage grounding monitoring system based on artificial intelligence, comprising a data acquisition module 1, a data processing module 2, a first data calculation module 3, a first data analysis module 4, a second data calculation module 5, a second data analysis module 6 and a feedback module 7;

[0077] The data acquisition module 1 is used to collect nuclear power medium voltage grounding data and form a multi-source data set;

[0078] The data processing module 2 is used to pre-process the multi-source data sets and reorganize them into a first data set, a second data set and a third data set;

[0079] The first data calculation module 3 is used to integrate and calculate the first data set, the second data set and the third data set respectively, so as to generate the first reference coefficient S1, the second reference coefficient S2 and the third reference coefficient S3, and integrate and calculate the first reference coefficient S1, the second reference coefficient S2 and the third reference coefficient S3, so as to generate the abnormal alarm coefficient YCX;

[0080] The first data analysis module 4 is used to compare the abnormal alarm coefficient YCX with the preset first threshold value Y, so as to generate a first comparison result, and determine whether the nuclear power medium voltage grounding data is abnormal according to the first comparison result;

[0081] If the first comparison result shows that the data is abnormal, the second data calculation module 5 calculates the first data set, the second data set and the third data set again to generate the first node alarm coefficient BJ1, the second node alarm coefficient BJ2 ​​and the third node alarm coefficient BJ3;

[0082] The second data analysis module 6 is used to compare the first node alarm coefficient BJ1, the second node alarm coefficient BJ2 ​​and the third node alarm coefficient BJ3 with the preset first node alarm threshold R1, the second node alarm threshold R2 and the third node alarm threshold R3 respectively, identify them through artificial intelligence, thereby generating a second comparison result, and judging the specific abnormal node according to the second comparison result;

[0083] The feedback module 7 is used to send the first comparison result and the second comparison result to the terminal and generate a visual report.

[0084] In this embodiment: Data acquisition module 1 is responsible for real-time acquisition of nuclear power medium-voltage grounding related data from on-site sensors and equipment, and through multi-channel data acquisition, converts this information into a multi-source data set to fully reflect the operating status of the nuclear power medium-voltage grounding system. This module provides reliable and accurate raw data support for subsequent data processing and analysis. Through comprehensive monitoring, it ensures that the system can obtain the operating data of the nuclear power plant grounding system in real time and comprehensively, improving the monitoring accuracy and timeliness.

[0085] Data processing module 2 is responsible for preprocessing multi-source data sets to ensure data quality. Since the collected data may contain noise, errors or missing, the module normalizes the original data through data cleaning, denoising, smoothing and other operations. The processed data is reorganized into the first data set, the second data set and the third data set to provide more stable and consistent data support for subsequent calculations and analysis. The role of this module is to eliminate uncertainty in the data, ensure the reliability of system analysis, and provide high-quality input data for subsequent calculations and anomaly detection.

[0086] The first data calculation module 3 integrates and calculates the first, second and third data sets to generate three reference coefficients: the first reference coefficient S1, the second reference coefficient S2 and the third reference coefficient S3. These reference coefficients can respectively reflect the change of ground current, the stability of ground resistance and the influence of external environmental factors on the grounding system. By calculating these coefficients, the system can quantitatively evaluate the operating status of the grounding system and provide a scientific basis for subsequent anomaly detection. By integrating multi-dimensional data, this module realizes multi-angle quantitative analysis of the grounding system status and improves the accuracy and efficiency of anomaly detection.

[0087] The first data analysis module 4 compares the calculated abnormal alarm coefficient YCX with the preset first threshold value Y to determine whether the grounding data is abnormal. When the abnormal alarm coefficient YCX exceeds the set threshold value Y, the system will determine that the grounding system is abnormal and trigger an alarm. The function of this module is to achieve early warning of abnormal conditions of the grounding system. By comparing with the preset threshold value, the module can quickly and accurately identify the abnormality of the grounding system, issue an alarm in time, and help the operation and maintenance personnel take corresponding measures to avoid further deterioration of the accident.

[0088] When the first data analysis module 4 determines that there is an abnormality, the second data calculation module 5 further calculates the first, second and third data sets to generate three node alarm coefficients BJ1, BJ2 and BJ3. These node alarm coefficients can indicate the location and abnormality of the specific grounding node, which helps the system to quickly locate the source of the fault. The second data calculation module 5 can locate the specific problem of the node through further analysis. The detailed calculation and positioning functions of this module make fault detection more accurate and help improve the efficiency of maintenance and processing.

