Multi-dimensional inductive leakage protection device

Abstract

The application relates to the technical field of electric leakage protection, and discloses a multi-dimensional induction type electric leakage protection device which is suitable for electrical equipment and comprises a thermal infrared image acquisition module, a digital twin module, an audio acquisition module, a light intensity acquisition module and an analysis module. The thermal infrared image acquisition module is used for acquiring thermal infrared image data. Then, the digital twin module is used for establishing a digital twin model. The audio acquisition module and the light intensity acquisition module are used for respectively acquiring audio information data and light intensity information data. Finally, the analysis module is used for analyzing the audio information data and the light intensity information data, determining whether an electric arc is generated, and if the electric arc is generated, analyzing the thermal infrared image data to determine the position of the electric arc, and analyzing the audio information data, the light intensity information data and the position of the electric arc to obtain an electric arc damage index. The electric arc damage index can be used for accurately evaluating the damage degree of the electric arc to the electrical equipment from multiple dimensions, thereby providing a basis for the maintenance of the electrical equipment.

Description

Technical Field

[0001] This invention relates to the field of leakage current protection technology, specifically to a multi-dimensional inductive leakage current protection device. Background Technology

[0002] In power systems, electrical equipment is a core component, and its stable operation is directly related to the safety and reliability of the entire power system. However, in actual operation, electrical equipment is inevitably affected by various factors, among which electric arc faults are relatively common and extremely dangerous. The generation of electric arcs is often accompanied by high temperature, strong light, and violent energy release, which can cause serious damage to electrical equipment in a very short time. It may not only cause partial burnout and insulation failure, but may even cause major safety accidents such as equipment explosions and fires, seriously threatening the stable operation of the power system and the safety of people's lives.

[0003] Currently, most traditional detection methods rely on data from a single sensor, such as monitoring changes in the surface temperature of an equipment using only a temperature sensor, or detecting abnormal fluctuations in current using only a current sensor.

[0004] However, while these single-dimensional detection methods can detect signs of arcing faults to some extent, they often fail to accurately determine whether an arc has actually occurred and the specific location and severity of the arcing fault due to their lack of comprehensiveness and systematicity. For example, in some cases, the increase in surface temperature of equipment may be due to environmental factors or load changes, rather than an arcing fault. Furthermore, it is difficult to accurately determine the specific location of the arc inside the equipment based solely on abnormal current fluctuations, thus failing to provide accurate guidance for subsequent maintenance and repair. Summary of the Invention

[0005] The purpose of this invention is to provide a multi-dimensional inductive leakage current protection device to solve the above-mentioned technical problems.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A multi-dimensional inductive leakage current protection device, suitable for electrical equipment, the leakage current protection device comprising:

[0008] Thermal infrared image acquisition module, used to acquire thermal infrared image data of electrical equipment;

[0009] The digital twin module is used to create a digital twin model based on the internal structure of electrical equipment; and to establish a one-to-one correspondence between the positions of each point in the thermal infrared image data and the positions of each point in the plan view of the digital twin model.

[0010] An audio acquisition module, installed inside electrical equipment, is used to collect audio information data;

[0011] A light intensity acquisition module is installed inside electrical equipment to collect light intensity information data inside the electrical equipment.

[0012] The analysis module is used to analyze audio and light intensity data to determine whether an electric arc has occurred. If an electric arc has occurred, the module analyzes the thermal infrared image data to determine the location of the electric arc in the digital twin model. Then, based on the audio and light intensity data of the electric arc and the location of the electric arc in the digital twin model, the module analyzes the data to obtain the electric arc damage index.

[0013] As a further aspect of the present invention: the process for determining whether an electric arc is generated is as follows:

[0014] S1: Obtain the real-time light radiation intensity through the light intensity acquisition module; then, analyze the light radiation intensity to obtain a first judgment index for determining whether the current time point is a characteristic time point; if the current time point is a characteristic time point, proceed to step S2; otherwise, repeat step S1.

