A method, recording medium and system for diagnosing defects and moisture conditions of xlpe cable insulation

Abstract

The application belongs to the technical field of cable defect diagnosis, and particularly relates to a method for diagnosing XLPE cable insulation defects and moisture state, which is modeled through a bidirectional neural network algorithm, in which the partial discharge state of the cable is formed into an array in a quantitative manner and added to the model as a variable parameter, so that the technical problem that the traditional cable defect detection method has a large error with the actual situation and is difficult to correct is solved. The application further provides a non-transient readable recording medium storing a program of the method and a system comprising the medium, and the program can be called by a processing circuit to execute the above method.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of cable defect diagnosis, and discloses a method for diagnosing insulation defects and moisture state of XLPE cable, a recording medium storing a program capable of executing the method, and a system. BACKGROUND

[0002] Cross-linked polyethylene (XLPE) cables are widely used in the field of high-voltage power transmission. Operating experience shows that power cables laid in rainy and humid areas often suffer from insulation performance degradation due to moisture, and these hidden dangers will accumulate and develop to a certain extent, which will quickly lead to cable failure. Due to the relatively complex installation and laying process of XLPE cables and the harsh operating environment, different degrees of moisture may occur at different positions of the cable, causing different degrees of partial discharge. The partial discharge phenomenon at the moisture site contains defect type information of the fault site, and therefore partial discharge measurement is considered an effective means to detect and identify the moisture level of XLPE cables.

[0003] However, cable moisture or insulation defects do not necessarily cause a certain degree of partial discharge. There are many ways to measure the degree of moisture, and although severe moisture may have a higher probability of causing insulation defects, it is not necessarily positively correlated with insulation defects. It is difficult for those skilled in the art to quantitatively correlate the complex relationship between cable moisture, insulation defects, and partial discharge conditions, so the traditional cable defect detection method has a large error compared to the actual situation and is difficult to correct. SUMMARY

[0004] To solve the above problems, the present application provides a method for diagnosing insulation defects and moisture state of XLPE cable, which models through a bidirectional neural network algorithm. In the modeling, the partial discharge state of the cable is formed into an array in a quantitative manner and added as a variable parameter to the model, thereby solving the technical problem that the traditional cable defect detection method has a large error compared to the actual situation and is difficult to correct.

[0005] The specific scheme includes the following steps:

[0006] S1. Measuring a set of moisture parameters of the same type and length of XLPE cable with different moisture levels and insulation defect types;

[0007] S2. Measuring a set of partial discharge parameters of the same type and length of XLPE cable with different moisture levels and insulation defect types;

[0008] S3. Establishing a fingerprint map of the partial discharge parameters and moisture parameters under different moisture levels and insulation defect types;

[0009] S4. Constructing a model using a bidirectional recurrent neural network according to the fingerprint map;

[0010] S5. Substituting the measured partial discharge value and the moisture parameter value into the model, identifying the insulation defect type of the cable and evaluating the moisture degree thereof.

[0011] The above technical solution incorporates the quantified partial discharge factor into the modeling category, and constructs a model through a bidirectional neural network. Such a model composed of multiple arrays can fit the originally uncertain complex correlation relationship, so that the accuracy of diagnosis is greatly improved.

[0012] Preferably, the partial discharge parameters include discharge frequency and discharge cumulative energy.

[0013] The two factors can simply summarize the degree of partial discharge.

[0014] Further, the moisture parameters include water content, dielectric loss and conductivity.

[0015] The three parameters of water content, dielectric loss and conductivity are quantities used to measure the moisture degree in traditional methods. Here, they are combined into the model, together with the partial discharge parameters, to exert a combined influence on the establishment of the model, so that the model is closer to reality.

[0016] Further, the insulation defect type is divided into five categories, and the step S5 further includes a step of measuring the diagnosis accuracy. The recognition rate is used as a measurement standard for the identified insulation defect type, and the error is evaluated by introducing a numerical integral calculation for the moisture degree, so as to obtain the following objective function:

[0017]

[0018] Wherein, NFi is the number of defects whose defect type is identified as Fi in the sequence F; NT is the length of the sequence F; is a phase number sequence converted into a five-bit binary hot code; E is a unit matrix of 5*5 specification; k is the recognition rate of the defect type; p is the confidence probability of the judgment result.

[0019] Such a setting can reduce the amount of calculation and the occupation of calculation resources on the premise of ensuring the accuracy requirement, and can also make scientific correction to the model through the evaluation error.

