A cable joint multi-cause short circuit fault simulation and early identification system and method

Abstract

A cable joint multi-inducing short circuit fault simulation and early identification system and method, the system comprises a cable joint fault simulation module, which is used for configuring different cable structures and material parameters, constructing an adjustable air gap defect to simulate different cable joint fault conditions; a multi-inducing simulation module is used for applying humidity, temperature, pressure and ultraviolet aging inducement; a data acquisition module is used for collecting pyrolysis particle data generated by the main insulation layer under the action of power-on and multi-inducing; a control module is used for providing a man-machine interface, adjusting the operating state of the multi-inducing simulation module, and controlling the on-off and power-on current value of the test main loop; an early identification module is used for constructing an early identification model according to the pyrolysis particle characteristics and corresponding fault condition parameters, early identifying potential faults and generating an explanation. The present application realizes the insulation degradation simulation and short circuit risk early identification of the cable joint under multi-inducing, and is suitable for power system fault evolution analysis and diagnosis training.

Description

Technical Field

[0001] This invention belongs to the field of cable fault simulation and identification technology, and in particular relates to a system and method for simulating and proactively identifying multi-cause short-circuit faults in cable joints. Background Technology

[0002] Cable joints, as critical components in power cable systems, directly impact the reliability and safety of the entire power transmission and distribution system. With the large-scale construction of urban underground cable lines, the number of cable joints has increased dramatically, and their failures have become one of the main causes of power outages. Cable joint failures are often triggered by a combination of factors, including excessively high ambient humidity, abnormally high temperatures, abnormal external compression, and material aging caused by long-term operation or ultraviolet radiation. These factors often couple together, leading to deterioration of the internal insulation medium of the joint, increased partial discharge, and ultimately, a short-circuit fault.

[0003] Existing cable fault simulation devices mainly employ centralized or fixed structures, typically simulating single fault types, making it difficult to realistically reproduce the fault evolution process under the coupled effects of multiple contributing factors. Furthermore, existing fault identification methods largely rely on post-event analysis, lacking the ability to effectively capture early signs of faults and provide proactive warnings. In addition, the test cable structures in existing simulation devices are mostly designed with fixed parameters, making it difficult to quickly configure cable joints with different structural and material parameters, and also difficult to quickly adjust cable joint defect parameters, thus limiting the versatility and reusability of the devices.

[0004] To address the aforementioned issues, while existing technologies have proposed some cable fault simulation systems, they primarily focus on simulating faults within the cable itself, and the fault types are relatively limited, failing to effectively simulate the multi-causal coupled short-circuit faults unique to cable joints. Other systems introduce the fusion of infrared thermography and acoustic vibration data to predict abnormal states of cable joints, but these are only applicable to scenarios after the abnormal state occurs and are not suitable for the early identification of short-circuit faults in cable joints, nor do they involve the collection and analysis of early features such as pyrolysis particles. Therefore, this invention urgently needs to design a system capable of simultaneously simulating multi-causal coupled faults, flexibly configuring the structural and material parameters and internal defect parameters of cable joints, and performing early identification based on pyrolysis particle data, to meet the needs of power system fault mechanism research, maintenance training, and early warning. Summary of the Invention

[0005] This invention aims to address the problems of existing technologies in simulating cable joint faults, such as the reliance on single-factor simulations, the inability to simulate multiple contributing factors, inconvenient configuration of cable structure and material parameters, and insufficient accuracy in advanced identification. It provides a system and method for simulating and advancedly identifying multi-factor short-circuit faults in cable joints. This system simulates internal air gap defects and their parameter changes by incorporating adjustable air gap defects within the main insulation layer; integrates a multi-factor simulation module to simulate contributing factors such as humidity, temperature, pressure, and UV aging; supports configuration of different conductor types, insulation materials, and sheath types, and allows adjustment of the current value; and constructs an advanced identification model by collecting pyrolysis particle data and correlating it with corresponding fault condition parameters, achieving advanced fault identification and interpretation.

[0006] The technical solution of this invention is:

[0007] The present invention provides a multi-cause short-circuit fault simulation and advance identification system for cable joints, characterized in that it comprises:

[0008] The cable joint fault simulation module is placed in a sealed test chamber and includes a cable joint body and an insulation defect parameter adjustment unit. The cable joint body includes several independently arranged phase units, each phase unit including a conductive core and a main insulation layer covering the conductive core. The cable joint fault simulation module is used to form different cable joint fault simulation conditions by configuring different cable structure and material parameters and constructing adjustable air gap defects with different parameters in the main insulation layer through the insulation defect parameter adjustment unit.

[0009] A multi-inducement simulation module is used to apply various external inducements to the cable joint body inside the sealed test chamber;

[0010] The data acquisition module is used to collect data on pyrolysis particles generated by the cable connector body under the action of power and external inducing factors.

[0011] The advanced identification module is used to construct an advanced identification model by using the pyrolysis particle features extracted from the pyrolysis particle data and the fault condition parameters corresponding to the fault simulation conditions as model input features, to perform advanced identification of potential short-circuit faults and generate explanations, and to output risk prediction results and corresponding explanations.

[0012] The control module is used to provide a human-machine interface, receive user input of fault simulation condition configuration instructions, adjust the operating status of the multi-cause simulation module, control the on / off state and current value of the test main circuit, and also record and display the pyrolysis particle data, fault condition parameters and risk prediction results.

[0013] Furthermore, the insulation defect parameter adjustment unit includes:

[0014] An air gap length adjustment unit includes an insulating slider and a first driving unit linked thereto. The insulating slider is disposed in a preset air gap channel within the main insulating layer, and the air gap channel extends radially along the main insulating layer. The outer surface of the insulating slider is provided with an external thread, and the opening section of the air gap channel or a mating part disposed within the air gap channel is provided with an internal thread that mates with the external thread. The first driving unit includes a first knob connected to the insulating slider. When the first knob is rotated, the insulating slider rotates in or out along the air gap channel, thereby controlling the depth to which the insulating slider extends into the preset air gap and changing the length of the air section not occupied by the insulating slider within the air gap channel.

[0015] The air gap radius adjustment unit includes several replaceable insulating sleeves. The insulating sleeves are set in the air gap channel. Their outer diameter is the same as the diameter of the air gap channel, and their inner diameter is one of various specifications. The radial cross-sectional size of the air gap can be changed by replacing the insulating sleeves with different inner diameters.

[0016] The defect position adjustment unit includes a rotating sleeve, a detachable rotor flange, a pin seat, a positioning pin, and a phase retaining component. The rotating sleeve is fitted and fixed to the outside of the main insulation layer of the target phase unit, for example, directly fixed to the outside of the main insulation layer, or fixed to the outside of the outer sheath layer. The detachable rotor flange is detachably connected to the rotating sleeve and is used to drive the main insulation layer and conductive core of the target phase unit to rotate synchronously around an axis. The detachable rotor flange has multiple positioning holes along its circumference, and the positioning pin can pass through the pin seat and be inserted into any of the positioning holes to limit the adjustment angle of the target phase unit. The phase retaining component is used to clamp and fix the insulating outer surface of the target phase unit after the angle adjustment is completed, so that after the detachable rotor flange, positioning pin, and temporary connection structure are removed, the target phase unit still maintains a preset circumferential position and allows its conductive core end to be connected to the test main circuit.

[0017] Furthermore, an outer sheath layer can be provided outside the main insulation layer; the cable joint fault simulation module also includes a positioning support, which has mounting holes for installing each phase unit, and each phase unit is detachably connected to the corresponding mounting hole; the structural and material parameters include one or more of the conductor type, main insulation material, and sheath type, and each phase unit can be installed in the corresponding mounting hole of the positioning support and can be removed from the positioning support to facilitate the replacement of phase units with different conductor types, main insulation materials, and sheath types; each phase unit has terminals at both ends of its conductive core for connecting to the test main circuit.

[0018] Furthermore, the multi-cause simulation module includes:

[0019] The humidity simulation unit includes a spray device, a humidity sensor, and a relay. The spray device is used to spray humidification into the sealed test chamber. The humidity sensor is used to detect the humidity value in the chamber and feed it back to the control module. The relay is controlled by the control module to control the start and stop of the spray device.

[0020] The temperature simulation unit includes a heater, a cooler, a temperature sensor, and a relay; the heater and cooler are used to regulate the temperature inside the sealed test chamber, the temperature sensor is used to detect the temperature value inside the chamber and feed it back to the control module, and the relay is controlled by the control module to control the start and stop of the heater and cooler;

[0021] The pressure simulation unit includes a pressure block, a second drive unit, a stress sensor, and a detachable limiting support frame. The second drive unit includes a pressure rod for moving the pressure block up and down. The detachable limiting support frame is disposed within the sealed test chamber and has a guide structure that cooperates with the pressure rod to limit the vertical movement of the pressure rod and prevent the pressure block from swaying. The detachable limiting support frame also provides auxiliary support for each phase unit. The second drive unit drives the pressure block to move toward the main insulation layer to change the pressure on the main insulation layer. The stress sensor detects the pressure value and feeds it back to the control module.

[0022] The ultraviolet aging simulation unit includes an ultraviolet lamp and a relay; the ultraviolet lamp is set in a sealed test chamber and is used to irradiate the cable joint body with ultraviolet light; the relay is controlled by a control module to control the start and stop of the ultraviolet lamp, and the control module is used to record the ultraviolet irradiation time of the ultraviolet lamp, and simulate different degrees of ultraviolet aging of the cable joint body by changing the ultraviolet irradiation time.

[0023] Furthermore, the data acquisition module includes a pyrolysis particle collector and a sampling circuit; the output of the pyrolysis particle collector is connected to the control module and the advanced identification module via the sampling circuit; the main test circuit, composed of the test power supply, connecting wires, and conductive cores of each phase unit, is equipped with an on / off switch, a current regulator, and a current sensor. The on / off switch is used to control the on / off of the main circuit, the current regulator is controlled by the control module and is used to adjust the energized current value of the cable joint body, and the current sensor is used to detect the energized current value and feed it back to the control module; the control module includes a microcontroller and a computer, and the microcontroller is electrically connected to the cable joint fault simulation module, the multi-cause simulation module, the data acquisition module, and the advanced identification module, respectively.

[0024] Furthermore, the advanced identification module includes:

[0025] An advanced identification model is used to receive the input features of the model and establish a risk identification relationship for judging low-risk, medium-risk, or high-risk based on the correspondence between fault condition parameters, pyrolysis particle characteristics, and short-circuit fault risk levels in historical test samples. The risk identification relationship represents the correspondence between pyrolysis particle response and low-risk, medium-risk, or high-risk levels under different cable structures, insulation defects, electrical operating states, and multiple inducing factors. The historical test samples are samples collected under different cable structure and material parameters, insulation defect parameters, electrical operating parameters, and multiple inducing factor parameters. The cable structure and material parameters, insulation defect parameters, electrical operating parameters, and multiple inducing factor parameters constitute the fault condition parameters.