[0015] S2: The recognition module identifies the audio information data of the preset time period after the feature time point and determines whether there is feature audio within the preset time period; if feature audio exists, the real-time frequency, volume and duration of each feature audio are obtained; proceed to step S3; otherwise, repeat step S1.

[0016] S3: Analyze the real-time frequency, volume, and duration of each characteristic audio signal to obtain a characteristic audio risk index used to determine whether an electric arc has occurred.

[0017] As a further aspect of the present invention: In S1, by formula:

[0018] ;

[0019] Calculate the first judgment index ;

[0020] in, As the first judgment function, when hour, ;when hour, ; The current time; The preset interval duration; The intensity of light radiation at the current time; The light radiation intensity at the moment corresponding to the basic time interval past the current time; Preset a contrast value for changes in light intensity.

[0021] As a further aspect of the present invention: the process for determining whether the current time is a characteristic time point is as follows:

[0022] when At that time, the current time is the characteristic time point;

[0023] when At that time, the current time is not a characteristic time point.

[0024] As a further aspect of the present invention: In step S2, the analysis module includes an analysis unit and an identification unit. The identification unit is a trained convolutional neural network model, used to identify the presence of characteristic audio and various characteristic audio information data based on the audio information data. The characteristic audio information data includes frequency, volume, and duration.

[0025] As a further aspect of the present invention: In S3, by formula:

[0026] ;

[0027] Calculate the characteristic audio risk index ;

[0028] in, This represents the number of feature audio segments within a preset time period following the feature time point. ; This is a curve showing the frequency changing over time. For the first The start time of the segment's characteristic audio; For the first The end time of the segment's characteristic audio; For the first The maximum frequency of the segment's characteristic audio; For the first The maximum volume of the segment's characteristic audio; For the first The duration of segment-specific audio; To remove the unit coefficient; The first weighting coefficient; This is the second weighting coefficient; This is the third weighting coefficient; This is the fourth weighting coefficient; This is the first preset constant; This is the second preset constant; This is the third preset constant; This is the fourth preset constant.

[0029] As a further aspect of the present invention: the process for determining whether an electric arc is generated is as follows:

[0030] when At this characteristic time point, an electric arc is generated;

[0031] when At that time, no electric arc was generated at this characteristic time point.

[0032] As a further aspect of the present invention: when The process for obtaining the arc damage index is as follows:

[0033] S10: By analyzing the thermal infrared image data within a preset time period after the characteristic time point of the electric arc generation, the location with the greatest temperature change in the thermal infrared image data within the preset time period is obtained as the characteristic location.

[0034] S20: Determine the preset light radiation intensity and the preset structural complexity within the preset range of the feature location based on the feature location;

[0035] S30: Based on the preset light radiation intensity, the structural complexity within the preset range of the characteristic location, the light radiation intensity at the characteristic time point, and the characteristic audio risk index, the arc damage index is obtained through analysis.

[0036] As a further aspect of the present invention: through the formula:

[0037] ;

[0038] Calculating the arc damage index ;

[0039] in, The preset light radiation intensity at the characteristic location, The structural complexity is preset within a certain range for the feature locations; This is the first weighting coefficient; This is the second weighting coefficient; This is the third weighting coefficient; This is the first preset constant; This is the second preset constant; This is the third preset constant.

