High-voltage Cable Defect Identification Method and System Based on Cross-Connected Grounding Loop Current

Abstract

The present invention discloses a method and system for identifying high-voltage cable defects based on cross-connected grounding loop current, including: constructing an equivalent model of the sheath loop current of a three-phase cable cross-connected grounding system for a single-core cable model with a cross-connected grounding method, and calculating the value of the sheath loop current; obtaining the metal sheath loss and the internal temperature of the cable for different types of typical defects based on a field-circuit coupling model of multiple physical fields, and constructing a classification table of typical defect modes for the cross-connected system; establishing an integrated learning model based on the LightGBM algorithm, training the integrated learning model to obtain a trained integrated learning model; calculating the loop current data of the target high-voltage cable, obtaining the target defect feature quantity of the target high-voltage cable, and classifying the target high-voltage cable based on the trained integrated learning model to obtain the defect type of the target high-voltage cable. The present invention significantly improves the calculation accuracy of the system loop current and the accuracy and accuracy of high-voltage cable defect identification.

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and particularly to a method and system for identifying high-voltage cable defects based on cross-connected grounding loop current. Background Art

[0002] High-voltage cross-linked polyethylene (XLPE) cables are widely used in major power transmission projects such as ultra-high voltage and extra-high voltage. Their insulation performance is crucial for the safe and stable operation of the power system. As the operation years increase, the cables are prone to failures due to various factors, seriously affecting the safe and stable operation of the power grid. Since they are usually laid underground, traditional planned maintenance lacks pertinence, and it is difficult to diagnose and handle fault defects. At present, the operation state of the cable is mainly judged by analyzing the sheath loop current, but most only analyze the calculation method of the induced current component in the sheath loop current, directly ignoring the leakage current component, and not considering the influence of the changes in the parameters of the cross-connected system, resulting in problems such as low accuracy in identifying high-voltage cable defects, and low efficiency and accuracy.

[0003] In view of this, the present application is specifically proposed. Summary of the Invention

[0004] The technical problem to be solved by the present invention is that the prior art mainly judges the operation state of the cable by analyzing the sheath loop current, but most only analyze the calculation method of the induced current component in the sheath loop current, directly ignoring the leakage current component, and not considering the influence of the changes in the parameters of the cross-connected system, resulting in problems such as low accuracy in identifying high-voltage cable defects, and low efficiency and accuracy. The purpose of the present invention is to provide a method and system for identifying high-voltage cable defects based on cross-connected grounding loop current. This method significantly improves the calculation accuracy of the system loop current, meets the loop current analysis and monitoring requirements of various typical defects; the present invention achieves the ability to identify high-voltage cable defects with high accuracy, and also greatly improves the efficiency and accuracy of defect handling in the high-voltage cable cross-connected system.

[0005] The present invention is realized by the following technical solutions:

[0006] In the first aspect, the present invention provides a method for identifying high-voltage cable defects based on cross-connected grounding loop current, the method comprising:

[0007] For a single-core cable model with a cross-connected grounding method, construct an equivalent model of the sheath loop current of a three-phase cable cross-connected grounding system; based on the equivalent model of the sheath loop current, calculate the value of the sheath loop current;

[0008] According to the value of the sheath loop current, based on the field-circuit coupling model of multiple physical fields, obtain the metal sheath loss and the internal temperature of the cable for different types of typical defects, and construct a classification table of typical defect modes of the cross-connected system;

[0009] An integrated learning model based on the LightGBM algorithm is established, and the integrated learning model based on the LightGBM algorithm is trained according to the sheath circulating current value and the typical defect mode classification table to obtain a trained integrated learning model;

[0010] Calculate the circulating current data of the target high-voltage cable, obtain the target defect feature quantity of the target high-voltage cable, and classify the target high-voltage cable based on the trained integrated learning model to obtain the defect type of the target high-voltage cable.

[0011] In the above technical solution, considering that the existing technology uses a circuit method and only analyzes the calculation method of the induced current component in the sheath circulating current, while directly ignoring the leakage current component; therefore, the present invention is based on the field-circuit coupling idea, solves the field-circuit coupling model of multiple physical fields, and solves the problem of ignoring the leakage current component. Based on the sheath circulating current value and the typical defect mode classification table, the present invention considers the influence of the change of each parameter of the cross-bonding system, achieves a high-precision high-voltage cable defect recognition ability, and greatly improves the efficiency and accuracy of defect handling of the high-voltage cable cross-bonding system.

[0012] Further, for the single-core cable model with cross-bonding grounding method, an equivalent model of the sheath circulating current of the three-phase cable cross-bonding grounding system is constructed; based on the equivalent model of the sheath circulating current, the sheath circulating current value is calculated, including:

[0013] S1.1. Based on the three-phase cross-bonding grounding method, each cable part in the single-core cable model is divided into three sections with equal length, and each section is grounded through the sheath protector in the transposition box, and the two terminal metal sheaths of each cross-bonding section are directly grounded;

[0014] Among them, the three-phase high-voltage cable is divided into three phases a, b, and c. The a-phase cable is divided into three sections a1, a2, and a3, the b-phase cable is divided into three sections b1, b2, and b3, and the c-phase cable is divided into three sections c1, c2, and c3.

