Substation multi-terminal fusion inspection system and method based on artificial intelligence

Abstract

The invention discloses a transformer substation multi-terminal fusion inspection system and method based on artificial intelligence, relates to the technical field of transformer substation inspection, and solves the problems that in the prior art, the risk of equipment is calculated only based on real-time operation parameters of the equipment, but the influence of environmental data on the equipment performance is not combined, so that the risk assessment is not comprehensive enough; and meanwhile, the technical problem of low inspection efficiency caused by the fact that most existing inspection paths are fixed or generated based on a preset rule and cannot be dynamically adjusted according to real-time risk changes or sudden faults is solved. The method comprises the following steps: acquiring operation data and corresponding environment data of each device in a transformer substation; calculating a risk score of each device based on the operation data; correcting the risk score based on the environmental data to obtain a corrected risk score; obtaining an optimal inspection path of the inspection robot based on the corrected risk score; and re-planning the inspection path based on the inspection result of the inspection robot, so that the technical problem is solved.

Description

Technical Field

[0001] This invention belongs to the field of substation inspection, specifically a substation multi-terminal integrated inspection system and method based on artificial intelligence. Background Technology

[0002] With the continuous expansion and increasing complexity of power grids, traditional substation inspection methods are no longer sufficient to meet the demands for efficient and accurate operation and maintenance. Traditional manual inspections are not only time-consuming and labor-intensive, but also easily affected by environmental and personal factors, leading to low inspection quality and efficiency. Therefore, improving the level of intelligence in substation inspection work has become an important research direction.

[0003] Existing technologies typically calculate equipment risk based solely on real-time operating parameters, without considering the impact of environmental data on equipment performance, resulting in incomplete risk assessment. Furthermore, existing inspection paths are mostly fixed or generated based on preset rules, failing to dynamically adjust according to real-time risk changes or sudden failures, leading to low inspection efficiency.

[0004] Therefore, this invention proposes an artificial intelligence-based multi-terminal integrated inspection system and method for substations to solve the above problems. Summary of the Invention

[0005] This invention aims to solve at least one of the technical problems existing in the prior art. To this end, this invention proposes a substation multi-terminal integrated inspection system and method based on artificial intelligence, which is used to solve the technical problem that the existing technology usually only calculates the risk of the equipment based on the real-time operating parameters of the equipment, but does not take into account the impact of environmental data on the equipment performance, resulting in an incomplete risk assessment. At the same time, the existing inspection paths are mostly fixed or generated based on preset rules, which cannot be dynamically adjusted according to real-time risk changes or sudden failures, resulting in low inspection efficiency.

[0006] To achieve the above objectives, the first aspect of the present invention provides a substation multi-terminal integrated inspection system based on artificial intelligence, comprising: a data acquisition module, a data analysis module, and an inspection module;

[0007] Data acquisition module: used to acquire the operating data of each device in the substation and the corresponding environmental data;

[0008] Data analysis module: Calculates a risk score for each device based on operational data; corrects the risk score based on environmental data to obtain a corrected risk score;

[0009] Inspection module: Obtain the optimal inspection path for the inspection robot based on the corrected risk score; replan the inspection path based on the inspection results of the inspection robot.

[0010] Preferably, the acquisition of operating data and corresponding environmental data for each device in the substation includes:

[0011] The system collects real-time operational data of each piece of equipment in the substation, as well as environmental data of the environment in which the equipment is located. The operational data includes the equipment's temperature and vibration frequency, while the environmental data includes ambient temperature and humidity.

[0012] Preferably, the risk score for each device calculated based on operational data includes:

[0013] The temperature of each device is denoted as Wi, and the vibration frequency of each device is denoted as Di; where Wi refers to the temperature of the i-th device, i = {1,2,3,…,N}, and N is the total number of devices; Di refers to the vibration frequency of the i-th device.

[0014] Through formula The risk score for each device is calculated; where Li is the risk score of the i-th device, ZWi is the optimal operating temperature of the i-th device, and α and β are weighting coefficients.