[0040] The beneficial effects of this invention are:

[0041] This invention first acquires thermal infrared image data of electrical equipment using a thermal infrared image acquisition module; then, a digital twin model is established based on the internal structure of the electrical equipment using a digital twin module; and a one-to-one correspondence is established between the positions of each point in the thermal infrared image data and the positions of each point in the plan view of the digital twin model; next, audio information data and light intensity information data inside the electrical equipment are acquired by an audio acquisition module and a light intensity acquisition module installed inside the electrical equipment, respectively; finally, an analysis module analyzes the audio information data and light intensity information data to determine whether an electric arc has occurred. If an electric arc has occurred, further analysis of the thermal infrared image data determines the location of the electric arc in the digital twin model. The location of the digital twin model is analyzed based on audio and light intensity data of the generated arc, as well as the arc's position within the digital twin model, to obtain an arc damage index. This allows for accurate assessment of the extent of arc damage to electrical equipment from multiple dimensions, providing a clear and reliable basis for the maintenance and repair of electrical equipment. Maintenance personnel can quickly determine whether equipment requires immediate repair, component replacement, or routine inspection based on the damage index. This helps to rationally plan maintenance schedules, improve maintenance efficiency, reduce the risk of equipment failure or even safety accidents caused by arc damage, and ensure the stable operation of electrical equipment and the safety and reliability of the entire power system. Attached Figure Description

[0042] The invention will now be further described with reference to the accompanying drawings.

[0043] Figure 1 This is a system module framework diagram of one embodiment of the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] Please see Figure 1 As shown, in one embodiment, a multi-dimensional inductive leakage current protection device is provided, suitable for electrical equipment, including:

[0046] Thermal infrared image acquisition module, used to acquire thermal infrared image data of electrical equipment;

[0047] The digital twin module is used to create a digital twin model based on the internal structure of electrical equipment; and to establish a one-to-one correspondence between the positions of each point in the thermal infrared image data and the positions of each point in the plan view of the digital twin model.

[0048] An audio acquisition module, installed inside electrical equipment, is used to collect audio information data;

[0049] A light intensity acquisition module is installed inside electrical equipment to collect light intensity information data inside the electrical equipment.

[0050] The analysis module is used to analyze audio information data and light intensity information data to determine whether an electric arc has occurred. If an electric arc has occurred, the analysis is performed based on thermal infrared image data to determine the location of the electric arc in the digital twin model. Then, based on the audio information data, light intensity information data, and location of the electric arc in the digital twin model, the electric arc damage index is obtained.

[0051] Through the above technical solution, this embodiment first acquires thermal infrared image data of the electrical equipment using a thermal infrared image acquisition module; then, a digital twin model is established based on the internal structure of the electrical equipment using a digital twin module; and a one-to-one correspondence is established between the positions of each point in the thermal infrared image data and the positions of each point in the plan view of the digital twin model; next, audio information data and light intensity information data inside the electrical equipment are acquired by an audio acquisition module and a light intensity acquisition module installed inside the electrical equipment, respectively; finally, the audio information data and light intensity information data are analyzed by an analysis module to determine whether an electric arc has occurred. If an electric arc has occurred, further analysis based on the thermal infrared image data is used to determine the generation of the arc. The location of the electric arc in the digital twin model is analyzed, along with audio and light intensity data of the arc, to obtain an arc damage index. This allows for accurate assessment of the extent of arc damage to electrical equipment from multiple dimensions, providing a clear and reliable basis for the maintenance and repair of electrical equipment. Maintenance personnel can quickly determine whether equipment requires immediate repair, component replacement, or routine inspection based on the damage index. This helps to rationally plan maintenance, improve maintenance efficiency, reduce the risk of equipment failure or even safety accidents caused by arc damage, and ensure the stable operation of electrical equipment and the safety and reliability of the entire power system.

[0052] As one embodiment of the present invention, the process for determining whether an electric arc is generated is as follows:

[0053] S1: Obtain the real-time light radiation intensity through the light intensity acquisition module; then, analyze the light radiation intensity to obtain a first judgment index for determining whether the current time point is a characteristic time point; if the current time point is a characteristic time point, proceed to step S2; otherwise, repeat step S1.

[0054] S2: The recognition module identifies the audio information data of the preset time period after the feature time point and determines whether there is feature audio within the preset time period; if feature audio exists, the real-time frequency, volume and duration of each feature audio are obtained; proceed to step S3; otherwise, repeat step S1.