[0026] An interpretation model is used to determine the key factors affecting the risk prediction results after obtaining the risk prediction results, combined with the fault condition parameters and pyrolysis particle characteristics corresponding to the current sample, and to generate an interpretation that includes the risk level, key factors and their correspondence with the insulation degradation state.

[0027] The output of the advanced identification module is connected to the control module and is used to output the risk prediction results of low risk, medium risk or high risk and the corresponding explanations.

[0028] The present invention provides a method for simulating and proactively identifying multi-cause short-circuit faults in cable joints, comprising the following steps:

[0029] S1: Place the cable joint fault simulation module inside the sealed test chamber and configure the cable structure and material parameters and insulation defect parameters of the cable joint body according to the preset fault simulation conditions.

[0030] S2: Based on the pre-input multi-factor target parameters, control the multi-factor simulation module to apply humidity, temperature, pressure and ultraviolet aging factors to the cable joint body;

[0031] S3: After each inducing factor reaches the preset condition, adjust the current value of the cable joint body according to the preset electrical operating parameters, and control the cable joint fault simulation module to be energized, so that the air gap defect of the main insulation layer will undergo insulation deterioration and evolve into a short circuit fault under the combined action of electric field and multiple inducing factors, generating pyrolysis particles.

[0032] S4: During the power-on degradation process, pyrolysis particle data in the sealed test chamber is collected in real time, and fault condition parameters corresponding to the pyrolysis particle data are recorded simultaneously.

[0033] S5: Extract pyrolysis particle features from the pyrolysis particle data, use the pyrolysis particle features and corresponding fault condition parameters as model input features, construct an advanced identification model, perform advanced identification of potential short-circuit faults and generate explanations, and output risk prediction results and corresponding explanations.

[0034] Furthermore, step S5 specifically includes:

[0035] S5-1: Construct an experimental sample set based on the pyrolysis particle data and its corresponding fault condition parameters obtained in step S4, extract pyrolysis particle features from the pyrolysis particle data, use the pyrolysis particle features and fault condition parameters as model input features, and label the experimental samples as low-risk, medium-risk or high-risk based on the changes in pyrolysis particles, insulation degradation status and short-circuit fault occurrence.

[0036] S5-2: Based on experimental samples with calibrated risk levels, establish the risk identification relationship between fault condition parameters, pyrolysis particle characteristics and short-circuit fault risk levels, in order to determine the short-circuit fault risk levels corresponding to pyrolysis particle responses under different cable structures, insulation defects, electrical operating states and multiple inducing factors.

[0037] S5-3: During the new power-on test, the real-time obtained pyrolysis particle characteristics and corresponding fault condition parameters are input into the risk identification relationship, and the risk prediction results of low risk, medium risk or high risk are output, and the key factors that affect the risk prediction results are identified.

[0038] S5-4: Generate an interpretation based on the fault condition parameters and pyrolysis particle characteristics corresponding to the risk prediction results, and store the key factors that can reflect the changes in risk level in historical samples and their correspondence with the risk level into the interpretation knowledge base for real-time interpretation generation of new samples.

[0039] Furthermore, the process of establishing the risk identification relationship described in step S5-2 includes:

[0040] (1) Select historical test samples with different fault simulation conditions and risk levels from the experimental sample set, and use cable structure and material parameters, insulation defect parameters, electrical operation parameters, multi-cause parameters and pyrolysis particle characteristics as candidate judgment factors;

[0041] (2) Set candidate classification conditions for each candidate judgment factor; for numerical data such as current value, ultraviolet irradiation duration, temperature, humidity, pressure, air gap length, air gap radius, air gap circumferential position, particle concentration, and particle size distribution, set several candidate boundary values; for categorical data such as conductor type, main insulation layer material, sheath type, and particle type, set candidate classification conditions according to different categories or combinations of categories; respectively count the proportions of low-risk, medium-risk, and high-risk samples in the samples on both sides of each candidate classification condition to determine whether a certain working condition parameter or pyrolysis particle characteristics is suitable as a risk judgment basis, and according to the formula:

[0042] ;

[0043] Calculate the risk confluence of the sample set S formed under the candidate judgment factor;

[0044] In the formula, S is a set of historical test samples collected according to a certain candidate judgment factor and its classification conditions. The historical test samples include cable structure and air gap defect parameters determined by the cable joint fault simulation module, current value recorded by the control module, temperature, humidity, pressure and ultraviolet irradiation duration formed by the multi-cause simulation module, and pyrolysis particle characteristics obtained by the data acquisition module; Q(S) is the degree of mixing of low-risk, medium-risk and high-risk samples in the sample set S; K is the number of risk levels; r is the risk level sequence number; pᵣ is the proportion of samples belonging to the r-th risk level in the sample set S;

[0045] After obtaining Q(S), the sample set S before partitioning is divided into a first-side sample set S1 and a second-side sample set S2 according to the candidate judgment factors and their candidate partitioning conditions, and then according to the formula:

[0046] ;

[0047] Calculate the discrimination of the candidate judgment factor and its candidate classification conditions on the short-circuit fault risk level;

[0048] In the formula, ΔQ is the reduction in risk mixing before and after the split, which indicates whether the candidate judgment factor can separate low-risk, medium-risk and high-risk samples; |S| is the number of samples in the historical test sample set S before the split; |S1| and |S2| are the number of samples falling on both sides of the split condition, respectively; Q(S1) and Q(S2) are the risk mixing of the sample sets on both sides, respectively.

[0049] Select the candidate judgment factors that maximize ΔQ and their candidate division conditions as the judgment conditions in the risk identification relationship, so that the experimental samples can be gradually mapped to low risk, medium risk or high risk according to the actual working parameters and pyrolysis particle response.

[0050] (3) Establish corresponding judgment paths for each judgment condition, and mark the fault condition range, pyrolysis particle characteristic range and short circuit fault risk level corresponding to each judgment path;

[0051] (4) Repeat the above sub-steps (1) to (3) until the samples of different risk levels in the experimental sample set can be distinguished according to the corresponding fault condition parameters and pyrolysis particle characteristics.

[0052] Furthermore, the explanation generation method includes:

[0053] (1) Identification process record: Record the complete judgment process of each historical test sample in the risk identification relationship, record the fault condition parameters or pyrolysis particle characteristics used for each judgment condition, the corresponding division conditions and the judgment path of the sample, and obtain the identification process corresponding to the sample.

[0054] (2) Key factor identification: Based on the identification process, candidate judgment conditions that can change the short-circuit fault risk level of the sample are searched, and the key factors corresponding to the candidate judgment conditions are identified; based on the data type, the key factors are divided into numerical key factors and categorical key factors.

[0055] (3) Numerical Key Factor Analysis: For the numerical key factors among the current value, UV irradiation duration, temperature, humidity, pressure, air gap parameters, particle concentration or particle size distribution, the numerical key factors are adjusted beyond the division value according to the corresponding candidate judgment conditions to form an adjusted sample x′. The adjusted sample x′ is then re-inputted into the advanced identification model to verify whether the short-circuit fault risk level has changed before and after the adjustment. The following formula is used to measure the degree of closeness between the current test state and the adjacent risk level state, and the difference between sample x and adjusted sample x′ is calculated:

[0056] ;

[0057] In the formula, sample x is the current test sample to be interpreted, corresponding to a set of actual operating condition parameters and pyrolysis particle characteristics recorded by the control module; adjusted sample x′ is a reference sample formed by adjusting only a certain numerical key factor to the adjacent risk level while keeping other test conditions unchanged; d(x,x′) is the difference between the two in fault operating condition parameters and pyrolysis particle characteristics, used to represent the degree of proximity of the current test state to the risk level change state; p is the total number of numerical key factors participating in the comparison; j is the j-th numerical key factor; xⱼ and x′ⱼ are the values ​​of the factor in sample x and adjusted sample x′, respectively; that is, xⱼ is the recorded value of the j-th parameter or feature in sample x; x′ⱼ is the reference value of the corresponding parameter or feature in the adjusted reference sample x′; wⱼ is the scale conversion coefficient of the j-th factor, used to enable data of different units such as current, temperature, pressure, particle concentration and particle size to be compared in the same difference quantity;

[0058] From the adjusted samples that can change the risk level of short circuit faults, select the adjusted sample with the smallest difference from sample x, and use the numerical key factors, adjustment direction and adjustment range corresponding to the adjusted sample as the basis for interpretation to explain which actual working conditions or pyrolysis particle characteristics mainly affect the current risk level.

[0059] (4) Processing of Category-based Key Factors: For the category-based key factors in conductor type, main insulation material, sheath type or particle type, record the category value of the category-based key factor and its classification conditions in the identification process, use it as the correlation condition of the risk prediction result, and do not make continuous numerical adjustments to the category-based key factors.

[0060] (5) Explanation of knowledge base construction: Numerical key factors that can change the risk prediction results, adjustment direction, adjustment magnitude, and corresponding categorical key factors and risk levels are stored in the explanation knowledge base, and risk correlation relationships that meet the preset threshold conditions are extracted from it.

[0061] (6) Real-time interpretation generation: For a new sample, historical samples with the same or similar risk association conditions are selected in the interpretation knowledge base according to the categorical characteristics of the new sample. Then, similar historical samples are determined according to the comprehensive difference between numerical key factors. An interpretation is generated based on the risk association relationship corresponding to the similar historical samples. The interpretation includes the risk prediction result, the key factors that affect the risk prediction result, and the adjustment direction and adjustment range of the numerical key factors.

[0062] The beneficial effects of this invention are:

[0063] 1. By setting insulation defect parameter adjustment units within the main insulation layer, the length, transverse cross-sectional dimensions, and circumferential orientation of the air gap defect can be independently adjusted. The air gap length is controlled by screwing a threaded insulating slider into or out of a preset air gap channel. The circumferential position of the defect is maintained by rotating a single target phase unit through a detachable rotor flange. This method can simulate cable joint insulation defects of different shapes, severity, and spatial orientations, overcoming the limitation of existing technologies that can only simulate a single fault type. It provides a diverse defect model for studying the evolution of insulation degradation and short-circuit faults.