[0020] Another scheme of the present application is to provide a non-transitory readable recording medium for storing one or more programs containing a plurality of instructions, which when executed, will cause the processing circuit to perform the above-mentioned XLPE cable insulation defect and moisture state diagnosis method.

[0021] Still another aspect of the present application provides a system for diagnosing XLPE cable insulation defects and moisture state, comprising a processing circuit and a memory electrically coupled to the processing circuit, the memory is configured to store at least one program, the program comprises a plurality of instructions, and the processing circuit executes the program to perform the method for diagnosing XLPE cable insulation defects and moisture state. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 A schematic diagram of a bidirectional neural network algorithm in the present application;

[0023] Figure 2 A schematic diagram of measurement and data acquisition of partial discharge state in an embodiment of the present application;

[0024] Wherein, 1. voltage regulating unit; 2. acquisition unit; 3. measurement unit; 4. identification unit; 5. voltage regulating resistor; 6. no halo test transformer, 7. protection resistor, 8. capacitor voltage divider; 9. coupling capacitor; 10. measurement resistor; 11. oscilloscope; 12. synchronization control circuit. DETAILED DESCRIPTION

[0025] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be described below with reference to the drawings of the embodiments of the present application. The described embodiments are part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making innovative labor fall within the scope of protection of the present application.

[0026] As shown in Figure 1 The modeling in the present embodiment adopts the algorithm structure of bidirectional recurrent neural networks (BRNN for short). The BRNN algorithm is a new type of neural network form extended from the one-way neural network algorithm. Unlike the one-way neural network algorithm which only focuses on the forward transmission of data, the BRNN structure is composed of two opposite neural networks. The BRNN algorithm allows internal recurrent connections, and these repeated connections temporarily store the previous input data in the internal network, thereby affecting the output of the neural network, enabling the backward propagation of information flow, i.e. paying attention to the influence of previous and subsequent inputs on the neural network algorithm.

[0027] Prepare multiple groups of the same type of fixed-length XLPE power cables, each group of cable is 2 meters long, and the user can switch the cable sample to be detected through the cable switching button of the upper computer. After the button is started, the switching switch acts to switch the cable sample to be detected.

[0028] The RFID receiver of the upper computer is matched with the sheet-shaped sensor to realize the measurement of the moisture content, dielectric loss, conductivity and other parameters reflecting the moisture degree of the cable, and a set of moisture parameters is obtained; the effective detection distance of the sheet-shaped sensor is greater than or equal to 2.0 m, the temperature detection range is -40 DEG C to 100 DEG C, the moisture content detection range is 25% to 98%, the gain of the RFID receiver is 9dBi, the frequency range is 840 / 940MHz, and the impedance is 50Ω, so that the detection efficiency is improved.

[0029] As shown in Figure 2 The partial discharge measurement part includes a voltage regulating unit 1, an acquisition unit 2 and a measurement unit 3. The voltage regulating unit 1 includes a voltage regulating resistor 5, a no corona test transformer 6, a protection resistor 7 and a capacitor divider 8. The AC voltage generated by the voltage regulating resistor 5 and the no corona test transformer 6 provides voltage for the sample to be tested in the acquisition unit 2 after passing through the protection resistor 7 and the capacitor divider 8. The protection resistor 7 limits the current through the no corona test transformer 6 when the sample is broken down, avoiding the impact on it. The capacitor divider 8 monitors the voltage value on the output side of the no corona test transformer 6 in real time and inputs it into the measurement unit 3 as the phase reference of the discharge signal.

[0030] The measurement unit 3 includes a coupling capacitor 9, a measurement resistor 10, an oscilloscope 11 and a synchronous control circuit 12. The coupling capacitor 9 provides a path for the measurement resistor 10, which is a pulse current test resistor. The pulse current method is used to couple the pulse current to the measurement resistor 10, and then the partial discharge information is displayed on the oscilloscope 11. The Tektronix DPO 7000C oscilloscope with FastFrame function is selected. In addition to the pulse current method, the measurement unit 3 can also match high frequency method, ultra high frequency method, optical fiber method and other partial discharge measurement methods to detect the partial discharge of the sample to be tested, and the obtained data set of discharge frequency and discharge cumulative energy is transmitted to the identification unit 4.

[0031] The identification unit 4 uses the moisture degree detection and partial discharge measurement data set in the early database to establish a fingerprint spectrum, wherein the fingerprint parameters include moisture content, dielectric loss, conductivity, discharge frequency and discharge cumulative energy, the insulation defect type is divided into 5 categories from light to heavy, and the moisture degree follows the experience data in the early database. According to the fingerprint spectrum, a model is constructed by using a bidirectional recurrent neural network. Then, the measured partial discharge value and the moisture parameter value are substituted into the model to identify the insulation defect type of the cable and evaluate its moisture degree.