[0015] It should be noted that the weighting coefficients α and β are set by those skilled in the art based on experience, and can be obtained through expert scoring based on the Delphi method;

[0016] Excessive temperature usually indicates that the equipment may have problems such as overload, short circuit or poor heat dissipation, which will accelerate insulation aging and material fatigue, increasing the risk of failure; while excessively low temperature may indicate low equipment operating efficiency or heating system failure, which may also lead to unstable performance or damage; excessively high vibration frequency often reflects accelerated wear, loosening or imbalance of mechanical parts, indicating an increased risk of potential structural failure or malfunction.

[0017] Preferably, the correction of the risk score based on environmental data includes:

[0018] The ambient temperature is labeled as HW, and the ambient humidity is labeled as HS;

[0019] Through formula The risk score of each device is corrected to obtain the corrected risk score of each device; where XLi is the corrected risk score of the i-th device, ZHWi is the optimal ambient temperature of the i-th device, ZHSi is the optimal ambient humidity of the i-th device, and A and B are weighting coefficients.

[0020] It should be noted that the weighting coefficients A and B are set by those skilled in the art based on experience.

[0021] Preferably, obtaining the optimal inspection path for the inspection robot based on the modified risk score includes:

[0022] The equipment is sorted in descending order of its modified risk score to obtain the inspection ranking sequence;

[0023] The optimal inspection path for the inspection robot is obtained sequentially based on the inspection ranking sequence using a path planning algorithm; the path planning algorithm includes either the A* algorithm or the Dijkstra algorithm.

[0024] Preferably, the replanning of the inspection path based on the inspection results of the inspection robot includes:

[0025] S1: Based on historical fault records, obtain devices with strong transmission relationships and the strong transmission coefficients between devices;

[0026] S2: Determine whether the inspection robot has detected a device malfunction; if yes, extract the devices with a strong transmission relationship to the currently malfunctioning device and their corresponding strong transmission coefficients; if no, continue detecting and judging.

[0027] S3: Update the correction risk score of equipment that has a strong transmission relationship with the currently faulty equipment based on the corresponding strong transmission coefficient;

[0028] S4: Replan the inspection route based on the updated revised risk score.

[0029] Preferably, the step of obtaining the devices with strong transmission relationships and the strong transmission coefficients between devices based on historical fault records includes:

[0030] Based on the electrical connection diagram of the substation, the devices that are electrically connected through the same busbar are set as devices with strong conduction relationship, resulting in several groups of devices with strong conduction relationship;

[0031] Based on the historical fault records of the substation, obtain the total number of faults Ki for each device, and obtain the number of faults Y of devices with a strong transmission relationship with the original device from the total number of faults. i,j Where Ki is the total number of failures of the i-th device, and Y is... i,j It refers to the number of failures of the j-th device that has a strong transmission relationship with the i-th device out of the total number of failures of the i-th device, where j = {1,2,3,…,M}, and M is the total number of devices that have a strong transmission relationship with the i-th device;

[0032] Through formula F ij =(Y i,j The strong conduction coefficient between devices with a strong conduction relationship is calculated by ( / Ki)×θ; where F ij It refers to the strong transmission coefficient between the i-th device and the j-th device, θ is the risk score conversion coefficient, and 0 < θ < 0.1.

[0033] It should be noted that classifying devices electrically connected via the same busbar as having a strong fault conduction relationship based on the electrical connection diagram means that if two or more devices share the same busbar (i.e., their electrical connection paths are directly or indirectly connected via the same busbar), then there is a strong fault conduction relationship between them. This relationship indicates that when one device fails, the fault may quickly propagate to other devices via the busbar, leading to a chain reaction or increased risk. For example, if a device (such as a transformer) fails, the fault current may cause the circuit breaker to trip due to overcurrent through the busbar.

[0034] Preferably, updating the correction risk score of devices with a strong transmission relationship to the currently faulty device based on the corresponding strong transmission coefficient includes:

[0035] When the inspection robot detects a fault in a certain device, it acquires several devices that have a strong transmission relationship with the device that detected the fault.

[0036] By extracting the strong conduction coefficient between the faulty device and several devices with a strong conduction relationship, the strong conduction coefficient of each device among the several devices is obtained.