[0015] S1.2. According to the equivalent circuit diagram of the single-core cable, the loop current equation about the sheath circulating current is obtained. The loop current equation is specifically:

[0016]

[0017] In the formula, are the induced voltages generated by the core conductor currents of the a1, a2, and a3 sections of the cable respectively, is the induced voltage generated by the core conductor currents of the b1, b2, and b3 sections of the cable, is the induced voltage generated by the core conductor currents of the c1, c2, and c3 sections of the cable; is the in-ground circulating current, R eqis the equivalent resistance value; is the induced voltage generated by the circulating currents in the metal sheaths of phases b and c on the metal sheath of the i-th land cable of phase a, is the induced voltage generated by the circulating currents in the metal sheaths of phases a and c on the metal sheath of the i-th land cable of phase b, is the induced voltage generated by the circulating currents in the metal sheaths of phases a and b on the metal sheath of the i-th land cable of phase c; is the grounding circulating current of phase a cable, is the grounding circulating current of phase b cable, is the grounding circulating current of phase c cable; Z ai is the self-impedance of the metal sheath of the i-th section of phase a cable, Z bi is the self-impedance of the metal sheath of the i-th section of phase b cable, Z ci is the self-impedance of the metal sheath of the i-th section of phase c cable.

[0018] Among them, the self-impedances Z ai 、Z bi 、Z ci of the metal sheaths of the i-th sections of phase a, phase b, and phase c cables are specifically expressed as:

[0019]

[0020] In the formula, R ai is the sheath resistance of the i-th section of phase a cable, R bi is the sheath resistance of the i-th section of phase b cable, R ci is the sheath resistance of the i-th section of phase c cable; L aai is the self-inductance per unit length of the loop formed by the metal sheath of the i-th section of phase a cable and the ground, L bbi is the self-inductance per unit length of the loop formed by the metal sheath of the i-th section of phase b cable and the ground, L cci is the self-inductance per unit length of the loop formed by the metal sheath of the i-th section of phase c cable and the ground; l i is the length of the i-th cable line section.

[0021] Among them, the induced voltages generated by the core conductor currents of the i-th sections of phase a, phase b, and phase c cables are specifically expressed as:

[0022]

[0023] In the formula, M Ab is the mutual inductance per unit length between the core of phase a cable and the sheath of phase b cable, M Ac is the mutual inductance per unit length between the core of phase a cable and the sheath of phase c cable; M Ba is the mutual inductance per unit length between the core of phase b cable and the sheath of phase a cable, M BcThe mutual inductance per unit length between the core of the B-phase cable and the sheath of the C-phase cable; M Ca The mutual inductance per unit length between the core of the C-phase cable and the sheath of the A-phase cable, M Cb The mutual inductance per unit length between the core of the C-phase cable and the sheath of the B-phase cable; M Aa The mutual inductance per unit length between the core of the A-phase cable and the sheath of the A-phase cable, M Bb The mutual inductance per unit length between the core of the B-phase cable and the sheath of the B-phase cable, M Cc The mutual inductance per unit length between the core of the C-phase cable and the sheath of the C-phase cable; Is the ground loop current of the A-phase cable, Is the ground loop current of the B-phase cable, Is the ground loop current of the C-phase cable.

[0024] Among them, the induced voltage generated by the circulating current in the metal sheaths of the other two phases on the i-th section of the land cable of the A, B, and C phases The calculation formula is

[0025]

[0026] In the formula, M abi Is the mutual inductance per unit length of the loop formed by the sheaths of the i-th sections of the A-phase and B-phase cables and the ground, M aci Is the mutual inductance per unit length of the loop formed by the sheaths of the i-th sections of the A-phase and C-phase cables and the ground, M bci Is the mutual inductance per unit length of the loop formed by the sheaths of the i-th sections of the B-phase and C-phase cables and the ground; Is the ground loop current of the A-phase cable, Is the ground loop current of the B-phase cable, Is the ground loop current of the C-phase cable; l i Is the length of the i-th section of the cable line.

[0027] S1.3. Substitute formulas (2) to (4) into the loop equations of formula (1), derive the ground loop current matrix equation, and use the ground loop current matrix equation as the sheath circulating current equivalent model; based on the sheath circulating current equivalent model, calculate the numerical value of the sheath circulating current.