[0032] The recognition rate is used as the measurement standard for the identified insulation defect type, and the error is evaluated by introducing the numerical integral calculation for the moisture degree, so as to obtain the following objective function:

[0033]

[0034] wherein, NFi is the number of defects of the type Fi identified in the sequence F; NT is the length of the sequence F; is the sequence of the stage converted into five-bit binary hot code; E is a 5x5 unit matrix; k is the identification rate of the defect type; p is the confidence probability of the judgment result. If the values of k or p cannot meet the requirements, the division of the insulation defect type or the detection type of the damp data is adjusted, and the model is re-constructed for correction until the detection accuracy meets the requirements.

[0035] Those skilled in the art will appreciate that embodiments of the application can be supplied as methods, systems, or computer program products. Accordingly, the application can be embodied in the form of complete hardware embodiments, complete software embodiments, or embodiments combining software and hardware aspects. Furthermore, the application can be embodied in the form of a computer program product implemented on one or more computers containing computer usable program code, which can be embodied in but is not limited to magnetic storage media, optical storage media, and the like.

[0036] The application is described in reference to the flowcharts and / or block diagrams of the methods, apparatus (systems), and computer program products according to the embodiments of the application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce the functions specified in the flowcharts and / or block diagrams. Figure 1 The flow or flows and / or blocks in the flowcharts and / or block diagrams Figure 1 The means for carrying out the functions specified in the flow or flows or the block or blocks.

[0037] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including instruction means, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The flow or flows and / or blocks in the flowcharts and / or block diagrams Figure 1 The means for carrying out the functions specified in the flow or flows or the block or blocks.

[0038] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide the functions specified in the flowcharts and / or block diagrams. Figure 1 The flow or flows and / or blocks in the flowcharts and / or block diagrams Figure 1steps of the functions specified in one or more blocks.

[0039] The method steps are assembled into a program and stored in a hard disk or other non-transient storage medium, which constitutes the "non-transient readable recording medium" technical solution of the present application; and the storage medium is electrically connected with the computer processor, and through data processing, the diagnosis of the XLPE cable insulation defect and moisture state can be completed, which constitutes the "XLPE cable insulation defect and moisture state diagnosis system" technical solution of the present application.

[0040] Finally, it should be pointed out that: the above only for the preferred embodiments of the present application, and is not used to limit the present application, although the foregoing embodiments of the present application are described in detail, for those skilled in the art, it still can be modified to the technical scheme recorded in the foregoing each embodiment, or equivalent replacement of part of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for diagnosing insulation defects and moisture conditions in XLPE cables, characterized in that... Includes the following steps: S1. A set of moisture parameters for fixed-length XLPE cables of the same model with different degrees of moisture and insulation defect types; S2. Set of partial discharge parameters for fixed-length XLPE cables of the same model with different degrees of moisture and insulation defect types; S3. Establish fingerprints of the partial discharge parameters and moisture parameters under different degrees of moisture and types of insulation defects; S4. A model is constructed using a bidirectional recurrent neural network based on the fingerprint spectrum; S5. Substitute the partial discharge value and moisture parameter value measured on site into the model to identify the type of insulation defect of the cable and assess its degree of moisture. The partial discharge parameters include discharge frequency and cumulative discharge energy; The humidity parameters include moisture content, dielectric loss, and conductivity; The insulation defect types are divided into 5 categories. Step S5 also includes a step to measure the diagnostic accuracy. The identification rate is used as the measurement standard for the identified insulation defect types, and the product calculation is introduced to evaluate the error for the degree of moisture. Thus, the following objective function is obtained: ; where N Fi N represents the number of defects in sequence F whose defect type is identified as Fi; T S is the length of sequence F; Fi is the sequence of stages converted into a five-bit binary hot unique code; E is a 5×5 identity matrix; k is the defect type recognition rate; p is the confidence probability of defect type determination.

2. A non-transitory readable recording medium for storing one or more programs containing multiple instructions, characterized in that, When the instruction is executed, the processing circuit will perform the method for diagnosing insulation defects and moisture conditions of an XLPE cable as described in claim 1.

3. A system for diagnosing insulation defects and moisture conditions in XLPE cables, comprising a processing circuit and a memory electrically coupled thereto, characterized in that, The memory is configured to store at least one program, the program containing multiple instructions, and the processing circuit runs the program to perform the method for diagnosing insulation defects and moisture conditions of XLPE cables as described in claim 1.

Images