[0037] Extract the correction risk score for each device from a set of devices;

[0038] The correction term for each device is obtained by calculating the product of the corrected risk score of each device and the strong transmission coefficient of the corresponding device.

[0039] The updated correction risk score for each device is obtained by summing the correction risk score of each device with the corresponding correction item.

[0040] Preferably, the step of replanning the inspection route based on the updated corrected risk score includes:

[0041] The corrective risk scores of several devices that have a strong transmission relationship with the devices whose faults were detected by the inspection robot are replaced with the updated corrective risk scores to obtain the updated inspection ranking sequence.

[0042] Based on the updated inspection ranking sequence, the optimal inspection path for the inspection robot is obtained sequentially through a path planning algorithm.

[0043] A second aspect of the present invention provides an artificial intelligence-based multi-terminal fusion inspection method for substations, comprising:

[0044] Step 1: Obtain the operating data of each device in the substation and the corresponding environmental data;

[0045] Step 2: Calculate the risk score for each device based on the operational data;

[0046] Step 3: Correct the risk score based on environmental data to obtain the corrected risk score;

[0047] Step 4: Obtain the optimal inspection path for the inspection robot based on the corrected risk score;

[0048] Step 5: Replan the inspection path based on the inspection results of the inspection robot.

[0049] Compared with the prior art, the beneficial effects of the present invention are:

[0050] 1. Existing technologies typically calculate equipment risk based solely on real-time operating parameters, neglecting to consider the impact of environmental data on equipment performance, resulting in incomplete risk assessment. This invention constructs a more comprehensive risk assessment model by integrating equipment operating data and environmental data. By collecting real-time equipment operating data and environmental data, combined with the equipment's optimal operating parameters, a weighted formula is used to dynamically calculate and correct risk scores. This considers both the direct impact of abnormal equipment conditions on failures and quantifies the indirect impact of environmental factors on equipment performance, making risk assessment more comprehensive and accurate. Based on the corrected risk score ranking, a path planning algorithm is used to generate a dynamically optimal path, ensuring that high-risk equipment is inspected first, significantly improving inspection efficiency.

[0051] 2. Existing inspection paths are mostly fixed or generated based on preset rules, which cannot be dynamically adjusted according to real-time risk changes or sudden failures, resulting in low inspection efficiency. This invention obtains strong transmission relationships between devices (such as the fault propagation path of busbar-related devices) based on electrical connection diagrams and historical fault records, and quantitatively calculates the transmission coefficient through formulas to quantify the possibility of a fault spreading from one device to related devices. By extracting the transmission coefficient between the faulty device and the strongly related devices, the modified risk score of the related devices is multiplied by the transmission coefficient and then superimposed on the original score, thereby quickly improving the risk level of devices with potential cascading failures. Based on the updated risk score, the optimal inspection path is regenerated, prioritizing the re-inspection or emergency investigation of high-risk devices, achieving rapid fault location and key inspection of related risk devices, effectively suppressing fault propagation, reducing maintenance delays, significantly improving the utilization efficiency of inspection resources, and greatly reducing the systemic risks caused by cascading failures. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1This is a schematic diagram of the system modules according to an embodiment of the present invention;

[0054] Figure 2 This is a schematic diagram of the corrected risk score update method according to an embodiment of the present invention;

[0055] Figure 3 This is a schematic diagram of the method steps in an embodiment of the present invention. Detailed Implementation

[0056] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] Please see Figure 1 The first aspect of the present invention provides a substation multi-terminal fusion inspection system based on artificial intelligence, including: a data acquisition module, a data analysis module and an inspection module;

[0058] Data acquisition module: used to acquire the operating data of each device in the substation and the corresponding environmental data;

[0059] Data analysis module: Calculates a risk score for each device based on operational data; corrects the risk score based on environmental data to obtain a corrected risk score;

[0060] Inspection module: Obtain the optimal inspection path for the inspection robot based on the corrected risk score; replan the inspection path based on the inspection results of the inspection robot.

[0061] Acquire operational data and corresponding environmental data for each device in the substation, including:

[0062] The system collects real-time operational data of each piece of equipment in the substation, as well as environmental data of the environment in which the equipment is located. The operational data includes the equipment's temperature and vibration frequency, while the environmental data includes ambient temperature and humidity.