[0028] The expression of the sheath circulating current equivalent model is:

[0029]

[0030] In the formula, Is the ground loop current of the A-phase cable, Is the ground loop current of the B-phase cable, Is the ground loop current of the C-phase cable; They are the induced voltages generated by the conductor currents of the cable core segments a1, a2, and a3 respectively, are the induced voltages generated by the conductor currents of the cable core segments b1, b2, and b3, are the induced voltages generated by the conductor currents of the cable core segments c1, c2, and c3; Z 11 from Z 33 to Z are impedances;

[0031] Among them, the impedance Z 11 from Z 33 to Z has the following calculation formula:

[0032]

[0033] Furthermore, based on the sheath circulating current value and the field-circuit coupling model of multiple physical fields, the metal sheath losses and the internal cable temperature of different types of typical defects are obtained, including:

[0034] Establish a finite element calculation model for the single-core cable model and perform mesh division on the finite element calculation model;

[0035] Input the sheath circulating current value as the excitation source into the finite element calculation model to obtain the internal conductor conductivity and electromagnetic loss distribution results of the cable; calculate the internal temperature field distribution of the cable according to the electromagnetic loss distribution results;

[0036] Input the internal temperature field distribution of the cable into the finite element calculation model to recalculate the electromagnetic loss distribution; compare the electromagnetic loss distribution results before and after, and if the calculation error of the comparison exceeds the set error range, perform cyclic iteration until the set error range is satisfied and output the metal sheath loss and the internal cable temperature.

[0037] In the above technical solutions, the present invention combines the two-way coupling idea of the field and the circuit. First, the sheath circulating current value is input as the excitation source into the finite element calculation model to obtain the internal conductor conductivity and electromagnetic loss distribution results of the cable; calculate the internal temperature field distribution of the cable according to the electromagnetic loss distribution results; then input the internal temperature field distribution of the cable into the finite element calculation model to recalculate the electromagnetic loss distribution; compare the electromagnetic loss distribution results before and after until the calculation error of the comparison meets the set error range to obtain the metal sheath loss and the internal cable temperature. The present invention overcomes the influence of directly ignoring the leakage current component in the original circuit method.

[0038] Furthermore, construct a classification table of typical defect modes of the cross-bonding system, including:

[0039] Use the metal sheath loss and the internal cable temperature as defect characteristic quantities to encode different types of typical defects of the cross-bonding system to obtain defect type codes;

[0040] Construct a classification table of typical defect patterns for the cross-connected system according to the defect characteristic quantity and defect type coding.

[0041] Furthermore, according to the sheath circulating current value and the classification table of typical defect patterns, train the integrated learning model based on the LightGBM algorithm to obtain a trained integrated learning model, including:

[0042] Construct a sample set for defect classification training and testing according to the classification table of typical defect patterns;

[0043] Standardize the defect characteristic quantities in the sample set, and divide the sample set into a training set and a validation set according to a preset ratio;

[0044] Set the hyperparameter range of the integrated learning model based on the LightGBM algorithm, pre-train the integrated learning model based on the LightGBM algorithm using the fuzzy entropy algorithm, and adopt data augmentation technology to simulate different cable detection environments and conditions to find the optimal hyperparameter combination;

[0045] Take the optimal hyperparameter combination as the final model parameters, and retrain the integrated learning model based on the LightGBM algorithm using the training set to obtain a trained integrated learning model.

[0046] Furthermore, it also includes:

[0047] Verify the trained integrated learning model based on the validation set to obtain a verified integrated learning model.

[0048] Furthermore, classify the target high-voltage cable based on the trained integrated learning model, including:

[0049] Judge whether the target high-voltage cable is operating normally: If the cable status predicted from the input defect characteristic quantity shows normal, there is no response; if the cable status predicted from the input defect characteristic quantity shows abnormal, use the trained integrated learning model to classify the target high-voltage cable to obtain a specific defect category recognition result.

[0050] In a second aspect, the present invention also provides a high-voltage cable defect identification system based on cross-connected grounding circulating current, and the system includes:

[0051] A circulating current calculation unit, which is used to construct an equivalent model of the sheath circulating current of a three-phase cable cross-connected grounding system for a single-core cable model with a cross-connected grounding method; based on the equivalent model of the sheath circulating current, calculate the sheath circulating current value;

[0052] The field-circuit coupling unit is used to obtain the metal sheath loss and the internal cable temperature of different types of typical defects based on the field-circuit coupling model of multiple physical fields according to the sheath circulating current value, and construct a classification table of typical defect modes of the cross-bonding system;

[0053] The model construction and training unit is used to establish an integrated learning model based on the LightGBM algorithm, and train the integrated learning model based on the LightGBM algorithm according to the sheath circulating current value and the classification table of typical defect modes to obtain a trained integrated learning model;

[0054] The defect type identification unit is used to calculate the circulating current data of the target high-voltage cable, obtain the target defect feature quantity of the target high-voltage cable, and classify the target high-voltage cable based on the trained integrated learning model to obtain the defect type of the target high-voltage cable.

[0055] Furthermore, the field-circuit coupling unit includes:

[0056] The finite element model establishment subunit is used to establish a finite element calculation model for the single-core cable model and perform mesh division on the finite element calculation model;

[0057] The bidirectional coupling subunit is used to input the sheath circulating current value as an excitation source into the finite element calculation model to obtain the internal conductor conductivity and electromagnetic loss distribution results of the cable; calculate the internal temperature field distribution of the cable according to the electromagnetic loss distribution results; input the internal temperature field distribution of the cable into the finite element calculation model to recalculate the electromagnetic loss distribution;

[0058] The comparison subunit is used to compare the electromagnetic loss distribution results before and after, and if the calculation error of the comparison exceeds the set error range, iterate again until the set error range is satisfied and output the metal sheath loss and the internal cable temperature.