[0089] The second data analysis module 6 compares the node alarm coefficients BJ1, BJ2, and BJ3 with the preset node alarm thresholds R1, R2, and R3, and performs intelligent identification and analysis through artificial intelligence algorithms. Artificial intelligence technology is used in this module to identify and analyze complex abnormal patterns, and can automatically learn and identify potential abnormal behaviors of the system. With the support of intelligent algorithms, the system can accurately determine which nodes have abnormalities in more complex situations, and give the severity and possible causes of the faults. The application of artificial intelligence improves the intelligence level of the system, makes abnormality detection more accurate and flexible, reduces the need for manual intervention, and improves the intelligent management capabilities of the overall system.

[0090] The feedback module 7 feeds back the results obtained by the first data analysis module 4 and the second data analysis module 6 to the terminal and generates a visual report. These reports are presented in the form of charts, alarm information and detailed descriptions, which facilitates operation and maintenance personnel to quickly understand the operating status and potential problems of the grounding monitoring system. Through an intuitive interface and real-time updates, operators can promptly grasp the abnormal information and fault location of the nuclear power medium-voltage grounding system and respond quickly. This module not only improves the operability and transparency of the system, but also provides an intuitive decision-making basis for fault repair, further improving work efficiency and response speed.

[0091] Through multi-level processing of data collection, processing, calculation and analysis, this system realizes real-time monitoring and intelligent early warning of the grounding system. Compared with traditional manual judgment or simple threshold alarm methods, this system optimizes the anomaly detection process through artificial intelligence algorithms, and can more accurately identify and locate fault nodes, significantly improving the fault response speed and accuracy. In addition, the feedback module 7 enhances the ease of use of the system through visual reports, enabling operation and maintenance personnel to efficiently handle abnormal situations.

[0092] Example 2: Please refer to Figure 1 , the data acquisition module 1 includes a first acquisition unit 11, a second acquisition unit 12 and a third acquisition unit 13;

[0093] The first acquisition unit 11 obtains grounding electrical parameters including grounding current, grounding voltage and grounding resistance through sensors, and obtains standard values ​​of grounding current, grounding voltage and grounding resistance through a cloud platform;

[0094] The second acquisition unit 12 obtains environmental parameters including soil temperature, soil moisture and soil salinity through sensors, and obtains standard values ​​of soil temperature, soil moisture and soil salinity of local soil through the cloud platform;

[0095] The third acquisition unit 13 obtains the grid operation parameters including grid frequency, power load and grid harmonics through sensors, and obtains the standard values ​​of grid frequency, power load and grid harmonics through the cloud platform;

[0096] The ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load and grid harmonics constitute the multi-source data set.

[0097] In this embodiment: the first acquisition unit 11 monitors the grounding electrical parameters, including grounding current, grounding voltage and grounding resistance, in real time through sensors, and obtains the standard values ​​of grounding current, grounding voltage and grounding resistance through the cloud platform. By comparing with the standard values, it is ensured that the grounding electrical parameters are within the normal range, and potential grounding system faults can be effectively identified to ensure system safety.

[0098] The second acquisition unit 12 collects environmental parameters, including soil temperature, soil moisture and soil salinity, through sensors, and obtains standard values ​​under the soil conditions of the region through the cloud platform. These environmental parameters have an important impact on the change of grounding resistance. Real-time monitoring and comparison of these parameters with standard values ​​can timely discover the impact of external environmental changes on the grounding system, further improving the adaptability of the system.

[0099] The third acquisition unit 13 obtains the grid operation parameters, including grid frequency, power load and grid harmonics, through sensors, and obtains the corresponding standard values ​​through the cloud platform. Changes in grid parameters may affect the normal operation of the grounding system. Real-time monitoring of the grid operation status can ensure that the grounding system can still operate stably under grid fluctuations and harmonic interference.

[0100] The data acquisition module 1 obtains grounding electrical parameters, environmental parameters and grid operation parameters through three independent acquisition units, and comprehensively monitors the operation status of the nuclear power medium-voltage grounding system from multiple dimensions. This multi-source data acquisition method can ensure a comprehensive understanding of the working environment of the grounding system from different angles, thereby improving the accuracy and reliability of data monitoring. By combining grounding electrical parameters, environmental factors and grid operation status, the overall state of the system can be better evaluated, avoiding the limitation that a single parameter cannot fully reflect the problem. Each acquisition unit obtains the standard value through the cloud platform and compares the collected real-time data with the standard value. The advantage of this approach is that it can provide a real-time reference benchmark to help determine whether the collected data is within the normal operating range. If the collected electrical parameters, environmental parameters or grid operation parameters have a large deviation from the standard value, the system can identify potential problems in a timely manner and issue an alarm. This standard value comparison method effectively improves the accuracy of anomaly detection, avoids misjudgment and missed judgment, and ensures that the grounding system of the nuclear power plant can operate under safe and reliable conditions.