[0055] S3: Analyze the real-time frequency, volume and duration of each characteristic audio to obtain the characteristic audio risk index used to determine whether an electric arc has occurred;

[0056] Through the above technical solution, this embodiment first obtains the real-time light radiation intensity through the light intensity acquisition module; then, by analyzing the light radiation intensity, a first judgment index is obtained to determine whether the current time point is a characteristic time point; if the current time point is a characteristic time point, the audio information data of a preset time period after the characteristic time point is identified by the recognition module to determine whether there is characteristic audio within the preset time period; if characteristic audio exists, the real-time frequency, volume, and duration of each characteristic audio are obtained; based on the real-time frequency, volume, and duration of each characteristic audio, a characteristic audio risk index is obtained to determine whether an electric arc is generated.

[0057] As one embodiment of the present invention, in S1, the formula is:

[0058] ;

[0059] Calculate the first judgment index ;

[0060] in, As the first judgment function, when hour, ;when hour, ; The current time; The preset interval duration; The intensity of light radiation at the current time; The light radiation intensity at the moment corresponding to the basic time interval past the current time; Preset contrast values ​​for changes in light intensity;

[0061] The process for determining whether the current time is a characteristic time point is as follows:

[0062] when At that time, the current time is the characteristic time point;

[0063] when At that time, the current time is not a characteristic time point;

[0064] Through the above technical solution, this embodiment The change in light radiation intensity at the current time and the time interval prior to the current time; The rate of change of light radiation intensity at the current time and the time interval prior to the current time; The difference between the rate of change of light radiation intensity at the current time and the preset comparison value of the light intensity change at the time corresponding to the preset interval elapsed since the current time is given in the formula. In the first judgment function In It refers to This is used to determine whether the rate of change of light radiation intensity at the current time and the time interval preceding the current time exceeds a preset comparison value for light intensity change. If the rate of change of light radiation intensity at the current time and the time interval preceding the current time exceeds the preset comparison value for light intensity change, an electric arc may be present. Therefore, the current time is a characteristic time point; when This indicates that the rate of change of light radiation intensity at the current time and the time interval preceding the current time has not exceeded the preset comparison value for light intensity change, and therefore no electric arc exists. Therefore, the current time is not a characteristic time point;

[0065] It should be noted that the preset interval duration As a preset value, The specific values ​​are derived from experience and will not be detailed here.

[0066] In one embodiment of the present invention, in step S2, the analysis module includes an analysis unit and an identification unit. The identification unit is a trained convolutional neural network model, used to identify the presence of characteristic audio and the characteristic audio information data based on the audio information data. The characteristic audio information data includes frequency, volume, and duration.

[0067] It should be noted that the training process of the convolutional neural network model is existing technology and will not be described in detail here; the feature audio is audio within a preset frequency range, which is determined based on the frequency of the electric arc, and the determination process is existing technology and will not be described in detail here.

[0068] As one embodiment of the present invention, in S3, the formula is:

[0069] ;

[0070] Calculate the characteristic audio risk index ;

[0071] in, This represents the number of feature audio segments within a preset time period following the feature time point. ; This is a curve showing the frequency changing over time. For the first The start time of the segment's characteristic audio; For the first The end time of the segment's characteristic audio; For the first The maximum frequency of the segment's characteristic audio; For the first The maximum volume of the segment's characteristic audio; For the first The duration of segment-specific audio; To remove the unit coefficient; The first weighting coefficient; This is the second weighting coefficient; This is the third weighting coefficient; This is the fourth weighting coefficient; This is the first preset constant; This is the second preset constant; This is the third preset constant; This is the fourth preset constant;