[0059] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0060] 1. The high-voltage cable defect identification method and system based on the cross-bonding grounding circulating current of the present invention. Based on the field-circuit coupling idea, the present invention solves the problem of ignoring the leakage current component by solving the field-circuit coupling model of multiple physical fields; based on the sheath circulating current value and the classification table of typical defect modes, the present invention considers the influence of the changes of various parameters of the cross-bonding system. This method significantly improves the calculation accuracy of the system circulating current, meets the circulating current analysis and monitoring requirements of various typical defects; achieves high-precision high-voltage cable defect identification ability, and greatly improves the efficiency and accuracy of defect handling of the high-voltage cable cross-bonding system.

[0061] 2. The present invention provides a calculation method for the cross-connected grounding loop current loss of high-voltage cables based on the field-circuit coupling idea, which has the characteristics of high accuracy, simple operation, good reliability, etc. By performing parameter setting operations in the field-circuit coupling model, it is not necessary to repeatedly model and input, and more complex analysis and judgment can be carried out.

[0062] 3. The present invention constructs a typical defect recognition model based on the LightGBM algorithm (i.e., an ensemble learning model based on the LightGBM algorithm) for detecting the operating state of high-voltage cables with cross-connected grounding. The LightGBM algorithm shows better classification and recognition effects in this field, can efficiently analyze multi-dimensional feature data, and has distributed support, which has great advantages in dealing with multi-classification problems of large-capacity data samples.

[0063] 4. The technical method of the present invention fills the technical gaps at home and abroad in the industry: At present, for the typical defect recognition methods of high-voltage cables with cross-connected grounding, most focus on the online monitoring method to judge the operating state of the cables, and the online monitoring systems mainly include temperature online monitoring, partial discharge online monitoring, and sheath circulating current online monitoring. However, the current online monitoring methods are mostly limited to over-limit threshold judgment and local defect recognition, which limits their ability to determine the exact location of defects and quickly identify defect types. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, form a part of this application, and do not constitute a limitation to the embodiments of the present invention. In the drawings:

[0065] Figure 1 is the flow chart of the high-voltage cable defect recognition method based on cross-connected grounding loop current of the present invention;

[0066] Figure 2 is the schematic diagram of the sheath circulating current equivalent model of the three-phase cable cross-connected grounding system of the present invention;

[0067] Figure 3 is the structural block diagram of the high-voltage cable defect recognition system based on cross-connected grounding loop current of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0068] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in combination with the embodiments and the drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and do not constitute a limitation to the present invention.

[0069] The prior art mainly judges the operating state of cables by analyzing the sheath circulating current, but mostly only analyzes the calculation method of the induced current component in the sheath circulating current, directly ignores the leakage current component, and does not consider the influence of the change of each parameter of the cross-bonding system, resulting in problems such as low accuracy of high-voltage cable defect identification, low efficiency and low accuracy. Therefore, the present invention designs a high-voltage cable defect identification method and system based on the cross-bonding grounding circulating current. Based on the field-circuit coupling idea, the present invention solves the field-circuit coupling model of multiple physical fields, and solves the problem of ignoring the leakage current component; based on the sheath circulating current value and the typical defect mode classification table, the present invention considers the influence of the change of each parameter of the cross-bonding system, achieves high-precision high-voltage cable defect identification ability, and greatly improves the efficiency and accuracy of defect handling of the high-voltage cable cross-bonding system.

[0070] Embodiment 1

[0071] As Figure 1 shown, the high-voltage cable defect identification method based on the cross-bonding grounding circulating current of the present invention includes:

[0072] S1. For the single-core cable model with cross-bonding grounding method, construct the equivalent model of the sheath circulating current of the three-phase cable cross-bonding grounding system; based on the equivalent model of the sheath circulating current, calculate the value of the sheath circulating current;

[0073] S2. According to the value of the sheath circulating current, based on the field-circuit coupling model of multiple physical fields, obtain the metal sheath loss and the internal temperature of the cable of different types of typical defects, and construct a typical defect mode classification table of the cross-bonding system;

[0074] S3. Establish an integrated learning model based on the LightGBM algorithm, and train the integrated learning model based on the LightGBM algorithm according to the value of the sheath circulating current and the typical defect mode classification table to obtain a trained integrated learning model;

[0075] S4. Calculate the circulating current data of the target high-voltage cable, obtain the target defect feature quantity of the target high-voltage cable, and classify the target high-voltage cable based on the trained integrated learning model to obtain the defect type of the target high-voltage cable.