[0101] Example 3: Please refer to Figure 1 The data processing module 2 includes a data preprocessing unit 21 and a data integration unit 22. The data preprocessing unit 21 is used to preprocess and dimensionless the multi-source data set, and the data integration unit 22 is used to reorganize the processed multi-source data set to generate a first data set, a second data set, and a third data set;

[0102] The first data set includes ground current, ground voltage, and ground resistance;

[0103] The ground current includes a ground current standard value BZA and a ground current real-time value, wherein the ground current real-time values ​​are recorded as A1, A2, A3, ..., An according to timestamps;

[0104] The ground voltage includes a ground voltage standard value BZB and a ground voltage real-time value, wherein the ground voltage real-time values ​​are recorded as B1, B2, B3, ..., Bn according to timestamps;

[0105] The grounding resistance includes a standard value of the grounding resistance BZC and a real-time value of the grounding resistance, wherein the real-time values ​​of the grounding resistance are recorded as C1, C2, C3, ..., Cn according to the timestamps;

[0106] The second data set includes soil temperature, soil moisture, and soil salinity;

[0107] The soil temperature includes a soil temperature standard value BZD and a soil temperature real-time value, wherein the soil temperature real-time values ​​are recorded as D1, D2, D3, ..., Dn according to timestamps;

[0108] The soil moisture includes a soil moisture standard value BZE and a soil moisture real-time value, wherein the soil moisture real-time values ​​are recorded as E1, E2, E3, ..., En according to the timestamps;

[0109] Soil salinity includes soil salinity standard value BZF and soil salinity real-time value, wherein the soil salinity real-time value is recorded as F1, F2, F3, ..., Fn according to the timestamp;

[0110] The third data set includes grid frequency, power load, and grid harmonics;

[0111] The grid frequency includes a grid frequency standard value BZG and a grid frequency real-time value, wherein the grid frequency real-time values ​​are recorded as G1, G2, G3, ..., Gn according to timestamps;

[0112] The power load includes the power load standard value BZH and the power load real-time value, wherein the power load real-time value is recorded as H1, H2, H3, ..., Hn according to the timestamp;

[0113] The power grid harmonics include power grid harmonic standard values ​​BZI and power grid harmonic real-time values, wherein the power grid harmonic real-time values ​​are recorded as K1, K2, K3, ..., Kn according to timestamps.

[0114] In this embodiment: the data processing module 2 includes a data preprocessing unit 21 and a data integration unit 22. The data preprocessing unit 21 is used to preprocess the multi-source data set and perform dimensionless processing to ensure the comparability of different types of data and eliminate the impact caused by dimensional differences. The data integration unit 22 reorganizes the processed multi-source data to generate a first data set, a second data set, and a third data set.

[0115] Dimensionless processing ensures that when different types of data such as ground current, ground voltage, and ground resistance are compared, the impact of dimension differences can be eliminated, which improves the uniformity of data processing and the accuracy of analysis. By comparing the real-time value of each type of data with the standard value, real-time monitoring of the grounding system, environmental parameters, and power grid operation status can be achieved. This method can timely detect data deviations, early warn of potential faults, and improve the reliability and safety of the system. By marking and classifying real-time data by timestamp, the change trend of each data point can be accurately traced. This is crucial for data analysis, trend prediction, and fault diagnosis. It can improve the timeliness and accuracy of system fault prediction, clearly classify and organize data, and generate data sets suitable for subsequent analysis, avoiding data mixing and ensuring the efficiency and accuracy of subsequent analysis.

[0116] Example 4: Please refer to Figure 1 , the first data calculation module 3 calculates and obtains the first reference coefficient S1, the second reference coefficient S2 and the third reference coefficient S3 respectively through the following formulas:

[0117] S1=a1×PA+a2×PB+a3×PC;

[0118] S2=b1×PD+b2×PE+b3×PF;

[0119] S3 = c1 × PG + c2 × PH + c3 × PK;

[0120]

[0121]

[0122] Where: BZA is the standard value of ground current, BZB is the standard value of ground voltage, BZC is the standard value of ground resistance, BZD is the standard value of soil temperature, BZE is the standard value of soil moisture, BZF is the standard value of soil salinity, BZG is the standard value of grid frequency, BZH is the standard value of power load, and BZK is the standard value of grid harmonics;

[0123] An, Bn, Cn, Dn, En, Fn, Gn, Hn, and Kn are the real-time values ​​of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load, and grid harmonics at the nth timestamp, respectively;

[0124] Ai, Bi, Ci, Di, Ei, Fi, Gi, Hi, and Ki are the real-time values ​​of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load, and grid harmonics at the i-th timestamp, respectively;

[0125] i=n-m+1, m is the window value, n is the total number of timestamps;

[0126] PA is the ground current intervention coefficient, PB is the ground voltage intervention coefficient, PC is the ground resistance intervention coefficient, PD is the soil temperature intervention coefficient, PE is the soil moisture intervention coefficient, PF is the soil salinity intervention coefficient, PG is the grid frequency intervention coefficient, PH is the power load intervention coefficient, and PK is the grid harmonic intervention coefficient.