[0064] 2. By applying humidity, temperature, pressure, and UV aging factors to the cable joints in the sealed test chamber through a multi-factor simulation module, and combining the closed-loop adjustment of the factor parameters by the control module, the independent application or combined coupling of multiple factors such as humidity, temperature, pressure, and UV aging can be achieved. Among them, the pressure simulation unit limits the pressure rod and provides auxiliary support for the three-phase unit through a detachable limiting support frame, so that the pressure block can stably apply pressure to the target phase unit. This solves the problem that the existing technology cannot simulate the short-circuit fault process of cable joints under the coupling of multiple factors, thus realistically reproducing the synergistic effect of multiple environmental factors such as damp heat, mechanical stress, and UV aging on insulation degradation.

[0065] 3. Each phase unit is installed independently by positioning support components, and each phase unit is set as a detachable structure. Phase units with different conductor types, main insulation layer materials and sheath types can be replaced according to the test needs, and air gap defect parameters can be reset, thereby forming fault simulation conditions with a variety of structural, material and insulation defect combinations, improving the applicability and reusability of the experimental device.

[0066] 4. The current carrying current of the cable joint body is set, detected and fed back through the current regulator and current sensor. The actual current carrying current, temperature, humidity, pressure and ultraviolet irradiation time are associated with the pyrolysis particle data at the same sampling time and recorded. This ensures that each set of pyrolysis particle data has clear test condition information, which is beneficial for comparing the pyrolysis response and short circuit fault evolution law under different cable structures, insulation defects, electrical operating conditions and external factors.

[0067] 5. By using the advanced identification module to correlate pyrolysis particle characteristics with cable structure and material parameters, insulation defect parameters, electrical operating parameters, and multiple inducing factors, it can output low-risk, medium-risk, or high-risk short-circuit fault risk levels. It further explains which factors among the following—current, UV irradiation duration, temperature, humidity, pressure, air gap parameters, or pyrolysis particle characteristics—have the main impact on the risk level. This solves the problem of existing technologies that only output risk results without providing explanations based on actual test conditions. It enables operators to understand the main reasons for insulation degradation and increased short-circuit risk, thereby guiding subsequent test adjustments and maintenance decisions. Attached Figure Description

[0068] Figure 1 This is a circuit connection block diagram of the system according to an embodiment of the present invention;

[0069] Figure 2 This is a detailed circuit connection diagram of the system according to an embodiment of the present invention;

[0070] Figure 3 This is a three-dimensional structural schematic diagram of the sealing test chamber according to an embodiment of the present invention;

[0071] Figure 4 This is a schematic diagram of the internal structure of the sealing test chamber according to an embodiment of the present invention;

[0072] Figure 5 This is a schematic diagram of the air gap length adjustment unit and pressure simulation unit according to an embodiment of the present invention;

[0073] Figure 6 yes Figure 5 A partially enlarged schematic diagram of the embodiment shown;

[0074] Figure 7 This is a schematic diagram of the specific structure of the defect location adjustment unit in an embodiment of the present invention;

[0075] Figure 8 This is a schematic diagram illustrating the working principle of the advanced identification module in an embodiment of the present invention.

[0076] Explanation of reference numerals in the attached diagram:

[0077] 1. Cable connector body; 2. Sealing test chamber; 3. First fixed support; 4. Humidity sensor; 5. Atomizing nozzle; 6. Temperature sensor; 7. Heating tube; 8. Pressure block; 9. Limiting support frame; 10. Ultraviolet lamp; 11. Copper core conductor; 12. Main insulation layer; 13. Insulating slider; 14. First knob; 15. Insulating sleeve; 16. Rotating sleeve; 17. Rotor flange; 31. Pin seat; 32. Positioning pin; 81. Pressure rod. Detailed Implementation

[0078] The following description, in conjunction with the accompanying drawings, provides further examples of the present invention to facilitate understanding of the invention, but does not limit the scope of the invention.

[0079] like Figures 1-4 As shown, this embodiment provides a multi-cause short-circuit fault simulation and advanced identification system for cable joints, including a cable joint fault simulation module, a multi-cause simulation module, a data acquisition module, an advanced identification module, and a control module. The cable joint fault simulation module is installed inside a sealed test chamber 2, including a cable joint body 1 and a positioning support for mounting the cable joint body 1. The cable joint body further includes several independently arranged phase units, and the conductive core of each phase unit is connected to the main test circuit via terminals and connecting wires. The main circuit is equipped with an on / off switch, a current regulator, and a current sensor to control the on / off state of the main circuit and to adjust and detect the current value of the cable joint body.

[0080] Specifically, the cable joint fault simulation module is used to replace the cable joint body 1 with different structural and material parameters, and adjusts the length, radius, and circumferential position of the air gap defect in the main insulation layer through the insulation defect parameter adjustment unit, thereby forming different cable joint fault simulation conditions. The multi-inducing factor simulation module is used to apply humidity, temperature, pressure, and ultraviolet aging inducing factors to the cable joint body. The data acquisition module is used to collect pyrolysis particle data generated by the cable joint body under energization and multi-inducing factor effects. The control module is used to receive fault simulation condition parameters input by the user, control the operating status of the multi-inducing factor simulation module, control the energizing current value of the test main circuit, and record and display the pyrolysis particle data and fault condition parameters. The advanced identification module is used to construct an advanced identification model using the pyrolysis particle features extracted from the pyrolysis particle data and the corresponding fault condition parameters as model input features, perform advanced identification of potential short-circuit faults and generate interpretations, and finally send the risk prediction results and interpretations to the control module; the control module can also perform early warning or power-off operations based on the risk prediction results.

[0081] In this embodiment, the cable connector body 1 is an equivalent test body used to simulate the three-phase conductors and insulation structure of the cable connector area. Preferably, the cable connector body 1 includes three independently arranged phase units, each phase unit including a copper core conductor 11 and a main insulation layer 12 covering the copper core conductor 11. Depending on the test requirements, each phase unit may also use other conductor materials and main insulation layer materials, and a removable outer sheath layer may be provided outside the main insulation layer 12.

[0082] Three phase units are respectively installed in positioning supports made of insulating material. The positioning supports include three spaced-apart mounting holes to maintain the relative positions of the phase units. The diameter of each mounting hole is slightly larger than the outer diameter of the corresponding phase unit, allowing for a rotatable clearance and providing radial restraint for the phase unit. During circumferential air gap position adjustment, torque is applied to the target phase unit only through the rotor flange, causing the target phase unit to rotate around its own axis within the corresponding mounting hole; other phase units are not subjected to torque and remain in their original positioning state. When the target phase unit rotates to a preset angle, its insulating outer surface is clamped and fixed by a phase retaining component to prevent rotation when the flange adjustment structure is subsequently removed and the energized circuit is connected. The positioning supports are only used to support and position the phase units and are not used as the cable outer sheath.

[0083] To adjust the parameters of internal insulation defects, an insulation defect parameter adjustment unit is provided in the main insulation layer 12. This unit specifically includes an air gap length adjustment unit, an air gap radius adjustment unit, and a defect position adjustment unit to adjust the corresponding defect parameters respectively. To achieve the above function, one or more air gap channels are first opened in the main insulation layer. The channel extends radially along the main insulation layer 12, pointing inward from the outer surface of the main insulation layer 12 toward the copper core conductor 11, and retains an impermeable insulating wall between the channel and the surface of the copper core conductor 11, so that the air gap channel forms a cavity closed at one end.

[0084] Specifically, such as Figure 5 and Figure 6As shown: The air gap length adjustment unit includes an insulating slider 13 disposed within the air gap channel and a first drive unit for controlling the movement of the insulating slider 13 along the air gap channel; the outer diameter of the insulating slider 13 matches the diameter of the air gap channel (when the insulating sleeve is not used), and the length of the insulating slider 13 is preferably greater than the length of the air gap channel. The outer surface of the insulating slider 13 is provided with external threads, and the opening section of the air gap channel, the inner wall of the insulating sleeve 15, or a threaded fitting at the channel opening is provided with corresponding internal threads. The first drive unit is preferably a first knob 14, which is connected to the insulating slider 13; when the operator rotates the first knob 14, the insulating slider 13 rotates radially into or out along the main insulation layer 12 under the influence of the threaded side effect, thereby changing the depth of the insulating slider 13 entering the preset air gap and changing the length of the air section in the air gap channel, i.e., changing the air gap length. The first knob 14 is engraved with length markings to indicate the air gap length value corresponding to the current position of the insulating slider 13. The first knob 14 is disposed inside the sealed test chamber 2. During adjustment, the operator opens the sealing test chamber 2 and adjusts the air gap length to the preset value by rotating the first knob 14. After adjustment, the sealing test chamber 2 is closed before proceeding with the subsequent power-on test. Since the air gap length adjustment is completed before power-on, placing the first knob 14 inside the sealing test chamber 2 simplifies the structure and maintains the sealing performance of the sealing test chamber 2.

[0085] The air gap radius adjustment unit includes several replaceable insulating sleeves 15. Each insulating sleeve 15 is positioned within a pre-reserved air gap channel. Its outer diameter matches the diameter of the air gap channel, while its inner diameter is available in various specifications to provide insulating sleeves with different inner diameters. The radius of the air gap channel is changed by replacing insulating sleeves with different inner diameters. The length of the insulating sleeve 15 is greater than the depth of the air gap channel, allowing its end to protrude from the channel opening for easy insertion and removal by hand or tool.

[0086] It is understood that the insulating sleeve 15 and the insulating slider 13 can share the same air gap channel, or multiple radially extending air gap channels can be opened within the main insulation layer 12, with adjustments between the channels not interfering with each other. To simplify the structure, this embodiment preferably uses only one air gap channel, allowing the insulating sleeve 15 and the insulating slider 13 to share the same channel. In this way, the air gap length adjustment and air gap radius adjustment can be simulated separately or simultaneously. When simultaneous adjustment is required, first select and insert the insulating sleeve 15 with the corresponding inner diameter according to the required radius, and then screw the insulating slider 13 into or out along the inner cavity of the insulating sleeve 15. This allows the length and radius of the air gap to be changed simultaneously, thereby simulating air gap defects with different lengths and radius combinations in the same cable joint fault simulation module, for studying fault characteristics under different combinations of air gap parameters.