[0076] For the above technical solutions, considering that the prior art uses the circuit method and only analyzes the calculation method of the induced current component in the sheath circulating current, while directly ignoring the leakage current component; therefore, the present invention is based on the field-circuit coupling idea, solves the field-circuit coupling model of multiple physical fields, and solves the problem of ignoring the leakage current component. Based on the sheath circulating current value and the typical defect mode classification table, the present invention considers the influence of the change of each parameter of the cross-bonding system, achieves high-precision high-voltage cable defect identification ability, and greatly improves the efficiency and accuracy of defect handling of the high-voltage cable cross-bonding system.

[0077] In this embodiment, step S1 specifically includes:

[0078] S1.1. Based on the three-phase cross-bonding grounding method, each large section (i.e., the cable part) in the single-core cable model is divided into three equal-length sections. Each section is grounded through the sheath protector in the transposition box, and the two terminal metal sheaths of each cross-bonding section are directly grounded;

[0079] Among them, the three-phase high-voltage cable is divided into three phases: a, b, and c. The a-phase cable is divided into three sections: a1, a2, and a3. The b-phase cable is divided into three sections: b1, b2, and b3. The c-phase cable is divided into three sections: c1, c2, and c3.

[0080] S1.2. According to the equivalent circuit diagram of the single-core cable, the loop current equation for the sheath circulating current is obtained. The loop current equation is specifically:

[0081]

[0082] In the formula, are the induced voltages generated by the core conductor currents of the a1, a2, and a3 sections of the a-phase cable respectively, is the induced voltage generated by the core conductor currents of the b1, b2, and b3 sections of the b-phase cable, is the induced voltage generated by the core conductor currents of the c1, c2, and c3 sections of the c-phase cable; is the in-ground circulating current, and R eq is the equivalent resistance value; is the induced voltage on the i-th land cable metal sheath of the a-phase cable generated by the circulating currents of the b and c phase metal sheaths, is the induced voltage on the i-th land cable metal sheath of the b-phase cable generated by the circulating currents of the a and c phase metal sheaths, is the induced voltage on the i-th land cable metal sheath of the c-phase cable generated by the circulating currents of the a and b phase metal sheaths; is the grounding circulating current of the a-phase cable, is the grounding circulating current of the b-phase cable, is the grounding circulating current of the c-phase cable; Z ai is the self-impedance of the i-th metal sheath of the a-phase cable, Z bi is the self-impedance of the i-th metal sheath of the b-phase cable, Z ci is the self-impedance of the i-th metal sheath of the c-phase cable.

[0083] Among them, the expressions of the self-impedances Z ai 、Z bi 、Z ci of the i-th metal sheaths of the a-phase cable, b-phase cable, and c-phase cable are specifically:

[0084]

[0085] In the formula, R ai is the sheath resistance of the i-th section of the a-phase cable, R bi is the sheath resistance of the i-th section of the b-phase cable, R ci is the sheath resistance of the i-th section of the c-phase cable; L aai is the self-inductance per unit length of the loop formed by the metal sheath of the i-th section of the a-phase cable and the earth, L bbi is the self-inductance per unit length of the loop formed by the metal sheath of the i-th section of the b-phase cable and the earth, L cci is the self-inductance per unit length of the loop formed by the metal sheath of the i-th section of the c-phase cable and the earth; l i is the length of the i-th section of the cable line.

[0086] Among them, the induced voltage generated by the core conductor current of the i-th section of the a, b, and c phase cables

[0087]

[0088] In the formula, M Ab is the mutual inductance per unit between the core of the a-phase cable and the sheath of the b-phase cable, M Ac is the mutual inductance per unit between the core of the a-phase cable and the sheath of the c-phase cable; M Ba is the mutual inductance per unit between the core of the b-phase cable and the sheath of the a-phase cable, M Bc are respectively the mutual inductance per unit between the core of the b-phase cable and the sheath of the c-phase cable; M Ca is the mutual inductance per unit between the core of the c-phase cable and the sheath of the a-phase cable, M Cb are respectively the mutual inductance per unit between the core of the c-phase cable and the sheath of the b-phase cable; M Aa is the mutual inductance per unit between the core of the a-phase cable and the sheath of the a-phase cable, M Bb is the mutual inductance per unit between the core of the b-phase cable and the sheath of the b-phase cable, M Cc is the mutual inductance per unit between the core of the c-phase cable and the sheath of the c-phase cable; is the ground circulation of the a-phase cable, is the ground circulation of the b-phase cable, is the ground circulation of the c-phase cable.

[0089] Among them, the induced voltage generated on the metal sheath of the i-th section of the a, b, and c phase cables by the circulating current of the metal sheaths of the other two phases

[0090]

[0091] In the formula, M abiThe mutual inductance per unit length of the loop formed by the sheath of the i-th section of the phase-a and phase-b cables and the earth, M aci The mutual inductance per unit length of the loop formed by the sheath of the i-th section of the phase-a and phase-c cables and the earth, M bci The mutual inductance per unit length of the loop formed by the sheath of the i-th section of the phase-b and phase-c cables and the earth; The ground circulating current of the phase-a cable, The ground circulating current of the phase-b cable, The ground circulating current of the phase-c cable; l i The length of the i-th section of the cable line.