[0127] The first data calculation module 3 calculates and obtains the abnormal alarm coefficient YCX through the following formula:

[0128] YCX=d1×S1+d2×S2+d3×S3;

[0129] Where: S1 is the first reference coefficient, S2 is the second reference coefficient, and S3 is the third reference coefficient;

[0130] d1, d2 and d3 are weight values, and the values ​​of d1, d2 and d3 are adjusted and set by the user.

[0131] In this embodiment: by integrating multiple sources of data such as ground current, voltage, resistance, soil conditions and grid parameters, combined with user-adjustable weight values, the system can achieve more accurate and flexible abnormal monitoring. The impact of each data source on the monitoring results can be dynamically adjusted according to the actual operating conditions, thereby improving the accuracy of abnormal detection.

[0132] The introduction of timestamp and window mechanism in data calculation ensures that the system can process historical data in real time and respond based on the latest data. This dynamic calculation method solves the problem of slow response of traditional systems in the face of long-term changes and emergencies, and can reflect abnormal conditions of the grounding system more quickly.

[0133] The system generates various intervention coefficients by calculating the deviation between the standard value and the real-time data, which enables the system to detect changes in the environment and system parameters in real time and promptly discover potential abnormalities in the grounding system. The introduction of standardization eliminates the differences between data dimensions and ensures the consistency of data calculation.

[0134] In the calculation of the abnormal alarm coefficient YCX, users can freely adjust the weight values ​​of various data sources to optimize the monitoring results according to different operating conditions. This flexibility provides stronger adaptability and can be personalized for different operating environments and needs.

[0135] The real-time calculation and threshold comparison of the abnormal alarm coefficient YCX, combined with intelligent algorithms, enables the system to issue an alarm in a timely manner when potential risks are discovered. This intelligent abnormality detection can help operators identify possible faults or risks in the grounding system earlier, take measures in advance, and improve the safety and stability of nuclear power plants.

[0136] Example 5: Please refer to Figure 1 The first comparison result generated by the first data analysis module 4 is as follows:

[0137] When YCX≤Y, it means that there is no abnormality in the current nuclear power medium-voltage grounding. When YCX≤Y, it means that there is an abnormality in the current nuclear power medium-voltage grounding.

[0138] In this embodiment: the first data analysis module 4 provides a clear and easy-to-understand abnormality judgment standard by comparing with the preset threshold value Y. When YCX≤Y, it means that the nuclear power medium-voltage grounding system is in a normal state and the system does not need to be intervened; when YCX>Y, it means that the grounding system is abnormal and further diagnosis or measures are required. This mechanism simplifies the abnormality judgment process, facilitates the operator to respond quickly, and improves the overall monitoring efficiency.

[0139] The abnormality judgment mechanism allows the threshold Y to be adjusted to adapt to different monitoring needs and environmental conditions, thus providing flexible configuration options. According to the specific operating environment or monitoring requirements, users can adjust the setting of threshold Y so that the system can achieve the best monitoring effect according to different situations. This flexibility enhances the adaptability of the system, enabling it to maintain efficient and accurate monitoring capabilities under different operating conditions.

[0140] By comprehensively considering multiple source data such as ground current, ground voltage, and ground resistance, the first data analysis module 4 calculates YCX and compares it with the preset threshold value Y. This method effectively avoids the limitation of relying on a single data, can accurately judge abnormalities by combining the changes of multiple parameters, reduce the risk of false alarms or missed alarms, and further improve the monitoring accuracy and reliability of the system.

[0141] By calculating YCX in real time and comparing it with the threshold value Y, the system can continuously monitor the status of the grounding system. When an abnormality occurs, the system can quickly feedback to ensure that the fault can be discovered and handled in the first time. Based on this abnormality judgment mechanism, the nuclear power medium-voltage grounding monitoring system can automatically trigger an alarm when an abnormality occurs in the grounding system without manual intervention. This automation function greatly improves the response speed, especially for high-risk environments such as nuclear power. The automatic alarm system reduces the complexity of manual operation and ensures that corresponding measures can be taken in time when an abnormality occurs to reduce safety hazards.