[0087] like Figure 7As shown: The defect position adjustment unit adopts a detachable flange-type rotating structure, mainly used for circumferential position adjustment of the target phase unit in the three phases without driving the other phase units to rotate synchronously. Specifically, a rotating sleeve 16 is provided outside the main insulation layer 12 of the target phase unit, and the rotating sleeve 16 is fixed relative to the main insulation layer 12; a rotor flange 17 is detachably connected to the end of the rotating sleeve 16. The rotor flange 17 can be connected to the rotating sleeve 16 by means of screw connection, snap-fit, or keyway engagement, so as to transmit torque to the rotating sleeve 16 during the adjustment stage. The rotor flange 17 has multiple positioning holes evenly distributed along the circumferential direction. In this embodiment, eight positioning holes are preferably provided; a first fixed support 3 is provided in the sealing test chamber 2. The first fixed support 3 has a pin seat 31 at a position opposite to the rotor flange 17. The positioning pin 32 can pass through the pin seat 31 and be inserted into any positioning hole on the rotor flange 17. It can be understood that the pin seat 31 can also extend and retract along the first fixed support 3 to accommodate phase units of different heights. During adjustment, the operator pulls out the positioning pin 32 and rotates the rotor flange 17. The rotor flange 17 drives the rotating sleeve 16, the main insulation layer 12 and the copper core conductor 11 to rotate synchronously around the axis of the target phase unit, so that the air gap channel changes the circumferential orientation with the main insulation layer 12. When it rotates to the preset angle, the positioning pin 32 is inserted into the corresponding positioning hole to temporarily limit the rotor flange 17.

[0088] To avoid occupying the end space of the target phase unit during the energization phase, the defect position adjustment unit is also equipped with a phase holding unit. The phase holding unit includes a second fixed support, an insulating clamp, a flexible insulating gasket, and an anti-rotation limiting component. The second fixed support is connected to the sealed test chamber 2 or the positioning support. The insulating clamp is an openable structure, fitted onto the outside of the main insulation layer 12 or the outer sheath layer of the target phase unit. The flexible insulating gasket is located inside the insulating clamp to prevent damage to the outer insulating surface of the target phase unit when it is clamped. The anti-rotation limiting component is fixed to the second fixed support and abuts against the outer surface of the insulating clamp. After the insulating clamp closes and clamps the target phase unit, the insulating clamp remains relatively fixed to the target phase unit. The anti-rotation limiting component restricts the insulating clamp from moving circumferentially along the target phase unit, preventing the target phase unit from rotating around its own axis. When adjusting the circumferential position of the air gap, first disconnect the main test circuit and complete the discharge. Then, drive the target phase unit to rotate to the preset angle via the rotor flange 17. After the angle is determined, close and lock the phase holding unit to keep the circumferential position of the target phase unit unchanged by the insulating clamp and anti-rotation limiting component. Subsequently, remove the rotor flange 17, positioning pin 32, and their temporary connection structure to expose the copper core conductor 11 or terminal block at the end of the target phase unit for energization connection. In this way, the circumferential position of the air gap can be accurately changed during the adjustment stage without affecting subsequent energization due to the flange structure occupying the end.

[0089] Each phase unit is detachably connected to the positioning support. When it is necessary to change the cable structure and material parameters, the operator can remove the corresponding phase unit while the power is off and replace it with a phase unit that has a different conductor type, main insulation material, or sheath type. After replacement, each phase unit is re-fixed in the positioning support and connected to the test main circuit through the wiring terminals.

[0090] With the above structure, operators can adjust the length, radius, and spatial location of air gap defects inside the cable joint. The air gap length is controlled by the screw-in depth of the insulating slider, and the air gap position is adjusted by a single-phase detachable flange structure. After adjustment, the phase holding unit maintains the angle of the target phase unit, thereby simulating fault conditions under different combinations of air gap parameters and severity levels.

[0091] Understandably, the cable joint fault simulation module employs a replaceable phase unit structure. The conductor type can be selected from copper, aluminum, or other conductive cores depending on the test requirements; the main insulation material can be cross-linked polyethylene, ethylene propylene rubber, or other insulating materials; and the outer sheath can also be selected from different materials or structures depending on the test requirements. By replacing different phase units and adjusting the length, radius, and circumferential position of the air gap defect, test conditions with different combinations of structures, materials, and insulation defect parameters can be formed. When an energized test is required, the circumferential adjustment of the target phase unit can be completed first using the rotor flange, and then the phase unit can be fixed using the phase holding unit, and the flange structure can be removed to ensure sufficient space for end wiring.

[0092] In this embodiment, the multi-factor simulation module is used to simulate the effects of factors such as humidity, temperature, pressure and ultraviolet aging on the cable joint body. It includes a humidity simulation unit, a temperature simulation unit, a pressure simulation unit and an ultraviolet aging simulation unit, all of which are installed in the sealed test chamber.

[0093] The humidity simulation unit includes a spray device and a humidity sensor 4. The spray device includes at least one atomizing nozzle 5, which is installed on the top of the sealed test chamber 2 or other preset positions. The spray volume is controlled by the control module to adjust the ambient humidity. The humidity sensor 4 is used to feed back the humidity value inside the sealed test chamber 2 to the control module. The humidity adjustment range is 20%RH to 95%RH.

[0094] The temperature simulation unit includes a heater, a cooler, and a temperature sensor 6, all of which are connected to a control module. The module regulates the ambient temperature inside the sealed test chamber by adjusting the power of the heater and controlling the start / stop of the cooler. Temperature information is fed back via the temperature sensor 6, supporting a temperature range of -20℃ to 100℃. Preferably, the heater includes several heating tubes 7 arranged on the inner wall of the sealed test chamber.

[0095] The pressure simulation unit is installed inside the sealed test chamber 2, directly above the cable joint fault simulation module. This unit includes a pressure block 8, a second drive unit, a stress sensor, and a detachable limiting support frame 9. The pressure block 8 is used to contact and apply pressure to the outer surface of the main insulation layer of a target phase in the cable joint body 1. The second drive unit drives the pressure block 8 to move up and down to change the pressure on the main insulation layer 12, and monitors this pressure using the stress sensor. Preferably, the second drive unit includes a second knob and a pressure rod 81 for moving the pressure block up and down. The detachable limiting support frame 9 is located near the cable joint body and includes multiple horizontal support plates and vertical support plates connected to both ends of the horizontal support plates. Preferably, four horizontal support plates are provided. The top two horizontal support plates each have guide holes to restrict the pressure rod 81 to move only in the vertical direction, preventing the pressure block 8 from laterally swaying during pressure application. The third horizontal support plate is located between three phase units, for example, between two upper-layer phase units arranged side-by-side and one lower-layer phase unit. The fourth horizontal support plate supports the lower-layer phase unit. This structure allows the detachable limiting support frame 9 to form a support part adapted to the shape of the three-phase unit, providing auxiliary support for the three-phase unit and making the relative position between the three-phase units more stable during pressure application. The upper end of the pressure rod 81 is connected to the second knob, and the lower end of the pressure rod passes through the guide hole and connects to the pressure block 8. The inner wall of the guide hole is provided with internal threads, and the pressure rod 81 is preferably a lead screw with external threads, forming a rotatable threaded pair with the internal threads of the guide hole. When the operator rotates the second knob, the pressure rod 81 moves smoothly axially under the constraint of the detachable limiting support frame 9, thereby driving the pressure block 8 to move up and down, changing the pressure on the main insulation layer 12 of the cable joint. A stress sensor can be attached to the bottom surface of the pressure block to monitor the pressure value in real time and feed it back to the control module.

[0096] Since pressure adjustment is completed before the main test circuit is energized, it is preferable to place the second knob inside the sealed test chamber 2. During pressure adjustment, the operator opens the sealed test chamber 2, rotates the second knob, causing the pressure rod 81 to move axially under the influence of the threaded connection, and moving the pressure block 8 towards the main insulation layer 12 of the target phase unit. Based on the pressure value detected by the stress sensor, the pressure block 8 is adjusted to the position corresponding to the preset pressure. After adjustment, the second knob is stopped, and the pressure rod 81 maintains its current position through the threaded connection, allowing the pressure block 8 to continuously apply the preset pressure to the main insulation layer 12 after the sealed test chamber 2 is subsequently closed and the main test circuit is connected. The sealed test chamber 2 is then closed, and other induced factor adjustments and energization tests are performed.

[0097] The ultraviolet aging simulation unit includes an ultraviolet lamp 10 and a relay. The ultraviolet lamp 10 is fixedly installed on the top or side wall of the sealed test chamber 2, with its irradiation direction facing the part of the cable joint body 1 to be aged, and is used to irradiate the main insulation layer 12 or the sheath surface with ultraviolet light. A relay is connected in series in the power supply circuit of the ultraviolet lamp 10, and the control terminal of the relay is connected to the control module, which controls the start and stop of the ultraviolet lamp 10 and records the ultraviolet irradiation duration. In the same set of comparative tests, the irradiation intensity, irradiation distance, and irradiation position of the ultraviolet lamp 10 are kept consistent. By setting different ultraviolet irradiation durations, different ultraviolet aging states are formed in the cable joint body. In this embodiment, the preset duration of ultraviolet irradiation is preferably completed before the cable joint body is energized; when it is necessary to simulate the simultaneous action of ultraviolet aging and other inducing factors, the ultraviolet lamp can also be kept running during the energization process.

[0098] Understandably, the humidity simulation unit, temperature simulation unit, and UV aging simulation unit are each independently controlled by relays, while the pressure simulation unit is independently controlled by the second knob, enabling individual or combined simulation of the causative factors. Corresponding relays are installed in the power supply circuits of the spray device, heater, cooler, and UV lamp. The control terminals of each relay are connected to the I / O interface of the control module. The control module controls the on / off state of the relays to start and stop the corresponding actuators. The pressure block 8 of the pressure simulation unit is connected to the second knob via a pressure rod 81. When the operator rotates the second knob, the pressure block 8 moves closer to or further away from the cable connector body under the guidance of the detachable limiting support frame 9, thereby changing the pressure applied to the main insulation layer 12. The detachable limiting support frame 9 can be removed when pressure is not required or when phase units need to be replaced, facilitating assembly and maintenance.

[0099] In this embodiment, the main test circuit is also equipped with a current regulator and a current sensor. The current regulator is connected between the test power supply and the conductive core of each phase unit, and its control terminal is connected to the control module to adjust the energizing current value of the cable connector body according to preset electrical operating parameters. The detection terminal of the current sensor is connected to the main test circuit to detect the actual energizing current in the main test circuit in real time and feeds the detection result back to the control module. The control module compares the actual energizing current value with the preset value and adjusts the energizing current through the current regulator to keep the actual energizing current within the preset range.

[0100] In this embodiment, the data acquisition module includes a pyrolysis particle collector and a sampling circuit. The pyrolysis particle collector is installed on the inner wall of the sealed test chamber 2, with its sampling port facing the surface of the main insulation layer of the cable joint fault simulation module. Under the influence of energization of the cable joint body and multiple inducing factors, the main insulation layer undergoes pyrolysis, generating tiny solid particles and characteristic gases. The pyrolysis particle collector is used to collect the corresponding pyrolysis particle data in real time. The output terminal of the pyrolysis particle collector is connected to the control module and the advanced identification module via the sampling circuit. The sampling circuit is used to condition the acquired signal and reduce interference. The pyrolysis particle data includes at least particle concentration, particle type, and particle size distribution data.