[0072] Through the above technical solution, this embodiment For the first Cumulative frequency values ​​of segment-specific audio; For the first The average frequency of the characteristic audio segment; because during the generation and continuation of the electric arc, the motion state of charged particles inside the arc is closely related to the frequency. The higher the average frequency, the more frequent and intense the vibration, collision, and other movements of the charged particles. This directly reflects the more intense dynamic changes in the electric and magnetic fields inside the arc, thus indicating that the arc is in a more active state with more rapid energy release; therefore, the characteristic audio risk index The larger; the first Maximum frequency of segment feature audio The larger the value, the more pronounced the instantaneous extreme situations occurring during the movement of charged particles within the electric arc during the corresponding time period of the audio segment; the more intense the vibration or high-speed collision of charged particles at a particular moment, the higher the degree. This instantaneous high-intensity motion is often accompanied by rapid changes in the local electric and magnetic fields of the electric arc, which may trigger brief abrupt changes in the arc's shape or instantaneous peaks in energy release; therefore, the characteristic audio risk index... The larger; the first Maximum volume of segment characteristic audio The higher the sound pressure level, the higher the sound pressure level generated by the electric arc within the corresponding time interval of the audio segment. Physically, the intensity of energy release from the electric arc is closely related to the sound pressure level. A higher sound pressure level means a larger scale of energy released instantaneously by the electric arc. This high-intensity energy release causes more drastic changes in the physical state of the arc's interior and surrounding environment, such as intense heating of the air and increased ionization. Simultaneously, a high volume also reflects potentially more violent arc flickering and other unstable phenomena during the arc's combustion process. These unstable states increase the risk of arc runaway, potentially leading to more serious electrical faults or even safety accidents. Therefore, the characteristic audio risk index... The larger; the first Duration of segmented audio features The larger the value, the more continuously the electric arc is active within the corresponding time span of the audio segment. From the perspective of the arc's dynamic characteristics, a longer duration means continuous movement of charged particles within the arc, and the sustained high level of interaction between the electric and magnetic fields. This leads to continuous heating and ionization of the surrounding medium (such as air), causing the arc's energy to accumulate. With this continuous energy accumulation, the arc's stability gradually decreases, and its shape and position may undergo unpredictable changes, such as shifting, elongating, or even splitting. Furthermore, prolonged arc activity causes more severe thermal and electrical damage to the insulation materials of surrounding equipment, reducing its insulation performance and increasing the probability of equipment failure. Therefore, the characteristic audio risk index... The larger.

[0073] It should be noted that the unit coefficient is removed. First weighting coefficient Second weighting coefficient Third weighting coefficient Fourth weighting coefficient First preset constant Second preset constant Third preset constant and the fourth preset constant These are preset values, obtained based on experience, and will not be detailed here.

[0074] It should be noted that the preset time period is a preset value, and the range of values ​​is within [specified range]. The specific values ​​are derived from experience and will not be detailed here.

[0075] As one embodiment of the present invention, the process for determining whether an electric arc is generated is as follows:

[0076] when At this characteristic time point, an electric arc is generated;

[0077] when At that time point, no electric arc was generated;

[0078] Through the above technical solution, in this embodiment when At this characteristic time point, an electric arc is generated, and the power is immediately cut off; when At that time point, no electric arc was generated, and the machine continued to operate.

[0079] As one embodiment of the present invention, when The process for obtaining the arc damage index is as follows:

[0080] S10: By analyzing the thermal infrared image data within a preset time period after the characteristic time point of the electric arc generation, the location with the greatest temperature change in the thermal infrared image data within the preset time period is obtained as the characteristic location.

[0081] S20: Determine the preset light radiation intensity and the preset structural complexity within the preset range of the feature location based on the feature location;

[0082] S30: Based on the preset light radiation intensity, the structural complexity within the preset range of the characteristic location, the light radiation intensity at the characteristic time point, and the characteristic audio risk index, the arc damage index is obtained through analysis.