[0142] Example 6: Please refer to Figure 1 The second data calculation module 5 calculates and obtains the first node alarm coefficient BJ1, the second node alarm coefficient BJ2 ​​and the third node alarm coefficient BJ3 respectively through the following formulas:

[0143]

[0144] Where: An, Bn, Cn, Dn, En, Fn, Gn, Hn, Kn are the real-time values ​​of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load and grid harmonic at the nth timestamp respectively;

[0145] JA is the mean value of ground current, JB is the mean value of ground voltage, JC is the mean value of ground resistance, JD is the mean value of soil temperature, JE is the mean value of soil moisture, JF is the mean value of soil salinity, JG is the mean value of grid frequency, JH is the mean value of power load, and JK is the mean value of grid harmonics;

[0146] BA is the standard deviation of ground current, BB is the standard deviation of ground voltage, BC is the standard deviation of ground resistance, BD is the standard deviation of soil temperature, BE is the standard deviation of soil moisture, BF is the standard deviation of soil salinity, BG is the standard deviation of grid frequency, BH is the standard deviation of power load, and BK is the standard deviation of grid harmonics.

[0147] The second data calculation module 5 calculates and obtains the ground current average JA, ground voltage average JB, ground resistance average JC, soil temperature average JD, soil moisture average JE, soil salinity average JF, grid frequency average JG, power load average JH, and grid harmonic average JK through the following formulas:

[0148]

[0149] Where: Ao, Bo, Co, Do, Eo, Fo, Go, Ho and Ko are the real-time value of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load and grid harmonic of the oth node in the timestamp respectively, and n is the total number of timestamps.

[0150] The second data calculation module 5 calculates and obtains the ground current standard deviation BA, ground voltage standard deviation BB, ground resistance standard deviation BC, soil temperature standard deviation BD, soil moisture standard deviation BE, soil salinity standard deviation BF, grid frequency standard deviation BG, power load standard deviation BH, and grid harmonic standard deviation BK through the following formulas:

[0151]

[0152] Where: Ao, Bo, Co, Do, Eo, Fo, Go, Ho and Ko are the real-time value of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load and grid harmonic of the oth node in the timestamp respectively, and n is the total number of timestamps;

[0153] JA is the mean value of ground current, JB is the mean value of ground voltage, JC is the mean value of ground resistance, JD is the mean value of soil temperature, JE is the mean value of soil moisture, JF is the mean value of soil salinity, JG is the mean value of grid frequency, JH is the mean value of power load, and JK is the mean value of grid harmonics;

[0154] n is the total number of timestamps.

[0155] In this embodiment, by calculating the alarm coefficients of various parameters, the fluctuation of each parameter can be diagnosed more carefully, thereby improving the system's detection accuracy of grounding anomalies. Through this process, the system can conduct comprehensive analysis from multiple dimensions and multiple parameters to ensure more accurate early warning.

[0156] By calculating the mean and standard deviation of multiple parameters such as ground current and ground voltage, the normal fluctuation range of these parameters at different time points can be reflected, thereby improving the stability of the monitoring system. The standard deviation can help identify the fluctuation range and anomalies of the data and avoid misdiagnosis caused by occasional extreme data.

[0157] Based on the real-time monitoring data, through the dynamic update of these means and standard deviations, the system can timely adjust its diagnostic strategy, timely warn of grounding system faults, and reduce the probability of potential accidents. Especially in a changing environment, it can make adaptive adjustments based on past data.

[0158] After introducing the multi-parameter weighting coefficient, different weights can be set according to the importance of different monitoring items, making the alarm response more intelligent and avoiding the problem of over-reliance on a single indicator.

[0159] Example Seven: Please refer to Figure 1 , the specific second comparison results generated by the second data analysis module 6 are as follows:

[0160] When BJ1 ≤ R1, it represents a first-level abnormality in the grounding electrical parameters. When R1 < BJ1 ≤ R1 × 105%, it represents a second-level abnormality in the grounding electrical parameters. When BJ1 > R1 × 105%, it represents a third-level abnormality in the grounding electrical parameters;

[0161] When BJ2 ≤ R2, it represents a first-level abnormality in the environmental parameters. When R2 < BJ2 ≤ R2 × 105%, it represents a second-level abnormality in the environmental parameters. When BJ2 > R2 × 105%, it represents a third-level abnormality in the environmental parameters;

[0162] When BJ3 ≤ R3, it represents a first-level abnormality in the power grid operation parameters. When R3 < BJ3 ≤ R3 × 105%, it represents a second-level abnormality in the power grid operation parameters. When BJ3 > R3 × 105%, it represents a third-level abnormality in the power grid operation parameters.