[0101] Temperature, humidity, and pressure are acquired by temperature sensor 6, humidity sensor 4, and stress sensor, respectively. The actual energized current value is acquired by a current sensor. The UV irradiation duration is recorded by the control module based on the running time of the UV aging simulation unit. Cable structure and material parameters, as well as insulation defect parameters, are input into the control module by the operator when setting up the fault simulation conditions. The control module correlates the pyrolysis particle data with the corresponding fault condition parameters according to the sampling time, so that each set of pyrolysis particle data has its corresponding cable structure and material parameters, insulation defect parameters, electrical operation parameters, and multi-cause parameters, forming experimental samples for constructing an advanced identification model.

[0102] like Figure 8 As shown: In this embodiment, the advanced identification module constructs an advanced identification model based on the collected pyrolysis particle data and its corresponding fault condition parameters to identify potential faults in advance. The input of the advanced identification module is connected to the output of the data acquisition module, and the output is connected to the control module to output low-risk, medium-risk, or high-risk risk prediction results and corresponding explanations. Specifically, the advanced identification module uses an embedded processor to run the advanced identification model and the interpretation model. The advanced identification model is used to establish the correspondence between different fault simulation conditions, pyrolysis particle responses, and short-circuit fault risk levels; the interpretation model is used to explain the main reasons for the formation of risk levels by combining the current experimental conditions and pyrolysis particle characteristics.

[0103] In this embodiment, the control module includes a microcontroller and a computer. The microcontroller connects to the cable joint fault simulation module, the multi-cause simulation module, the data acquisition module, and the advanced identification module via serial port, digital interface, or analog interface, respectively, and controls or acquires data from the on / off switches, current regulators, and current sensors in the main test circuit. The computer provides a human-machine interface for inputting cable structure and material parameters, insulation defect parameters, electrical operating parameters, and multi-cause parameters, and displays pyrolysis particle data, actual current flow, temperature, humidity, pressure, UV irradiation duration, and risk prediction results. The microcontroller controls the operation of each relay and current regulator, and receives data from each sensor and data acquisition module. For example, the microcontroller can adjust the heater power via PWM signals, regulate the current flow by controlling the current regulator, and execute an early warning or disconnect the main circuit when the risk prediction results reach preset conditions.

[0104] The multi-cause short-circuit fault simulation and advance identification method in this embodiment specifically includes the following steps:

[0105] S101: Place the cable joint fault simulation module inside the sealed test chamber 2, and configure the structural and material parameters and insulation defect parameters of the cable joint body 1.

[0106] Specifically, with the main test circuit de-energized and discharged, the operator selects phase units with corresponding conductor types, main insulation materials, and sheath types according to the preset fault simulation conditions, and installs each phase unit into the corresponding mounting holes of the positioning support. Then, the length, radius, and circumferential position of the air gap are adjusted using the insulation defect parameter adjustment unit. The air gap length is determined by rotating the first knob 14 to change the depth of the threaded insulating slider 13 into the preset air gap channel; the circumferential position is determined by the rotor flange 17 rotating the target phase unit. After the circumferential position is adjusted, the target phase unit is fixed using the phase holding unit. Then, the rotor flange 17, positioning pin 32, and temporary connection structure are removed, and the conductive cores of each phase unit are connected to the main test circuit via terminals and connecting wires. After configuration, the sealed test chamber 2 is closed.

[0107] S102: Based on the pre-input multi-factor target parameters, control the multi-factor simulation module to apply humidity, temperature, pressure and ultraviolet aging factors to the cable joint body.

[0108] Specifically, the computer receives multi-factor target parameters input by the user and sends corresponding control commands to the microcontroller. The microcontroller controls the on / off switching of the relays corresponding to the spray device, heater, cooler, and ultraviolet lamp 10, so that the humidity and temperature in the sealed test chamber 2 reach preset values, and the cable joint body is irradiated with ultraviolet light for a preset duration. The operator adjusts the position of the pressure block 8 through the second knob, so that it applies preset pressure to the main insulation layer 12 of the target phase under the guidance of the detachable limiting support frame 9, while the detachable limiting support frame 9 provides auxiliary support for the three-phase unit. In the same group of ultraviolet aging comparison tests, the irradiation intensity, irradiation distance, and irradiation position of the ultraviolet lamp are kept consistent, and different ultraviolet aging states are formed by changing the ultraviolet irradiation duration. Humidity, temperature, pressure, and ultraviolet aging factors can be applied individually or in combination according to preset fault simulation conditions.

[0109] S103: After the cable joint body 1 reaches the preset multi-cause simulation working condition, adjust the current value of the cable joint body according to the preset electrical operating parameters, and control the cable joint fault simulation module to be energized, so that the air gap defect of the main insulation layer will undergo insulation degradation and generate pyrolysis particles under the combined action of electric field and multiple causes.

[0110] Specifically, when the temperature, humidity, and pressure reach preset values, and the ultraviolet irradiation lasts for a preset duration, the control module sends a current setting command to the current regulator based on preset electrical operating parameters. Subsequently, the microcontroller controls the on / off switches in the main circuit to close, forming an electrical path between the test power supply, connecting wires, and the conductive cores of each phase unit. After the phase unit is energized, the current sensor detects the actual current in real time and feeds it back to the control module. The control module uses the current regulator to maintain the actual current within a preset range. Artificial air gap defects within the main insulation layer 12 accelerate insulation degradation under the combined action of the electric field and multiple inducing factors, gradually evolving into a short-circuit fault and generating pyrolysis particles.

[0111] S104: During the power-on degradation process, pyrolysis particle data in the sealed test chamber is collected in real time, and fault condition parameters corresponding to the pyrolysis particle data are recorded simultaneously.

[0112] Specifically, the pyrolysis particle collector collects particle concentration, particle type, and particle size distribution data at preset sampling intervals, and sends the data to the control module and the advanced identification module via the sampling circuit. The control module simultaneously records the actual energized current, temperature, humidity, pressure, and ultraviolet irradiation duration, and correlates this data with pre-inputted cable structure and material parameters and insulation defect parameters according to the sampling time. The computer displays and stores the pyrolysis particle data and fault condition parameters, thereby forming an experimental sample with a unified time correspondence.

[0113] S105: Construct an advanced identification model based on pyrolysis particle data and its corresponding fault condition parameters, perform advanced identification of potential short-circuit faults, generate explanations, and finally output low-risk, medium-risk, or high-risk risk prediction results and corresponding explanations.

[0114] Specifically, the advanced identification module extracts pyrolysis particle features from the pyrolysis particle data and uses these features, along with the corresponding fault condition parameters, as input features to the model. The fault condition parameters indicate the cable structure and material parameters, insulation defects, electrical operating conditions, and multiple contributing factors that caused the current sample. The pyrolysis particle features represent the pyrolysis response of the main insulation layer under these conditions. The advanced identification module establishes a correspondence between these two types of data and the short-circuit fault risk level, outputting low, medium, or high risk, and the interpretation model generates a description corresponding to that risk level. This includes the following steps:

[0115] S105-1: Construct an experimental sample set, use fault condition parameters and pyrolysis particle characteristics as model input features, and label the short-circuit fault risk level as low risk, medium risk or high risk based on the changes in pyrolysis particles, insulation degradation state and short-circuit fault occurrence in the experiment.

[0116] Specifically, the fault condition parameters include cable structure and material parameters, insulation defect parameters, electrical operating parameters, and multi-cause parameters. Among these, cable structure and material parameters include at least conductor type, main insulation material, and sheath type; insulation defect parameters include air gap length, air gap radius, and circumferential air gap position; electrical operating parameters include at least the current value; and multi-cause parameters include temperature, humidity, pressure, and duration of ultraviolet irradiation. The pyrolysis particle characteristics are obtained by processing pyrolysis particle data and include at least particle concentration, particle type, and particle size distribution.

[0117] The control module records the fault condition parameters and pyrolysis particle characteristics according to the sampling time, so that each experimental sample corresponds to a specific experimental state. Based on the changes in pyrolysis particle concentration, particle type, and particle size distribution under this experimental state, and in conjunction with the degree of main insulation layer degradation during the experiment, whether short circuit precursors appear, and whether a short circuit eventually occurs, the experimental sample is classified into risk levels, including low risk, medium risk, and high risk.

[0118] Preferably, low risk indicates that the pyrolysis particle response is weak under the current fault simulation condition, and the main insulation layer does not show a significant deterioration trend; medium risk indicates that the pyrolysis particle response has changed significantly compared to the low-risk state, and the main insulation layer shows a tendency to continue deteriorating and develop into a short-circuit fault; high risk indicates that the pyrolysis particle response is significantly enhanced, or the test state is close to a short-circuit precursor or short-circuit fault state. The above risk levels can be determined comprehensively based on test results, historical samples, and manual interpretation results.

[0119] This embodiment constructs an experimental sample set, using cable structure and material parameters, insulation defect parameters, electrical operation parameters, multi-cause parameters, and pyrolysis particle characteristics as model input features. It establishes the correspondence between different fault simulation conditions, insulation material pyrolysis response, and short-circuit fault risk, providing a data foundation for advanced identification models.

[0120] S105-2: Establish risk identification relationships based on experimental samples with calibrated risk levels. These risk identification relationships are used to map new experimental samples to low-risk, medium-risk, or high-risk categories.

[0121] Specifically, this step involves organizing the correspondences that have occurred in historical experiments into reusable judgment criteria. Each historical sample includes the cable structure and material parameters used in the experiment, air gap defect parameters, current value, temperature, humidity, pressure, UV irradiation duration, and pyrolysis particle characteristics such as particle concentration, particle type, and particle size distribution, and has calibration results for low-risk, medium-risk, or high-risk. After comparing these historical samples, the advanced identification module determines which operating condition parameters or pyrolysis particle characteristics can distinguish different risk levels, thereby forming risk identification relationships for subsequent test samples. The process of establishing risk identification relationships specifically includes:

[0122] (1) Select historical test samples covering different cable structures, different defect parameters, different current values ​​and different combinations of multiple causes from the experimental sample set to form a sample set for establishing risk identification relationships;

[0123] (2) Set candidate classification conditions for candidate judgment factors such as current value, temperature, humidity, pressure, UV irradiation duration, air gap length, air gap radius, air gap circumferential position, particle concentration, particle type and particle size distribution. For example, set several candidate boundary values ​​for numerical data such as current value, UV irradiation duration, temperature, humidity, pressure, air gap length, air gap radius, air gap circumferential position, particle concentration and particle size distribution; set candidate classification conditions according to different categories or combinations of categories for categorical data such as conductor type, main insulation material, sheath type and particle type; and observe whether the samples on both sides of the classification conditions mainly fall into low risk, medium risk or high risk.