[0083] Through the above technical solution, this embodiment analyzes thermal infrared image data within a preset time period after the characteristic time point of arc generation to obtain the location with the greatest temperature change within that preset time period as the characteristic location; based on the characteristic location, it obtains a preset light radiation intensity and a preset structural complexity within a preset range of the characteristic location; finally, it analyzes the preset light radiation intensity, the structural complexity within the preset range of the characteristic location, the light radiation intensity at the characteristic time point, and the characteristic audio risk index to obtain the arc damage index; the arc damage index provides a clear understanding of the degree of damage to the electrical equipment caused by the arc, providing a clear and reliable basis for the maintenance and repair of the electrical equipment; maintenance personnel can quickly determine whether the equipment needs immediate repair, component replacement, or routine inspection based on the damage index, which helps to rationally arrange maintenance plans, improve maintenance efficiency, reduce the risk of equipment failure or even safety accidents caused by arc damage, and ensure the stable operation of electrical equipment and the safety and reliability of the entire power system;

[0084] It should be noted that the process of obtaining the location of the largest temperature change in the thermal infrared image data within the preset time period as the feature location is existing technology and will not be described in detail here.

[0085] It should be noted that the preset range of the feature location is a circular range with the feature location as the center and a fixed length (based on the staff's preset) as the radius. The structural complexity within the preset range of the feature location is evaluated based on the digital twin model within the preset range of the feature location. This evaluation process is existing technology and will not be described in detail here.

[0086] As one embodiment of the present invention, the process of obtaining the preset light radiation intensity is as follows:

[0087] S100: Before starting work, place the light sources with fixed parameters in each position of the electrical equipment in sequence;

[0088] S200: The light intensity acquisition module sequentially acquires the light radiation intensity of a light source with fixed parameters at various locations.

[0089] S300: Sets the light radiation intensity of a light source with fixed parameters at each location to the preset light radiation intensity at each location;

[0090] Through the above technical solution, this embodiment first places a light source with fixed parameters at various positions of the electrical equipment in sequence before operation; then, the light intensity acquisition module sequentially acquires the light radiation intensity of the light source with fixed parameters at each position; finally, the light radiation intensity of the light source with fixed parameters at each position is set to the preset light radiation intensity of each position; by setting the preset light radiation intensity of each position, it is convenient to compare and analyze the light radiation intensity with the light radiation intensity of the characteristic position at the characteristic time point, so as to accurately evaluate the arc intensity.

[0091] It should be noted that the installation position of the light intensity acquisition module is set based on experience. The light radiation intensity of the light source with fixed parameters at each position should reach the preset light intensity value. The specific position will not be detailed here.

[0092] As one embodiment of the present invention, the formula is as follows:

[0093] ;

[0094] Calculating the arc damage index ;

[0095] in, The preset light radiation intensity at the characteristic location, The structural complexity is preset within a certain range for the feature locations; This is the first weighting coefficient; This is the second weighting coefficient; This is the third weighting coefficient; This is the first preset constant; This is the second preset constant; This is the third preset constant;

[0096] Through the above technical solution, the characteristic audio risk index of this embodiment The larger the value, the more intense the arc and the greater the degree of damage; therefore, the arc damage index... The larger; This refers to the difference between the light radiation intensity at a specific location and a specific time point, and the preset light radiation intensity at that location. The larger the value, the higher the intensity of the electric arc, and therefore the greater the degree of damage. (Earth arc damage index) Larger; structural complexity within the preset range of feature locations The higher the value, the greater the degree of damage; arc damage index. The larger the value, the more accurate the assessment of arc intensity. This formula allows for the evaluation of the damage level of electrical equipment from multiple dimensions, providing a clear and reliable basis for the maintenance and repair of electrical equipment. Maintenance personnel can quickly determine whether equipment needs immediate repair, component replacement, or routine inspection based on the damage index. This helps to rationally arrange maintenance plans, improve maintenance efficiency, reduce the risk of equipment failure or even safety accidents caused by arc damage, and ensure the stable operation of electrical equipment and the safety and reliability of the entire power system.