[0163] In this embodiment: By classifying abnormal situations into three levels and setting thresholds suitable for different abnormal situations, the system can real-time identify changes in grounding electrical parameters, environmental parameters, and power grid operation parameters, and judge the severity of abnormalities based on the degree of change. The hierarchical response mechanism optimizes the monitoring system's ability to handle abnormalities, and can take different measures according to the level of abnormalities, improving the flexibility and fault tolerance of the system. For relatively minor abnormalities, the system can provide warnings without immediate intervention, while for severe abnormalities, the system can trigger an emergency response to avoid the expansion of risks.

[0164] The independent threshold setting of grounding electrical parameters, environmental parameters, and power grid operation parameters ensures that the characteristics of each parameter in monitoring can be fully utilized. This method avoids misjudgment that may occur when all parameters are processed under the same standard, and ensures that the fluctuations of each parameter can be accurately evaluated. Especially in complex environments, it can independently monitor various parameters such as grounding current, voltage, and soil temperature, greatly improving the accuracy and reliability of the system. By using dynamic threshold adjustment, combining real-time data with historical fluctuation conditions, it ensures accurate judgment of abnormal events. This design does not rely solely on fixed thresholds, but through dynamic calculation methods such as standard deviation and mean, enabling the system to adaptively adjust the thresholds and flexibly change according to the actual situation. For different types of abnormalities, the system can real-time adjust the response mechanism, improving the intelligence and adaptability of the system.

[0165] After the introduction of the three-level response mechanism, the system can provide refined abnormal classification responses: Level 1 abnormalities represent minor problems that do not require immediate intervention; Level 2 abnormalities are moderately serious problems that require preventive measures; Level 3 abnormalities are major problems that require the system to activate the emergency response mechanism in a timely manner. Through this hierarchical early warning and emergency response mechanism, nuclear power plant operators can identify potential risks in advance and optimize emergency response processes to ensure equipment safety and stable operation.

[0166] The contents not described in detail in this specification belong to the prior art known to professional and technical personnel in this field.

[0167] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A nuclear power medium voltage grounding monitoring system based on artificial intelligence, characterized by: It comprises a data acquisition module (1), a data processing module (2), a first data calculation module (3), a first data analysis module (4), a second data calculation module (5), a second data analysis module (6) and a feedback module (7); The data acquisition module (1) is used to collect nuclear power medium voltage grounding data and form a multi-source data set; The data processing module (2) is used to pre-process the multi-source data sets and reorganize them into a first data set, a second data set and a third data set; The first data calculation module (3) is used to integrate and calculate the first data set, the second data set and the third data set respectively, so as to generate a first reference coefficient S1, a second reference coefficient S2 and a third reference coefficient S3, and integrate and calculate the first reference coefficient S1, the second reference coefficient S2 and the third reference coefficient S3, so as to generate an abnormal alarm coefficient YCX; The first data analysis module (4) is used to compare the abnormal alarm coefficient YCX with a preset first threshold value Y, thereby generating a first comparison result, and judging whether the nuclear power medium voltage grounding data is abnormal according to the first comparison result; If the first comparison result shows that the data is abnormal, the second data calculation module (5) calculates the first data set, the second data set and the third data set again, thereby generating a first node alarm coefficient BJ1, a second node alarm coefficient BJ2 ​​and a third node alarm coefficient BJ3; The second data analysis module (6) is used to compare the first node alarm coefficient BJ1, the second node alarm coefficient BJ2 ​​and the third node alarm coefficient BJ3 with the preset first node alarm threshold R1, the second node alarm threshold R2 and the third node alarm threshold R3 respectively, identify them through artificial intelligence, thereby generating a second comparison result, and judging the specific abnormal node according to the second comparison result; The feedback module (7) is used to send the first comparison result and the second comparison result to the terminal and generate a visual report.

2. According to the artificial intelligence-based nuclear power medium-voltage grounding monitoring system of claim 1, it is characterized by: The data acquisition module (1) comprises a first acquisition unit (11), a second acquisition unit (12) and a third acquisition unit (13); The first acquisition unit (11) acquires grounding electrical parameters including grounding current, grounding voltage and grounding resistance through sensors, and acquires standard values ​​of grounding current, grounding voltage and grounding resistance through a cloud platform; The second acquisition unit (12) acquires environmental parameters including soil temperature, soil moisture and soil salinity through sensors, and acquires standard values ​​of soil temperature, soil moisture and soil salinity of local soil through a cloud platform; The third acquisition unit (13) acquires power grid operation parameters including power grid frequency, power load and power grid harmonics through sensors, and acquires standard values ​​of power grid frequency, power load and power grid harmonics through a cloud platform; The ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load and grid harmonics constitute the multi-source data set.