[0092] S1.3. Substitute formulas (2) to (4) into the loop equations of formula (1), derive the ground circulating current matrix equation, and use the ground circulating current matrix equation as the sheath circulating current equivalent model; based on the sheath circulating current equivalent model, calculate the value of the sheath circulating current.

[0093] The expression of the sheath circulating current equivalent model is:

[0094]

[0095] In the formula, The ground circulating current of the phase-a cable, The ground circulating current of the phase-b cable, The ground circulating current of the phase-c cable; The induced voltages generated by the core conductor currents of the a1, a2, and a3 sections of the cable respectively, The induced voltages generated by the core conductor currents of the b1, b2, and b3 sections of the cable, The induced voltages generated by the core conductor currents of the c1, c2, and c3 sections of the cable; Z 11 To Z 33 Are impedances;

[0096] Among them, the impedance Z 11 To Z 33 The calculation formula is:

[0097]

[0098] As Figure 2 Shown, Figure 2 Is the sheath circulating current equivalent model of a three-phase cable cross-connected grounding system. Figure 2 In, The ground circulating current of the phase-a cable, The ground circulating current of the phase-b cable, The ground circulating current of the phase-c cable; The in-ground current, R eq Is the equivalent resistance value, It represents the induced voltage generated by the current of the i-th segment of the core conductor. It represents the induced voltage on the i-th segment of the land cable metal sheath generated by the circulating current of the metal sheaths of the other two phases; Z ai 、Z bi 、Z ci respectively represent the self-impedances of the i-th segment of the metal sheaths of the a-phase cable, b-phase cable, and c-phase cable.

[0099] In this embodiment, step S2 specifically includes:

[0100] S2.1. Establish a finite element calculation model for the single-core cable model and perform mesh division on the finite element calculation model;

[0101] S2.2. Input the sheath circulating current value as the excitation source into the finite element calculation model to obtain the results of the conductivity and electromagnetic loss distribution of the internal conductor of the cable; calculate the internal temperature field distribution of the cable according to the electromagnetic loss distribution results;

[0102] S2.3. Input the internal temperature field distribution of the cable into the finite element calculation model to recalculate the electromagnetic loss distribution; compare the electromagnetic loss distribution results of the two times before and after, and if the calculation error of the comparison exceeds the set error range, perform cyclic iteration until the set error range is satisfied and output the metal sheath loss and the internal temperature of the cable;

[0103] S2.4. Use the metal sheath loss and the internal temperature of the cable as defect characteristic quantities to encode typical defects of different types of cross-bonding systems to obtain defect type codes; construct a typical defect mode classification table of the cross-bonding system according to the defect characteristic quantities and defect type codes, as shown in Table 1:

[0104] Table 1 Typical Defect Mode Classification Table

[0105]

[0106] In the above technical solution, the present invention combines the bidirectional coupling idea of field and circuit. First, input the sheath circulating current value as the excitation source into the finite element calculation model to obtain the results of the conductivity and electromagnetic loss distribution of the internal conductor of the cable; calculate the internal temperature field distribution of the cable according to the electromagnetic loss distribution results; then input the internal temperature field distribution of the cable into the finite element calculation model to recalculate the electromagnetic loss distribution; compare the electromagnetic loss distribution results of the two times before and after until the calculation error of the comparison meets the set error range to obtain the metal sheath loss and the internal temperature of the cable. The present invention overcomes the influence of directly ignoring the leakage current component in the original circuit method.

[0107] In this embodiment, in step S3, according to the sheath circulating current value and the typical defect mode classification table, the integrated learning model based on the LightGBM algorithm is trained to obtain a trained integrated learning model, which specifically includes:

[0108] S3.1. According to the typical defect mode classification table, that is, the defect feature quantity and its corresponding defect category code are used together as the sample set for defect classification training and testing; in order to reduce the influence of data differences on the diagnosis effect, the defect feature quantity in the sample set is standardized, and the sample set is split into a training set and a validation set according to a preset ratio;

[0109] S3.2. Set the hyperparameter range of the integrated learning model based on the LightGBM algorithm, pre-train the integrated learning model based on the LightGBM algorithm using the fuzzy entropy algorithm, and adopt the data augmentation technology to simulate different cable detection environments and conditions to find the optimal hyperparameter combination;

[0110] S3.3. Use the optimal hyperparameter combination as the final model parameter, and retrain the integrated learning model based on the LightGBM algorithm using the training set to obtain a trained integrated learning model;

[0111] S3.4. Based on the validation set, validate the trained integrated learning model to obtain a validated integrated learning model, thereby verifying the accuracy of the model.

[0112] In this embodiment, classifying the target high-voltage cable based on the trained integrated learning model includes:

[0113] Judge whether the target high-voltage cable is operating normally: if the cable status predicted from the input defect feature quantity shows normal, there is no response; if the cable status predicted from the input defect feature quantity shows abnormal, use the trained integrated learning model to classify the target high-voltage cable to obtain the specific defect category recognition result.

[0114] In this embodiment, in step S4, the circulating current data of the target high-voltage cable can be calculated by a method similar to that in step S1.