[0063] The risk score for each device is calculated based on operational data, including:

[0064] The temperature of each device is denoted as Wi, and the vibration frequency of each device is denoted as Di; where Wi refers to the temperature of the i-th device, i = {1,2,3,…,N}, and N is the total number of devices; Di refers to the vibration frequency of the i-th device.

[0065] Through formula The risk score for each device is calculated; where Li is the risk score of the i-th device, ZWi is the optimal operating temperature of the i-th device, and α and β are weighting coefficients.

[0066] Risk scores are revised based on environmental data, including:

[0067] The ambient temperature is labeled as HW, and the ambient humidity is labeled as HS;

[0068] Through formula The risk score of each device is corrected to obtain the corrected risk score of each device; where XLi is the corrected risk score of the i-th device, ZHWi is the optimal ambient temperature of the i-th device, ZHSi is the optimal ambient humidity of the i-th device, and A and B are weighting coefficients.

[0069] The optimal inspection path for the inspection robot is obtained based on the revised risk score, including:

[0070] The equipment is sorted in descending order of its modified risk score to obtain the inspection ranking sequence;

[0071] The optimal inspection path for the inspection robot is obtained sequentially based on the inspection ranking sequence using a path planning algorithm; the path planning algorithm includes either the A* algorithm or the Dijkstra algorithm.

[0072] For example: Suppose a substation has 3 critical pieces of equipment:

[0073] Equipment 1: Main transformer (ZWi = 80℃, ZHWi = 25℃, ZHSi = 50%);

[0074] Equipment 2: Circuit breaker (ZWi=60℃, ZHWi=25℃, ZHSi=50%);

[0075] Equipment 3: Disconnecting switch (ZWi=50℃, ZHWi=25℃, ZHSi=50%);

[0076] Real-time data acquisition:

[0077] Running data:

[0078] Equipment 1 temperature (W1) = 90℃, vibration frequency (D1) = 45Hz;

[0079] Equipment 2 has a temperature (W2) of 65℃ and a vibration frequency (D2) of 35Hz.

[0080] Equipment 3 has a temperature (W3) of 55℃ and a vibration frequency (D3) of 25Hz.

[0081] Environmental data:

[0082] Ambient temperature (HW) = 30℃, ambient humidity (HS) = 70%;

[0083] Weighting coefficients (determined via the Delphi method):

[0084] α = 0.7 (temperature weight), β = 0.3 (vibration weight);

[0085] A = 0.6 (weight of ambient temperature), B = 0.4 (weight of ambient humidity);

[0086] 2. Calculation of initial risk score (Li);

[0087] Through formula Calculate the risk score for each device;

[0088] Device 1: Substituting the data, we get L1≈0.919;

[0089] Device 2: Substituting the data, we get L2≈0.913;

[0090] Device 3: Substituting the data, we get L3≈0.915;

[0091] 3. Correct the calculation of the risk score (XLi);

[0092] Through formula The risk score for each device was revised.

[0093] Device 1: Substituting the data, we get XL1≈1;

[0094] Device 2: Substituting the data, we get XL1≈0.993;

[0095] Device 3: Substituting the data, we get XL1≈0.995;

[0096] 4. Inspection route planning;

[0097] Correct risk score ranking:

[0098] The equipment is ordered from highest to lowest according to the revised risk score: Equipment 1, Equipment 3, Equipment 2;

[0099] The optimal inspection path for the inspection robot is obtained sequentially using a path planning algorithm based on the sorting process.

[0100] Optimal path generation:

[0101] Starting point → Device 1 (highest priority);

[0102] Equipment 1 → Equipment 3;

[0103] Equipment 3 → Equipment 2;

[0104] By using real-time data and a dynamic correction model, the system accurately identifies high-risk equipment (such as main transformers) and optimizes the inspection sequence by combining path planning algorithms. In this example, equipment 1 has the highest correction score and is prioritized for robot inspection, effectively preventing the spread of potential faults. This method takes into account both data-driven and physical correlation, significantly improving the initiative and efficiency of substation operation and maintenance.