[0124] (3) Retain the judgment factors and their classification conditions that can clearly distinguish the risk level, form a judgment path from actual test conditions and pyrolysis particle response to risk level, and mark the corresponding working condition range, particle characteristic range and risk level in the judgment path;

[0125] (4) Repeat the above process to obtain multiple risk assessment paths. When a new test sample enters, the advanced identification module gives the risk level according to these assessment paths, and then determines the final risk level by combining the assessment results.

[0126] In order to determine whether a set of historical test records can be clearly classified as low-risk, medium-risk, or high-risk, in the risk identification relationship establishment stage, for any sample set S, the following formula is used:

[0127] ;

[0128] Calculate the risk mixing degree of low-risk, medium-risk, and high-risk samples in sample set S;

[0129] In the formula, S represents a set of historical test records to be classified in a certain judgment step. Each historical test record is formed by the control module according to the sampling time and includes at least cable structure and material parameters, air gap defect parameters, current value, temperature, humidity, pressure, ultraviolet irradiation duration, particle concentration, particle type, and particle size distribution. Q(S) represents the degree of mixing of risk levels within the set of historical test records, used to determine whether the set of records clearly corresponds to low risk, medium risk, or high risk. K represents the total number of short-circuit fault risk levels, and in this embodiment, K=3. r represents the risk level sequence number, corresponding to low risk, medium risk, or high risk, respectively. pᵣ represents the proportion of samples belonging to the r-th risk level in the sample set S. When Q(S) is small, it indicates that the risk levels of the set of records are relatively concentrated and can be used as a basis for risk judgment. When Q(S) is large, it indicates that the set of records still contains multiple risk states and needs to be further distinguished by combining other operating condition parameters or pyrolysis particle characteristics.

[0130] Subsequently, in order to determine whether a certain operating condition parameter or pyrolysis particle characteristic has a distinguishing effect, the sample set S is divided into a first-side sample set S1 and a second-side sample set S2 according to the candidate judgment factor and its candidate partitioning conditions, and then according to the formula:

[0131] ;

[0132] Calculate the effect of this classification condition on distinguishing risk levels.

[0133] In the formula, ΔQ represents the reduction in risk mixing after classification using a certain operating condition parameter or pyrolysis particle characteristics, indicating whether the classification condition helps to separate low-risk, medium-risk, and high-risk levels; |S| is the number of records in the historical test record set S before classification; |S1| and |S2| are the number of records falling into the first and second sides after classification, respectively; Q(S1) and Q(S2) are the degree of mixing of risk levels within the record sets on both sides, respectively. The larger ΔQ is, the better the operating condition parameter or pyrolysis particle characteristics can reflect changes in short-circuit fault risk, and it can be used as an effective judgment basis in risk identification relationships.

[0134] When establishing risk identification relationships in practice, the judgment factors and their classification conditions that result in a large ΔQ are selected as effective judgment criteria, so that historical test records can be gradually mapped to low-risk, medium-risk, or high-risk levels. The judgment criteria obtained in this way include operating conditions such as current, temperature, humidity, pressure, UV irradiation duration, and air gap parameters, as well as pyrolysis response factors such as particle concentration, particle type, and particle size distribution.

[0135] When new test samples enter the advanced identification module, multiple risk assessment paths each provide a risk level. To ensure the final result integrates multiple historical assessment criteria, the final risk level is determined using the following formula:

[0136] ;

[0137] Determine the final risk level of the current test sample;

[0138] In the formula, Y represents the short-circuit fault risk level of the current test sample, which includes low risk, medium risk, and high risk; T represents the number of risk judgment paths involved in the judgment, and each risk judgment path is derived from a set of correspondences between operating parameters, pyrolysis particle characteristics, and risk levels formed in historical test records; y1, y2, ..., y T These represent the risk levels for risk assessment paths 1 through T, respectively; mode(·) indicates selecting the risk level that appears most frequently in each assessment path. The result of this formula is output from the advanced identification module to the control module as the risk prediction result for the current cable joint test status.

[0139] Through the above processing, the advanced identification module can put the pyrolysis particle data back into the corresponding actual test conditions for judgment, rather than drawing conclusions based on the value of a single particle. This allows the risk level judgment to correspond to the insulation degradation process of the cable joint under multiple inducing factors.

[0140] S105-3: Based on the risk identification relationship, form an explanation basis, determine which operating parameters or pyrolysis particle characteristics mainly affect the current risk level, and generate an explanation accordingly.

[0141] Specifically, the explanatory model is used to explain the reasons for the formation of risk levels. Its output is not simply a calculated score, but rather an explanation of the causes for the operator, such as factors like high current, prolonged UV irradiation time, larger air gap size, increased pressure or temperature and humidity conditions, increased particle concentration, or significant changes in particle size distribution, which make the current sample closer to a medium- or high-risk state. These factors include:

[0142] (1) Record the basis for judgment: For each historical test sample, record its judgment process in each risk judgment path; record the formation process of the risk level as a set of traceable working conditions and pyrolysis particle judgment conditions. For sample x, its judgment process in the t-th risk judgment path can be expressed as:

[0143] ;

[0144] In the formula, P t (x) represents the judgment record of sample x in the t-th risk judgment path, where sample x represents the cable joint test sample to be identified or interpreted; L is the number of judgment conditions passed sequentially in this judgment path; k is the judgment condition number; f k The actual factors used for the k-th judgment condition can be the current value, temperature, humidity, pressure, ultraviolet irradiation duration, air gap length, air gap radius, circumferential position of the air gap, particle concentration, particle type, or particle size distribution; θ k For the engineering classification conditions corresponding to the k-th judgment condition, such as current setting range, temperature and humidity range, ultraviolet irradiation duration range, air gap size range, or particle concentration range; d k The side of sample x relative to the dividing condition indicates the direction of subsequent risk assessment for that sample along the judgment condition. This judgment record is used to trace the final risk level back to specific operating parameters and pyrolysis particle characteristics.

[0145] (2) Identifying the critical points of risk change: For a sample x that has already yielded low-risk, medium-risk, or high-risk results, the interpretative model further identifies which judgment factors only require minor changes to potentially shift the sample to another risk level. To this end, an adjusted sample x′ is generated, and the difference between sample x and adjusted sample x′ is calculated using the formula:

[0146] ;

[0147] In the formula, d(x,x′) represents the difference in experimental state between the current sample x and the adjusted reference sample x′, indicating how far the current experimental state differs from the adjacent risk level state; sample x is the actual collected or recorded experimental sample, and sample x′ is the reference sample obtained by making minor adjustments to a key parameter or pyrolysis particle characteristic without changing other conditions; p is the total number of parameters or characteristics involved in the comparison; j is the j-th parameter or characteristic; xⱼ is the actual recorded value of the j-th factor in sample x; x′ⱼ is the reference value of the corresponding factor in the adjusted reference sample x′; wⱼ is the scale conversion coefficient of the j-th factor, which can be set according to the range, unit, or experimental recording range of the factor, to enable the comparison of quantities in different units such as current value, temperature, pressure, ultraviolet irradiation duration, air gap size, particle concentration, and particle size. The smaller d(x,x′) is, the closer the current sample is to the risk level change position, and the more suitable this factor is as the focus of interpretation.

[0148] (3) Explanation generation:

[0149] ① Determine key judgment conditions: For sample x, based on the above judgment records, identify the judgment conditions that may lead to changes in the risk level; this formula is used to select the risk change location closest to the current experimental state from multiple possible factors, and follows the formula:

[0150] ;

[0151] Identify the location of the risk change that is closest to sample x;

[0152] In the formula, v* represents the key judgment condition in the t-th risk judgment path that is closest to sample x, i.e., the actual factor most likely to cause a change in the current risk level; V t (x) represents the set of all judgment conditions that could potentially change the risk level of sample x along this risk assessment path; B v The boundary sample is obtained by adjusting the corresponding operating parameters or pyrolysis particle characteristics in sample x according to the judgment condition v. The boundary sample represents the reference state that just enters the adjacent risk level side; d(x,B v Let x be the sample and b be the boundary sample B. v The difference between them. The calculation result of this formula is used to determine the priority of the interpretation content, that is, to give priority to the current, air gap parameters, multi-cause parameters or pyrolysis particle characteristics that are closest to the location of risk change.

[0153] ② Forming an adjustment reference sample: For the key judgment condition v* determined in step ①, if its corresponding judgment factor is f j The partitioning condition is θ jIf other operating parameters and pyrolysis particle characteristics in sample x remain unchanged, only the j-th parameter or characteristic is slightly adjusted beyond the partitioning condition to form an adjusted reference sample x′ for interpretation.

[0154] When x j ≤θ j At that time, according to the formula:

[0155] ;

[0156] Generate the value of the j-th parameter or feature in the adjusted reference sample;

[0157] When x j >θ j At that time, according to the formula:

[0158] ;

[0159] Generate the value of the j-th parameter or feature in the adjusted reference sample;

[0160] In the formula, x′ j To adjust the value of the j-th parameter or feature in the reference sample x′; x j θ is the actual recorded value of the j-th parameter or feature in sample x; j The j-th parameter or feature represents the engineering classification condition, such as current threshold, temperature range, UV irradiation duration range, air gap size range, particle concentration range, or particle size distribution range. ε is the adjustment margin, used to ensure that the adjusted parameter or feature just exceeds the corresponding classification condition, preventing the adjusted value from remaining on the original risk level side. The j-th parameter or feature is any one of the following: current value, temperature, humidity, pressure, UV irradiation duration, air gap parameter, particle concentration, or particle size distribution. This adjustment is only used to explain how close the current sample is to the adjacent risk state and does not change the actual collected experimental data or the data recorded by the control module.

[0161] ③ Verify whether key factors affect the risk level: Input the adjusted reference sample x′ into the advanced identification model, and follow the formula:

[0162] ;

[0163] Determine whether the short-circuit fault risk level has changed before and after the adjustment;

[0164] In the formula, Y(x) represents the short-circuit fault risk level corresponding to sample x; Y(x′) represents the short-circuit fault risk level corresponding to the adjusted reference sample x′. When Y(x′) ≠ Y(x), it indicates that the risk level will change after the parameter or feature crosses the corresponding classification condition; therefore, this parameter or feature is the key factor that needs to be highlighted in the interpretation. The result of this formula is used to generate the interpretation content, enabling the risk results output by the control module to simultaneously explain the main influencing factors and their direction of change.