3. According to the artificial intelligence-based nuclear power medium-voltage grounding monitoring system of claim 2, it is characterized by: The data processing module (2) comprises a data preprocessing unit (21) and a data integration unit (22), wherein the data preprocessing unit (21) is used to preprocess and dimensionlessly transform the multi-source data set, and the data integration unit (22) is used to rearrange the processed multi-source data set, thereby generating a first data set, a second data set, and a third data set; The first data set includes ground current, ground voltage, and ground resistance; The ground current includes a ground current standard value BZA and a ground current real-time value, wherein the ground current real-time value is recorded as A1, A2, A3, ..., An according to the timestamp; The ground voltage includes a ground voltage standard value BZB and a ground voltage real-time value, wherein the ground voltage real-time values ​​are recorded as B1, B2, B3, ..., Bn according to timestamps; The grounding resistance includes a grounding resistance standard value BZC and a grounding resistance real-time value, wherein the grounding resistance real-time values ​​are recorded as C1, C2, C3, ..., Cn according to timestamps; The second data set includes soil temperature, soil moisture, and soil salinity; The soil temperature includes a soil temperature standard value BZD and a soil temperature real-time value, wherein the soil temperature real-time values ​​are recorded as D1, D2, D3, ..., Dn according to timestamps; The soil moisture includes a soil moisture standard value BZE and a soil moisture real-time value, wherein the soil moisture real-time value is recorded as E1, E2, E3, ..., En according to the timestamp; The soil salinity includes a soil salinity standard value BZF and a soil salinity real-time value, wherein the soil salinity real-time value is recorded as F1, F2, F3, ..., Fn according to the timestamp; The third data set includes grid frequency, power load, and grid harmonics; The grid frequency includes a grid frequency standard value BZG and a grid frequency real-time value, wherein the grid frequency real-time values ​​are recorded as G1, G2, G3, ..., Gn according to timestamps; The electric load includes an electric load standard value BZH and an electric load real-time value, wherein the electric load real-time value is recorded as H1, H2, H3, ..., Hn according to the timestamp; The power grid harmonics include power grid harmonic standard values ​​BZI and power grid harmonic real-time values, wherein the power grid harmonic real-time values ​​are recorded as K1, K2, K3, . . . , Kn according to timestamps.

4. The artificial intelligence-based nuclear power medium voltage grounding monitoring system according to claim 3 is characterized in that: The first data calculation module (3) calculates and obtains the first reference coefficient S1, the second reference coefficient S2 and the third reference coefficient S3 respectively through the following formulas: S1=a1×PA+a2×PB+a3×PC; S2=b1×PD+b2×PE+b3×PF; S3 = c1 × PG + c2 × PH + c3 × PK; Where: BZA is the standard value of ground current, BZB is the standard value of ground voltage, BZC is the standard value of ground resistance, BZD is the standard value of soil temperature, BZE is the standard value of soil moisture, BZF is the standard value of soil salinity, BZG is the standard value of grid frequency, BZH is the standard value of power load, and BZK is the standard value of grid harmonics; An, Bn, Cn, Dn, En, Fn, Gn, Hn, and Kn are the real-time values ​​of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load, and grid harmonics at the nth timestamp, respectively; Ai, Bi, Ci, Di, Ei, Fi, Gi, Hi, and Ki are the real-time values ​​of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load, and grid harmonics at the i-th timestamp, respectively; i=n-m+1, m is the window value, n is the total number of timestamps; PA is the ground current intervention coefficient, PB is the ground voltage intervention coefficient, PC is the ground resistance intervention coefficient, PD is the soil temperature intervention coefficient, PE is the soil moisture intervention coefficient, PF is the soil salinity intervention coefficient, PG is the grid frequency intervention coefficient, PH is the power load intervention coefficient, and PK is the grid harmonic intervention coefficient.

5. The artificial intelligence-based nuclear power medium voltage grounding monitoring system according to claim 4 is characterized in that: The first data calculation module (3) calculates and obtains the abnormal alarm coefficient YCX through the following formula: YCX=d1×S1+d2×S2+d3×S3; Where: S1 is the first reference coefficient, S2 is the second reference coefficient, and S3 is the third reference coefficient; d1, d2 and d3 are weight values, and the values ​​of d1, d2 and d3 are adjusted and set by the user.

6. The artificial intelligence-based nuclear power medium voltage grounding monitoring system according to claim 5 is characterized in that: The first comparison result generated by the first data analysis module (4) is specifically as follows: When YCX≤Y, it means that there is no abnormality in the current nuclear power medium-voltage grounding. When YCX≤Y, it means that there is an abnormality in the current nuclear power medium-voltage grounding.