[0105] Based on the inspection results from the inspection robot, the inspection path is replanned, including:

[0106] S1: Based on historical fault records, obtain devices with strong transmission relationships and the strong transmission coefficients between devices;

[0107] S2: Determine whether the inspection robot has detected a device malfunction; if yes, extract the devices with a strong transmission relationship to the currently malfunctioning device and their corresponding strong transmission coefficients; if no, continue detecting and judging.

[0108] S3: Update the correction risk score of equipment that has a strong transmission relationship with the currently faulty equipment based on the corresponding strong transmission coefficient;

[0109] S4: Replan the inspection route based on the updated revised risk score.

[0110] Based on historical fault records, obtain the devices with strong transmission relationships and the strong transmission coefficients between devices, including:

[0111] Based on the electrical connection diagram of the substation, the devices that are electrically connected through the same busbar are set as devices with strong conduction relationship, resulting in several groups of devices with strong conduction relationship;

[0112] Based on the historical fault records of the substation, obtain the total number of faults Ki for each device, and obtain the number of faults Y of devices with a strong transmission relationship with the original device from the total number of faults. i,j Where Ki is the total number of failures of the i-th device, and Y is... i,j It refers to the number of failures of the j-th device that has a strong transmission relationship with the i-th device out of the total number of failures of the i-th device, where j = {1,2,3,…,M}, and M is the total number of devices that have a strong transmission relationship with the i-th device;

[0113] Through formula F ij =(Y i,j The strong conduction coefficient between devices with a strong conduction relationship is calculated by ( / Ki)×θ; where F ij It refers to the strong transmission coefficient between the i-th device and the j-th device, θ is the risk score conversion coefficient, and 0 < θ < 0.1.

[0114] See Figure 2Based on the corresponding strong conduction coefficient, the correction risk score of equipment with a strong conduction relationship with the currently faulty equipment is updated, including:

[0115] When the inspection robot detects a fault in a certain device, it acquires several devices that have a strong transmission relationship with the device that detected the fault.

[0116] By extracting the strong conduction coefficient between the faulty device and several devices with a strong conduction relationship, the strong conduction coefficient of each device among the several devices is obtained.

[0117] Extract the correction risk score for each device from a set of devices;

[0118] The correction term for each device is obtained by calculating the product of the corrected risk score of each device and the strong transmission coefficient of the corresponding device.

[0119] The updated correction risk score for each device is obtained by summing the correction risk score of each device with the corresponding correction item.

[0120] The inspection route was replanned based on the updated revised risk score, including:

[0121] The corrective risk scores of several devices that have a strong transmission relationship with the devices whose faults were detected by the inspection robot are replaced with the updated corrective risk scores to obtain the updated inspection ranking sequence.

[0122] Based on the updated inspection ranking sequence, the optimal inspection path for the inspection robot is obtained sequentially through a path planning algorithm.

[0123] For example: Suppose there are 4 pieces of equipment in a substation:

[0124] Equipment A: Main transformer;

[0125] Equipment B: Circuit breaker;

[0126] Equipment C: Disconnecting switch;

[0127] Equipment D: Surge arrester;

[0128] Parameter settings:

[0129] The risk score conversion coefficient θ = 0.02;

[0130] Assuming an initial revised risk score:

[0131] Equipment A: 1;

[0132] Device B: 0.996;

[0133] Device C: 0.997;

[0134] Device D: 0.995;

[0135] The inspection sequence is: Equipment A, Equipment C, Equipment B, Equipment D.

[0136] Historical fault records (past 3 years):

[0137] Faulty equipment Conducting fault equipment frequency Equipment A Equipment B 30 Equipment C Device D 20

[0138] Total number of equipment failures:

[0139] Equipment A: 60 times (self-failure); of which, the number of times equipment A and equipment B failed simultaneously is 30 times, as shown in the table above.

[0140] Equipment C: 50 times; of which, the number of times that Equipment C and Equipment D failed simultaneously is 30 times, as shown in the table above.