[0165] (4) Explanation of Knowledge Base Construction: A knowledge base is formed by summarizing key factors that have been verified to change the risk level from a large number of historical samples. For each key factor j, according to the formula:

[0166] ;

[0167] Calculate the proportion of times this key factor caused a change in risk level in historical trials;

[0168] In the formula, Pⱼ represents the proportion of the j-th key factor that causes a change in risk level in historical trials, indicating the importance of this factor in the explanatory knowledge base; M represents the total number of historical sample pairs participating in the statistics, including the original sample and the adjusted reference sample constructed around it; m represents the m-th historical sample pair; ΔY m Let Δx be the change in risk level before and after adjustment for the m-th sample pair; m,ⱼ P represents the numerical change of the j-th key factor in the m-th sample pair; I is a counting function, which takes the value 1 when the risk level of the m-th sample pair changes and the j-th key factor is adjusted, and 0 otherwise. Pⱼ is used to determine which factors cause risk level changes more frequently in historical trial records, and serves as a priority indicator for real-time interpretation.

[0169] For numerical key factors, to determine the extent to which a change in this factor typically needs to occur to alter the risk level, further calculations are performed using the formula:

[0170] ;

[0171] In the formula, Δⱼ represents the typical adjustment range of the j-th key factor when the risk level changes, indicating the extent to which this factor typically changes to alter the risk level; x m,ⱼ x′ represents the actual recorded value of the j-th key factor in the m-th original sample. m,ⱼ This is used to adjust the reference value of the j-th key factor in the reference sample; `median` indicates taking the median value among multiple adjustment ranges that meet the conditions, to avoid individual abnormal test records having an excessive impact on the typical range. `Δⱼ` is used to ensure that the interpretation results not only indicate the influencing factors, but also explain the magnitude relationship between the factor and the change in risk level.

[0172] In the above way, the explanation model can transform risk prediction results into understandable engineering reasons, explaining which operating conditions or pyrolysis particle characteristics mainly affect the current sample, and the relationship between these factors and the increase or decrease of risk level.

[0173] S105-4: Construct an explanatory knowledge base to extract reusable risk relationships from historical samples for real-time explanation generation of new samples.

[0174] Specifically, after a new sample x yields a risk prediction result, the system first searches the interpretation knowledge base for historical samples that most closely resemble its experimental conditions. These samples must have similar actual factors, including current value, temperature, humidity, pressure, UV irradiation duration, air gap parameters, pyrolysis particle concentration, particle type, and particle size distribution. This formula is used to select the most suitable reference sample from historical experimental records as the basis for the current interpretation.

[0175] ;

[0176] In the formula, x* represents the historical sample in the interpretation knowledge base that is closest to the current new sample x. This historical sample has similar conditions to the current sample in terms of current value, temperature, humidity, pressure, UV irradiation duration, air gap parameters, and pyrolysis particle characteristics. x′* represents the adjustment reference sample corresponding to the historical sample x*, used to indicate the reference state when the historical sample enters an adjacent risk level. KB represents the interpretation knowledge base. u represents any historical sample in the interpretation knowledge base. u′ represents the adjustment reference sample corresponding to the historical sample u. d(x,u) represents the difference between the current new sample x and the historical sample u in terms of operating parameters and pyrolysis particle characteristics. The result of this formula is used to determine which type of historical interpretation basis should be reused for the current sample.

[0177] Then, the risk change direction that has been validated in the historical sample is used for the current new sample, and the formula is applied:

[0178] ;

[0179] Generate an explanatory reference state for the current new sample;

[0180] In the formula, x′ is the explanatory reference sample generated for the current new sample x; x is the new sample to be explained; x* and x′* are the historical samples matched from the explanation knowledge base and their adjusted reference samples, respectively; x′*-x* represents the direction and magnitude of change of the corresponding operating parameters or pyrolysis particle characteristics when the historical sample changes from its original state to an adjacent risk level; α is an adjustment coefficient used to adapt the magnitude of change based on the similarity between the current new sample x and the historical sample x*. The result of this formula is used to generate an explanatory reference state that matches the current energizing current, air gap parameters, multi-cause conditions, and pyrolysis particle response, thereby forming a natural language explanation for operators.

[0181] Finally, the above explanatory reference values ​​are converted into natural language interpretations. For example, if the current sample is judged to be high-risk, the explanation can be stated as: the current current value is high, the ultraviolet irradiation time is long, the air gap radius is large, the pressure or temperature and humidity conditions are aggravated, or the particle concentration is significantly increased and the particle size distribution is concentrated in the abnormal range, making it closer to the high-risk historical samples; if the current sample is judged to be medium-risk, the explanation can be stated as: the pyrolysis particle response has changed relative to the low-risk state, but the relevant operating parameters or particle characteristics have not yet reached the range corresponding to the high-risk samples.

[0182] This embodiment constructs an explanatory knowledge base, enabling the advanced identification module to not only output low-risk, medium-risk, or high-risk results, but also to explain the actual experimental reasons corresponding to the results. This facilitates operators' understanding of the main factors contributing to cable joint insulation degradation and increased short-circuit risk, and provides a basis for subsequent test adjustments, early warning measures, and maintenance decisions.

[0183] In summary, this embodiment, through the aforementioned system and method, achieves insulation degradation simulation and early short-circuit risk identification of cable joints under multiple inducing factors. Specifically, by employing multi-inducing factor simulation, replaceable phase units, and adjustable air gap structures, it improves the realism and flexibility of fault simulation. By combining the acquisition of pyrolysis particle data with the early identification model, it achieves early warning of faults. It is effectively applicable to power system fault evolution analysis, diagnostic training, and related research.

[0184] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications are also within the protection scope of the present invention.

Claims

1. A system for simulating and proactively identifying multi-cause short-circuit faults in cable joints, characterized in that, include: The cable joint fault simulation module is placed in a sealed test chamber and includes a cable joint body and an insulation defect parameter adjustment unit. The cable joint body includes several independently arranged phase units, each phase unit including a conductive core and a main insulation layer covering the conductive core. The cable joint fault simulation module is used to form different cable joint fault simulation conditions by configuring different cable structure and material parameters and constructing adjustable air gap defects with different parameters in the main insulation layer through the insulation defect parameter adjustment unit. A multi-inducement simulation module is used to apply various external inducements to the cable joint body inside the sealed test chamber; The data acquisition module is used to collect data on pyrolysis particles generated by the cable connector body under the action of power and external inducing factors. The advanced identification module is used to construct an advanced identification model by using the pyrolysis particle features extracted from the pyrolysis particle data and the fault condition parameters corresponding to the fault simulation conditions as model input features, to perform advanced identification of potential short-circuit faults and generate explanations, and to output risk prediction results and corresponding explanations. The control module is used to provide a human-machine interface, receive user input of fault simulation condition configuration instructions, adjust the operating status of the multi-cause simulation module, control the on / off state and current value of the test main circuit, and also record and display the pyrolysis particle data, fault condition parameters and risk prediction results.

2. The cable joint multi-cause short-circuit fault simulation and advance identification system according to claim 1, characterized in that, The insulation defect parameter adjustment unit includes: An air gap length adjustment unit includes an insulating slider and a first driving unit linked thereto. The insulating slider is disposed in a preset air gap channel within the main insulating layer, and the air gap channel extends radially along the main insulating layer. The outer surface of the insulating slider is provided with an external thread, and the opening section of the air gap channel or a mating part disposed within the air gap channel is provided with an internal thread that mates with the external thread. The first driving unit includes a first knob connected to the insulating slider. When the first knob is rotated, the insulating slider rotates in or out along the air gap channel to adjust the depth of the insulating slider entering the preset air gap and change the length of the air section in the air gap. The air gap radius adjustment unit includes several replaceable insulating sleeves. The insulating sleeves are set in the air gap channel. Their outer diameter is the same as the diameter of the air gap channel, and their inner diameter is one of various specifications. The radial cross-sectional size of the air gap can be changed by replacing the insulating sleeves with different inner diameters. The defect position adjustment unit includes a rotating sleeve, a detachable rotor flange, a pin seat, a positioning pin, and a phase retaining component. The rotating sleeve is fitted and fixed to the outside of the main insulation layer of the target phase unit. The detachable rotor flange is detachably connected to the rotating sleeve and is used to drive the main insulation layer and conductive core of the target phase unit to rotate synchronously around an axis. The detachable rotor flange has multiple positioning holes along its circumference. The positioning pin can pass through the pin seat and be inserted into any positioning hole to limit the adjustment angle of the target phase unit. The phase retaining component is used to clamp and fix the insulating outer surface of the target phase unit after the angle adjustment is completed, so that after the detachable rotor flange, positioning pin, and temporary connection structure are removed, the target phase unit still maintains a preset circumferential position and allows its conductive core end to be connected to the test main circuit.

3. The cable joint multi-cause short-circuit fault simulation and advance identification system according to claim 1 or 2, characterized in that, An outer sheath layer can also be provided outside the main insulation layer; the cable joint fault simulation module also includes a positioning support, which has mounting holes for installing each phase unit, and each phase unit is detachably connected to the corresponding mounting hole; the cable structure and material parameters include one or more of the conductor type, main insulation material and sheath type; each phase unit has terminals at both ends of the conductive core for connecting to the test main circuit.

4. The cable joint multi-cause short-circuit fault simulation and advance identification system according to claim 1 or 2, characterized in that, The multi-cause simulation module includes: The humidity simulation unit includes a spray device, a humidity sensor, and a relay. The spray device is used to spray humidification into the sealed test chamber. The humidity sensor is used to detect the humidity value in the chamber and feed it back to the control module. The relay is controlled by the control module to control the start and stop of the spray device. The temperature simulation unit includes a heater, a cooler, a temperature sensor, and a relay; the heater and cooler are used to regulate the temperature inside the sealed test chamber, the temperature sensor is used to detect the temperature value inside the chamber and feed it back to the control module, and the relay is controlled by the control module to control the start and stop of the heater and cooler; The pressure simulation unit includes a pressure block, a second drive unit, a stress sensor, and a detachable limiting support frame. The second drive unit includes a pressure rod for moving the pressure block up and down. The detachable limiting support frame is disposed within the sealed test chamber and has a guide structure that cooperates with the pressure rod to limit the vertical movement of the pressure rod and prevent the pressure block from swaying. The detachable limiting support frame also provides auxiliary support for each phase unit. The second drive unit drives the pressure block to move toward the main insulation layer to change the pressure on the main insulation layer. The stress sensor detects the pressure value and feeds it back to the control module. The ultraviolet aging simulation unit includes an ultraviolet lamp and a relay; the ultraviolet lamp is set in a sealed test chamber and is used to irradiate the cable joint body with ultraviolet light; the relay is controlled by a control module to control the start and stop of the ultraviolet lamp, and the control module is used to record the ultraviolet irradiation time of the ultraviolet lamp, and simulate different degrees of ultraviolet aging of the cable joint body by changing the ultraviolet irradiation time.