7. The artificial intelligence-based nuclear power medium voltage grounding monitoring system according to claim 6 is characterized in that: The second data calculation module (5) calculates and obtains the first node alarm coefficient BJ1, the second node alarm coefficient BJ2 ​​and the third node alarm coefficient BJ3 respectively through the following formulas: Where: An, Bn, Cn, Dn, En, Fn, Gn, Hn, Kn are the real-time values ​​of ground current, ground voltage, ground resistance, soil temperature, soil moisture, soil salinity, grid frequency, power load and grid harmonic at the nth timestamp respectively; JA is the mean value of ground current, JB is the mean value of ground voltage, JC is the mean value of ground resistance, JD is the mean value of soil temperature, JE is the mean value of soil moisture, JF is the mean value of soil salinity, JG is the mean value of grid frequency, JH is the mean value of power load, and JK is the mean value of grid harmonics; BA is the standard deviation of ground current, BB is the standard deviation of ground voltage, BC is the standard deviation of ground resistance, BD is the standard deviation of soil temperature, BE is the standard deviation of soil moisture, BF is the standard deviation of soil salinity, BG is the standard deviation of grid frequency, BH is the standard deviation of power load, and BK is the standard deviation of grid harmonics.

8. The artificial intelligence-based nuclear power medium-voltage grounding monitoring system according to claim 7 is characterized in that: The second data calculation module (5) calculates and obtains the average value JA of the grounding current, the average value JB of the grounding voltage, the average value JC of the grounding resistance, the average value JD of the soil temperature, the average value JE of the soil humidity, the average value JF of the soil salinity, the average value JG of the grid frequency, the average value JH of the power load, and the average value JK of the grid harmonics respectively through the following formulas: Where: Ao, Bo, Co, Do, Eo, Fo, Go, Ho, and Ko are respectively the real-time value of the grounding current, the real-time value of the grounding voltage, the real-time value of the grounding resistance, the real-time value of the soil temperature, the real-time value of the soil humidity, the real-time value of the soil salinity, the real-time value of the grid frequency, the real-time value of the power load, and the real-time value of the grid harmonics of the o-th node within the time stamp, and n is the total number of time stamps.

9. The artificial intelligence-based nuclear power medium voltage grounding monitoring system according to claim 8 is characterized in that: The second data calculation module (5) calculates and obtains the standard deviation BA of the grounding current, the standard deviation BB of the grounding voltage, the standard deviation BC of the grounding resistance, the standard deviation BD of the soil temperature, the standard deviation BE of the soil humidity, the standard deviation BF of the soil salinity, the standard deviation BG of the grid frequency, the standard deviation BH of the power load, and the standard deviation BK of the grid harmonics respectively through the following formulas: Where: Ao, Bo, Co, Do, Eo, Fo, Go, Ho, and Ko are respectively the real-time value of the grounding current, the real-time value of the grounding voltage, the real-time value of the grounding resistance, the real-time value of the soil temperature, the real-time value of the soil humidity, the real-time value of the soil salinity, the real-time value of the grid frequency, the real-time value of the power load, and the real-time value of the grid harmonics of the o-th node within the time stamp, and n is the total number of time stamps; JA is the average value of the grounding current, JB is the average value of the grounding voltage, JC is the average value of the grounding resistance, JD is the average value of the soil temperature, JE is the average value of the soil humidity, JF is the average value of the soil salinity, JG is the average value of the grid frequency, JH is the average value of the power load, and JK is the average value of the grid harmonics; n is the total number of time stamps.

10. The artificial intelligence-based nuclear power medium voltage grounding monitoring system according to claim 9, characterized in that: The specific second comparison results generated by the second data analysis module (6) are as follows: When BJ1 ≤ R1, it represents a first-level abnormality in the grounding electrical parameters. When R1 < BJ1 ≤ R1 × 105%, it represents a second-level abnormality in the grounding electrical parameters. When BJ1 > R1 × 105%, it represents a third-level abnormality in the grounding electrical parameters; When BJ2 ≤ R2, it represents a first-level abnormality in the environmental parameters. When R2 < BJ2 ≤ R2 × 105%, it represents a second-level abnormality in the environmental parameters. When BJ2 > R2 × 105%, it represents a third-level abnormality in the environmental parameters; When BJ3 ≤ R3, it represents a first-level abnormality in the grid operation parameters. When R3 < BJ3 ≤ R3 × 105%, it represents a second-level abnormality in the grid operation parameters. When BJ3 > R3 × 105%, it represents a third-level abnormality in the grid operation parameters.

Images