[0141] Step S1: Calculate the strong conductivity coefficient;

[0142] Through formula F ij =(Y i,j Calculate the strong conduction coefficient between devices with a strong conduction relationship using / Ki)×θ;

[0143] Equipment A and Equipment B:

[0144] Substituting the data, the strong conduction coefficients of devices A and B are obtained as: 0.01;

[0145] Equipment C and Equipment D:

[0146] Substituting the data, the strong conductivity coefficients of devices C and D are obtained as follows: 0.008;

[0147] Step S2: Detect equipment malfunctions;

[0148] Suppose that the inspection robot detects a fault in device A, triggering the transmission relationship processing flow.

[0149] Step S3: Update and correct the risk score;

[0150] Because equipment A and equipment B have a strong transmission relationship, when equipment A fails, the corrective risk score of equipment B is updated:

[0151] The correction term is obtained by multiplying the corrected risk score of device B by the corresponding strong transmission coefficient; specifically: 0.996 × 0.01 = 0.00996;

[0152] The updated correction risk score for each device is obtained by summing the correction risk score of device B with the correction items of the corresponding device; specifically: 0.996 + 0.00996 ≈ 1.006.

[0153] The updated device fix risk score is:

[0154] Equipment A: Malfunctioning, marked as requiring maintenance;

[0155] Device B: 1.006;

[0156] Device C: 0.997;

[0157] Device D: 0.995.

[0158] Step S4: Replan the inspection route;

[0159] The updated inspection ranking sequence is: Equipment B, Equipment C, Equipment D;

[0160] Based on the updated inspection ranking sequence, the optimal inspection path for the inspection robot is obtained sequentially through a path planning algorithm.

[0161] By detecting faults in device A in real time, the system dynamically improves the risk score of device B based on transmission relationships and generates inspection paths that prioritize high-risk devices. This method effectively shortens fault response time (device B is detected 20 minutes earlier) and reduces the probability of secondary faults caused by transmission.

[0162] See Figure 3 The second aspect of this invention provides an artificial intelligence-based multi-terminal fusion inspection method for substations, comprising:

[0163] Step 1: Obtain the operating data of each device in the substation and the corresponding environmental data;

[0164] Step 2: Calculate the risk score for each device based on the operational data;

[0165] Step 3: Correct the risk score based on environmental data to obtain the corrected risk score;

[0166] Step 4: Obtain the optimal inspection path for the inspection robot based on the corrected risk score;

[0167] Step 5: Replan the inspection path based on the inspection results of the inspection robot.

[0168] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0169] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A substation multi-terminal fusion inspection system based on artificial intelligence, characterized in that, include: Data acquisition module, data analysis module, and inspection module; Data acquisition module: used to acquire the operating data of each device in the substation and the corresponding environmental data; Data analysis module: Calculates a risk score for each device based on operational data; The risk score is corrected based on environmental data to obtain a corrected risk score; Inspection module: Obtains the optimal inspection path for the inspection robot based on the corrected risk score; The inspection path is replanned based on the inspection results of the inspection robot.

2. The substation multi-terminal fusion inspection system based on artificial intelligence according to claim 1, characterized in that, The acquisition of operational data and corresponding environmental data for each device in the substation includes: The system collects real-time operational data of each piece of equipment in the substation, as well as environmental data of the environment in which the equipment is located. The operational data includes the equipment's temperature and vibration frequency, while the environmental data includes ambient temperature and humidity.

3. The substation multi-terminal fusion inspection system based on artificial intelligence according to claim 1, characterized in that, The risk score for each device, calculated based on operational data, includes: The temperature of each device is denoted as Wi, and the vibration frequency of each device is denoted as Di; where Wi refers to the temperature of the i-th device, i = {1,2,3,…,N}, and N is the total number of devices; Di refers to the vibration frequency of the i-th device. Through formula The risk score for each device is calculated; where Li is the risk score of the i-th device, ZWi is the optimal operating temperature of the i-th device, and α and β are weighting coefficients.

4. The substation multi-terminal fusion inspection system based on artificial intelligence according to claim 1, characterized in that, The correction of risk scores based on environmental data includes: The ambient temperature is labeled as HW, and the ambient humidity is labeled as HS; Through formula The risk score of each device is corrected to obtain the corrected risk score of each device; where XLi is the corrected risk score of the i-th device, ZHWi is the optimal ambient temperature of the i-th device, ZHSi is the optimal ambient humidity of the i-th device, and A and B are weighting coefficients.