[0115] The method of the present invention significantly improves the calculation accuracy of the system circulating current and meets the requirements of circulating current analysis and monitoring for various typical defects; the present invention achieves high-precision high-voltage cable defect recognition ability, and also greatly improves the efficiency and accuracy of defect handling in the high-voltage cable cross-bonding system.

[0116] Embodiment 2

[0117] Such as Figure 3As shown in the figure, the difference between this embodiment and Embodiment 1 is that this embodiment provides a high-voltage cable defect identification system based on cross-connected grounding loop current. This system uses the high-voltage cable defect identification method based on cross-connected grounding loop current in Embodiment 1, and the functions of this system and the high-voltage cable defect identification method based on cross-connected grounding loop current in Embodiment 1 correspond one by one. The system includes:

[0118] A loop current calculation unit, which is used to construct an equivalent model of the sheath loop current of a three-phase cable cross-connected grounding system for a single-core cable model with a cross-connected grounding method; and calculate the sheath loop current value based on the equivalent model of the sheath loop current.

[0119] A field-circuit coupling unit, which is used to obtain the metal sheath loss and the internal temperature of the cable for different types of typical defects based on the field-circuit coupling model of multiple physical fields according to the sheath loop current value, and construct a classification table of typical defect modes of the cross-connected system.

[0120] A model construction and training unit, which is used to establish an integrated learning model based on the LightGBM algorithm, and train the integrated learning model based on the LightGBM algorithm according to the sheath loop current value and the classification table of typical defect modes to obtain a trained integrated learning model.

[0121] A defect type identification unit, which is used to calculate the loop current data of the target high-voltage cable, obtain the target defect feature quantity of the target high-voltage cable, and classify the target high-voltage cable based on the trained integrated learning model to obtain the defect type of the target high-voltage cable.

[0122] As a further implementation, the field-circuit coupling unit includes:

[0123] A finite element model establishment sub-unit, which is used to establish a finite element calculation model for the single-core cable model and perform mesh division on the finite element calculation model.

[0124] A bidirectional coupling sub-unit, which is used to input the sheath loop current value as an excitation source into the finite element calculation model to obtain the internal conductor conductivity and electromagnetic loss distribution results of the cable; calculate the internal temperature field distribution of the cable according to the electromagnetic loss distribution results; input the internal temperature field distribution of the cable into the finite element calculation model to recalculate the electromagnetic loss distribution.

[0125] A comparison sub-unit, which is used to compare the electromagnetic loss distribution results before and after, and if the calculation error of the comparison exceeds the set error range, perform iterative calculation until the set error range is satisfied and output the metal sheath loss and the internal temperature of the cable.

[0126] Among them, the execution process of each unit can be carried out according to the process steps of the high-voltage cable defect identification method based on cross-connected grounding loop current in Embodiment 1, and will not be elaborated one by one in this embodiment.

[0127] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0128] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0129] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0130] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are performed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0131] The specific embodiments described above further elaborate on the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A high voltage cable defect identification method based on cross-connected ground loop current, characterized in that: The method includes: For the single-core cable model of the cross-connected grounding method, a sheath circulation equivalent model of the three-phase cable cross-connected grounding system is constructed; based on the sheath circulation equivalent model, the sheath circulation value is calculated; According to the sheath circulation value, based on the field-circuit coupling model of multi-physics fields, the metal sheath loss and the internal temperature of the cable of different types of typical defects are obtained, and a typical defect mode classification table of the cross-connection system is constructed; Establishing an integrated learning model based on the LightGBM algorithm, and training the integrated learning model based on the LightGBM algorithm according to the sheath circulation value and the typical defect mode classification table to obtain a trained integrated learning model; Calculating the circulating current data of the target high-voltage cable, obtaining the target defect characteristic quantity of the target high-voltage cable, and classifying the target high-voltage cable based on the trained integrated learning model to obtain the defect type of the target high-voltage cable; According to the sheath circulation value, based on the field-circuit coupling model of multi-physics fields, the metal sheath loss and cable internal temperature of different types of typical defects are obtained, including: Establishing a finite element calculation model for the single-core cable model, and performing meshing on the finite element calculation model; Input the sheath circulation value as an excitation source into the finite element calculation model to obtain the internal conductor conductivity and electromagnetic loss distribution results of the cable; calculate the internal temperature field distribution of the cable according to the electromagnetic loss distribution results; The temperature field distribution inside the cable is input into the finite element calculation model, and the electromagnetic loss distribution is recalculated; and the electromagnetic loss distribution results before and after are compared. If the calculation error of the comparison exceeds the set error range, the iteration is repeated until the set error range is met and the metal sheath loss and the internal temperature of the cable are output.