5. The cable joint multi-cause short-circuit fault simulation and advance identification system according to claim 1 or 2, characterized in that, The data acquisition module includes a pyrolysis particle collector and a sampling circuit; the output of the pyrolysis particle collector is connected to the control module and the advanced identification module via the sampling circuit; the main test circuit, consisting of a test power supply, connecting wires, and conductive cores of each phase unit, is equipped with an on / off switch, a current regulator, and a current sensor. The on / off switch is used to control the on / off state of the main circuit, the current regulator is controlled by the control module and is used to adjust the energized current value of the cable joint body, and the current sensor is used to detect the energized current value and feed it back to the control module; the control module includes a microcontroller and a computer, and the microcontroller is electrically connected to the cable joint fault simulation module, the multi-cause simulation module, the data acquisition module, and the advanced identification module, respectively.

6. The cable joint multi-cause short-circuit fault simulation and advance identification system according to claim 1 or 2, characterized in that, The advanced identification module includes: An advanced identification model is used to receive the input features of the model and establish a risk identification relationship for judging low-risk, medium-risk, or high-risk based on the correspondence between fault condition parameters, pyrolysis particle characteristics, and short-circuit fault risk levels in historical test samples. The historical test samples are samples collected under different cable structure and material parameters, insulation defect parameters, electrical operating parameters, and multiple inducing factor parameters. The cable structure and material parameters, insulation defect parameters, electrical operating parameters, and multiple inducing factor parameters constitute the fault condition parameters. An interpretation model is used to determine the key factors affecting the risk prediction results after obtaining the risk prediction results, combined with the fault condition parameters and pyrolysis particle characteristics corresponding to the current sample, and to generate an interpretation that includes the risk level, key factors and their correspondence with the insulation degradation state. The output of the advanced identification module is connected to the control module and is used to output the risk prediction results of low risk, medium risk or high risk and the corresponding explanations.

7. A method for simulating and proactively identifying multi-cause short-circuit faults in cable joints, characterized in that, Includes the following steps: S1: Place the cable joint fault simulation module inside the sealed test chamber and configure the cable structure and material parameters and insulation defect parameters of the cable joint body according to the preset fault simulation conditions. S2: Based on the pre-input multi-factor target parameters, control the multi-factor simulation module to apply humidity, temperature, pressure and ultraviolet aging factors to the cable joint body; S3: After each inducing factor reaches the preset condition, adjust the current value of the cable joint body according to the preset electrical operating parameters, and control the cable joint fault simulation module to be energized, so that the air gap defect of the main insulation layer will undergo insulation deterioration and evolve into a short circuit fault under the combined action of electric field and multiple inducing factors, generating pyrolysis particles. S4: During the power-on degradation process, pyrolysis particle data in the sealed test chamber is collected in real time, and fault condition parameters corresponding to the pyrolysis particle data are recorded simultaneously. S5: Extract pyrolysis particle features from the pyrolysis particle data, use the pyrolysis particle features and corresponding fault condition parameters as model input features, construct an advanced identification model, perform advanced identification of potential short-circuit faults and generate explanations, and output risk prediction results and corresponding explanations.

8. The method for simulating and proactively identifying multi-cause short-circuit faults in cable joints according to claim 7, characterized in that, Step S5 specifically includes: S5-1: Construct an experimental sample set based on the pyrolysis particle data and its corresponding fault condition parameters obtained in step S4, extract pyrolysis particle features from the pyrolysis particle data, use the pyrolysis particle features and fault condition parameters as model input features, and label the experimental samples as low-risk, medium-risk or high-risk based on the changes in pyrolysis particles, insulation degradation status and short-circuit fault occurrence. S5-2: Based on experimental samples with calibrated risk levels, establish the risk identification relationship between fault condition parameters, pyrolysis particle characteristics and short-circuit fault risk levels, in order to determine the short-circuit fault risk levels corresponding to pyrolysis particle responses under different cable structures, insulation defects, electrical operating states and multiple inducing factors. S5-3: During the new power-on test, the real-time obtained pyrolysis particle characteristics and corresponding fault condition parameters are input into the risk identification relationship, and the risk prediction results of low risk, medium risk or high risk are output, and the key factors that affect the risk prediction results are identified. S5-4: Generate an interpretation based on the fault condition parameters and pyrolysis particle characteristics corresponding to the risk prediction results, and store the key factors that can reflect the changes in risk level in historical samples and their correspondence with the risk level into the interpretation knowledge base for real-time interpretation generation of new samples.

9. The method for simulating and proactively identifying multi-cause short-circuit faults in cable joints according to claim 8, characterized in that, The process of establishing the risk identification relationship described in step S5-2 includes: (1) Select historical test samples with different fault simulation conditions and risk levels from the experimental sample set, and use cable structure and material parameters, insulation defect parameters, electrical operation parameters, multi-cause parameters and pyrolysis particle characteristics as candidate judgment factors; (2) Set candidate classification conditions for each candidate judgment factor; for numerical data such as current value, ultraviolet irradiation duration, temperature, humidity, pressure, air gap length, air gap radius, air gap circumferential position, particle concentration, and particle size distribution, set several candidate boundary values; for categorical data such as conductor type, main insulation layer material, sheath type, and particle type, set candidate classification conditions according to different categories or combinations of categories; respectively count the proportions of low-risk, medium-risk, and high-risk samples in the samples on both sides of each candidate classification condition to determine whether a certain working condition parameter or pyrolysis particle characteristics is suitable as a risk judgment basis, and according to the formula: ； Calculate the risk confluence of the sample set S formed under the candidate judgment factor; In the formula, S is a set of historical test samples collected according to a certain candidate judgment factor and its classification conditions. The historical test samples include cable structure and air gap defect parameters determined by the cable joint fault simulation module, current value recorded by the control module, temperature, humidity, pressure and ultraviolet irradiation duration formed by the multi-cause simulation module, and pyrolysis particle characteristics obtained by the data acquisition module; Q(S) is the degree of mixing of low-risk, medium-risk and high-risk samples in the sample set S; K is the number of risk levels; r is the risk level sequence number; pᵣ is the proportion of samples belonging to the r-th risk level in the sample set S; After obtaining Q(S), the sample set S before partitioning is divided into a first-side sample set S1 and a second-side sample set S2 according to the candidate judgment factors and their candidate partitioning conditions, and then according to the formula: ； Calculate the discrimination of the candidate judgment factor and its candidate classification conditions on the short-circuit fault risk level; In the formula, ΔQ is the reduction in risk mixing before and after the split, which indicates whether the candidate judgment factor can separate low-risk, medium-risk and high-risk samples; |S| is the number of samples in the historical test sample set S before the split; |S1| and |S2| are the number of samples falling on both sides of the split condition, respectively; Q(S1) and Q(S2) are the risk mixing of the sample sets on both sides, respectively. Select the candidate judgment factors that maximize ΔQ and their candidate division conditions as the judgment conditions in the risk identification relationship, so that the experimental samples can be gradually mapped to low risk, medium risk or high risk according to the actual working parameters and pyrolysis particle response. (3) Establish corresponding judgment paths for each judgment condition, and mark the fault condition range, pyrolysis particle characteristic range and short circuit fault risk level corresponding to each judgment path; (4) Repeat the above sub-steps (1) to (3) until the samples of different risk levels in the experimental sample set can be distinguished according to the corresponding fault condition parameters and pyrolysis particle characteristics.

10. The method for simulating and proactively identifying multi-cause short-circuit faults in cable joints according to claim 9, characterized in that, The explanation generation method includes: (1) Identification process record: Record the complete judgment process of each historical test sample in the risk identification relationship, record the fault condition parameters or pyrolysis particle characteristics used for each judgment condition, the corresponding division conditions and the judgment path of the sample, and obtain the identification process corresponding to the sample. (2) Key factor identification: Based on the identification process, candidate judgment conditions that can change the short-circuit fault risk level of the sample are searched, and the key factors corresponding to the candidate judgment conditions are identified; based on the data type, the key factors are divided into numerical key factors and categorical key factors. (3) Numerical Key Factor Analysis: For the numerical key factors among the current value, UV irradiation duration, temperature, humidity, pressure, air gap parameters, particle concentration or particle size distribution, the numerical key factors are adjusted beyond the division value according to the corresponding candidate judgment conditions to form an adjusted sample x′. The adjusted sample x′ is then re-inputted into the advanced identification model to verify whether the short-circuit fault risk level has changed before and after the adjustment. The following formula is used to measure the degree of closeness between the current test state and the adjacent risk level state, and the difference between sample x and adjusted sample x′ is calculated: ； In the formula, sample x is the current test sample to be interpreted, corresponding to a set of actual operating parameters and pyrolysis particle characteristics recorded by the control module; adjusted sample x′ is a reference sample formed by adjusting a certain numerical key factor to the adjacent risk level while keeping other test conditions unchanged; d(x,x′) is the difference between the two, used to represent the degree of proximity of the current test state to the risk level change state; p is the total number of numerical key factors participating in the comparison; j is the j-th numerical key factor; xⱼ and x′ⱼ are the values ​​of the factor in sample x and adjusted sample x′, respectively; wⱼ is the scaling factor of the j-th factor, used to enable data from different units to be compared in the same difference. From the adjusted samples that can change the risk level of short circuit faults, select the adjusted sample with the smallest difference from sample x, and use the numerical key factors, adjustment direction and adjustment range corresponding to the adjusted sample as the basis for interpretation to explain which actual working conditions or pyrolysis particle characteristics mainly affect the current risk level. (4) Processing of Category-based Key Factors: For the category-based key factors in conductor type, main insulation material, sheath type or particle type, record the category value of the category-based key factor and its classification conditions in the identification process, use it as the correlation condition of the risk prediction result, and do not make continuous numerical adjustments to the category-based key factors. (5) Explanation of knowledge base construction: Numerical key factors that can change the risk prediction results, adjustment direction, adjustment magnitude, and corresponding categorical key factors and risk levels are stored in the explanation knowledge base, and risk correlation relationships that meet the preset threshold conditions are extracted from it. (6) Real-time interpretation generation: For a new sample, historical samples with the same or similar risk association conditions are selected in the interpretation knowledge base according to the categorical characteristics of the new sample. Then, similar historical samples are determined according to the comprehensive difference between numerical key factors. An interpretation is generated based on the risk association relationship corresponding to the similar historical samples. The interpretation includes the risk prediction result, the key factors that affect the risk prediction result, and the adjustment direction and adjustment range of the numerical key factors.

Images