5. The substation multi-terminal fusion inspection system based on artificial intelligence according to claim 1, characterized in that, The method for obtaining the optimal inspection path for the inspection robot based on the modified risk score includes: The equipment is sorted in descending order of its modified risk score to obtain the inspection ranking sequence; The optimal inspection path for the inspection robot is obtained sequentially based on the inspection ranking sequence using a path planning algorithm; the path planning algorithm includes either the A* algorithm or the Dijkstra algorithm.

6. The substation multi-terminal fusion inspection system based on artificial intelligence according to claim 1, characterized in that, The replanning of inspection paths based on the inspection results of the inspection robot includes: S1: Based on historical fault records, obtain devices with strong transmission relationships and the strong transmission coefficients between devices; S2: Determine whether the inspection robot has detected a device malfunction; if yes, extract the devices with a strong transmission relationship to the currently malfunctioning device and their corresponding strong transmission coefficients; if no, continue detecting and judging. S3: Update the correction risk score of equipment that has a strong transmission relationship with the currently faulty equipment based on the corresponding strong transmission coefficient; S4: Replan the inspection route based on the updated revised risk score.

7. The substation multi-terminal fusion inspection system based on artificial intelligence according to claim 6, characterized in that, The process of obtaining devices with strong transmission relationships and the strong transmission coefficients between devices based on historical fault records includes: Based on the electrical connection diagram of the substation, the devices that are electrically connected through the same busbar are set as devices with strong conduction relationship, resulting in several groups of devices with strong conduction relationship; Based on the historical fault records of the substation, obtain the total number of faults Ki for each device, and obtain the number of faults Y of devices with a strong transmission relationship with the original device from the total number of faults. i,j Where Ki is the total number of failures of the i-th device, and Y is... i,j It refers to the number of failures of the j-th device that has a strong transmission relationship with the i-th device out of the total number of failures of the i-th device, where j = {1,2,3,…,M}, and M is the total number of devices that have a strong transmission relationship with the i-th device; Through formula F ij =(Y i,j The strong conduction coefficient between devices with a strong conduction relationship is calculated by ( / Ki)×θ; where F ij It refers to the strong transmission coefficient between the i-th device and the j-th device, θ is the risk score conversion coefficient, and 0 < θ < 0.

1.

8. The substation multi-terminal fusion inspection system based on artificial intelligence according to claim 6, characterized in that, The update of the correction risk score for devices with a strong transmission relationship to the currently faulty device based on the corresponding strong transmission coefficient includes: When the inspection robot detects a fault in a certain device, it acquires several devices that have a strong transmission relationship with the device that detected the fault. By extracting the strong conduction coefficient between the faulty device and several devices with a strong conduction relationship, the strong conduction coefficient of each device among the several devices is obtained. Extract the correction risk score for each device from a set of devices; The correction term for each device is obtained by calculating the product of the corrected risk score of each device and the strong transmission coefficient of the corresponding device. The updated correction risk score for each device is obtained by summing the correction risk score of each device with the corresponding correction item.

9. A substation multi-terminal fusion inspection system based on artificial intelligence according to claim 6, characterized in that, The process of replanning the inspection route based on the updated revised risk score includes: The corrective risk scores of several devices that have a strong transmission relationship with the devices whose faults were detected by the inspection robot are replaced with the updated corrective risk scores to obtain the updated inspection ranking sequence. Based on the updated inspection ranking sequence, the optimal inspection path for the inspection robot is obtained sequentially through a path planning algorithm.

10. A substation multi-terminal fusion inspection method based on artificial intelligence, applied to the substation multi-terminal fusion inspection system based on artificial intelligence as described in any one of claims 1-9, characterized in that, include: Step 1: Obtain the operating data of each device in the substation and the corresponding environmental data; Step 2: Calculate the risk score for each device based on the operational data; Step 3: Correct the risk score based on environmental data to obtain the corrected risk score; Step 4: Obtain the optimal inspection path for the inspection robot based on the corrected risk score; Step 5: Replan the inspection path based on the inspection results of the inspection robot.

Images