2. The high voltage cable defect identification method based on cross-connected ground loop current according to claim 1 is characterized in that: According to the single-core cable model of cross-connected grounding, the sheath circulation equivalent model of the three-phase cable cross-connected grounding system is constructed; Based on the sheath circulation equivalent model, the sheath circulation value is calculated, including: Based on the three-phase cross-interconnection grounding method, each cable part in the single-core cable model is divided into three sections of equal length. Each section is grounded through the sheath protector in the transposition box, and the two terminal metal sheaths of each cross-interconnection section are directly grounded. According to the equivalent circuit diagram of the single-core cable, the loop current equation about the sheath circulating current is obtained; Substitute the self-impedance of the i-th section of the metal sheath, the induced voltage and the induced voltage formed by the metal sheath circulating current of the remaining two phases into the loop current equation, derive the grounding circulation current matrix equation, and use the grounding circulation current matrix equation as the sheath circulating current equivalent model; Based on the sheath circulation equivalent model, the sheath circulation value is calculated.

3. The high voltage cable defect identification method based on cross-connected ground loop current according to claim 2 is characterized in that: The expression of the sheath circulation equivalent model is: In the formula, is the ground loop current of phase a cable, is the ground loop current of phase b cable, is the ground loop current of phase c cable; They are the induced voltages generated by the currents in the cable core conductors of sections a1, a2, and a3, respectively. is the induced voltage generated by the current in the cable core conductors of sections b1, b2, and b3. is the induced voltage generated by the current in the cable core conductors of sections c1, c2, and c3; Z 11 To Z 33 For impedance.

4. The high voltage cable defect identification method based on cross-connected ground loop current according to claim 1 is characterized in that: Construct a taxonomy of typical defect modes for cross-connect systems, including: Taking the metal sheath loss and the internal temperature of the cable as defect characteristic quantities, the typical defects of different types of cross-connection systems are coded to obtain the defect type coding; According to the defect characteristic quantity and the defect type code, a typical defect mode classification table of the cross-connection system is constructed.

5. The high voltage cable defect identification method based on cross-connected ground loop current according to claim 1 is characterized in that: According to the sheath circulation value and the typical defect mode classification table, the integrated learning model based on the LightGBM algorithm is trained to obtain a trained integrated learning model, including: According to the typical defect mode classification table, a sample set for defect classification training and testing is constructed; Standardizing the defect feature quantities in the sample set, and dividing the sample set into a training set and a validation set according to a preset ratio; Setting the hyperparameter range of the integrated learning model based on the LightGBM algorithm, pre-training the integrated learning model based on the LightGBM algorithm using the fuzzy entropy algorithm, and using data augmentation technology to simulate different cable detection environments and conditions to find the optimal hyperparameter combination; The optimal hyperparameter combination is used as the final model parameters, and the training set is reused to train the integrated learning model based on the LightGBM algorithm to obtain a trained integrated learning model.

6. The high voltage cable defect identification method based on cross-connected ground loop current according to claim 1 is characterized in that: Also includes: Based on the validation set, the trained ensemble learning model is validated to obtain a validated ensemble learning model.

7. The high voltage cable defect identification method based on cross-connected ground loop current according to claim 1 is characterized in that: Classifying the target high-voltage cable based on the trained integrated learning model includes: It is judged whether the target high-voltage cable is operating normally: if the cable status is predicted to be normal by the input defect feature quantity, there is no response; if the cable status is predicted to be abnormal by the input defect feature quantity, the trained ensemble learning model is used to classify the target high-voltage cable to obtain a specific defect category recognition result.

8. A high voltage cable defect identification system based on cross-connected ground loop current, characterized in that: The system includes: A circulating current calculation unit is used to construct a sheath circulating current equivalent model of a three-phase cable cross-connected grounding system for a single-core cable model of a cross-connected grounding mode; and calculate a sheath circulating current value based on the sheath circulating current equivalent model; A field-circuit coupling unit is used to obtain the metal sheath loss and the internal temperature of the cable of different types of typical defects according to the sheath circulating current value and based on the field-circuit coupling model of multi-physical fields, and to construct a typical defect mode classification table of the cross-connection system; A model building and training unit, used for establishing an integrated learning model based on the LightGBM algorithm, and training the integrated learning model based on the LightGBM algorithm according to the sheath circulation value and the typical defect mode classification table to obtain a trained integrated learning model; A defect type identification unit, used to calculate the circulating current data of the target high-voltage cable, obtain the target defect characteristic quantity of the target high-voltage cable, and classify the target high-voltage cable based on the trained integrated learning model to obtain the defect type of the target high-voltage cable; The field-circuit coupling unit comprises: A finite element model building subunit is used to build a finite element calculation model for the single-core cable model and to perform meshing on the finite element calculation model; A bidirectional coupling subunit is used to input the sheath circulating current value as an excitation source into the finite element calculation model to obtain the conductivity of the cable internal conductor and the electromagnetic loss distribution results; calculate the temperature field distribution inside the cable according to the electromagnetic loss distribution results; input the temperature field distribution inside the cable into the finite element calculation model to recalculate the electromagnetic loss distribution; The comparison subunit is used to compare the electromagnetic loss distribution results of the two times before and after. If the calculation error of the comparison exceeds the set error range, it will be iterated again until the set error range is met and the metal sheath loss and the internal temperature of the cable are output.

Images