[0097] It should be noted that the first weight coefficient Weighting coefficient No. 2 Weighting coefficient No. 3 Preset constant No. 1 Preset constant No. 2 and the third preset constant These are preset values, obtained based on experience, and will not be detailed here.

[0098] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A multi-dimensional inductive leakage current protection device, suitable for electrical equipment, characterized in that, The leakage protection device includes: Thermal infrared image acquisition module, used to acquire thermal infrared image data of electrical equipment; The digital twin module is used to create a digital twin model based on the internal structure of electrical equipment; and to establish a one-to-one correspondence between the positions of each point in the thermal infrared image data and the positions of each point in the plan view of the digital twin model. An audio acquisition module, installed inside electrical equipment, is used to collect audio information data; A light intensity acquisition module is installed inside electrical equipment to collect light intensity information data inside the electrical equipment. The analysis module is used to analyze audio information data and light intensity information data to determine whether an electric arc has occurred. If an electric arc has occurred, the analysis is performed based on thermal infrared image data to determine the location of the electric arc in the digital twin model. Then, based on the audio information data, light intensity information data, and location of the electric arc in the digital twin model, the electric arc damage index is obtained. The process for determining whether an electric arc has occurred is as follows: S1: Obtain the real-time light radiation intensity through the light intensity acquisition module; then, analyze the light radiation intensity to obtain a first judgment index for determining whether the current time point is a characteristic time point; if the current time point is a characteristic time point, proceed to step S2; otherwise, repeat step S1. S2: The recognition module identifies the audio information data of the preset time period after the feature time point and determines whether there is feature audio within the preset time period; if feature audio exists, the real-time frequency, volume and duration of each feature audio are obtained; proceed to step S3; otherwise, repeat step S1. S3: Analyze the real-time frequency, volume, and duration of each characteristic audio signal to obtain a characteristic audio risk index used to determine whether an electric arc has occurred.

2. The multi-dimensional inductive leakage current protection device according to claim 1, characterized in that, In S1, by the formula: ； Calculate the first judgment index ; in, As the first judgment function, when hour, ;when hour, ; The current time; The preset interval duration; The intensity of light radiation at the current time; The light radiation intensity at the moment corresponding to the basic time interval past the current time; Preset a contrast value for changes in light intensity.

3. The multi-dimensional inductive leakage current protection device according to claim 2, characterized in that, The process for determining whether the current time is a characteristic time point is as follows: when At that time, the current time is the characteristic time point; when At that time, the current time is not a characteristic time point.

4. A multi-dimensional inductive leakage current protection device according to claim 3, characterized in that, In step S2, the analysis module includes an analysis unit and an identification unit. The identification unit is a trained convolutional neural network model used to identify the presence of characteristic audio and the characteristic audio information data based on the audio information data. The characteristic audio information data includes frequency, volume, and duration.

5. A multi-dimensional inductive leakage current protection device according to claim 4, characterized in that, In S3, using the formula: ； Calculate the characteristic audio risk index ; in, This represents the number of feature audio segments within a preset time period following the feature time point. ; This is a curve showing the frequency changing over time. For the first The start time of the segment's characteristic audio; For the first The end time of the segment's characteristic audio; For the first The maximum frequency of the segment's characteristic audio; For the first The maximum volume of the segment's characteristic audio; For the first The duration of segment-specific audio; To remove the unit coefficient; The first weighting coefficient; This is the second weighting coefficient; This is the third weighting coefficient; This is the fourth weighting coefficient; This is the first preset constant; This is the second preset constant; This is the third preset constant; This is the fourth preset constant.

6. A multi-dimensional inductive leakage current protection device according to claim 5, characterized in that, The process for determining whether an electric arc has occurred is as follows: when At this characteristic time point, an electric arc is generated; when At that time, no electric arc was generated at this characteristic time point.

7. A multi-dimensional inductive leakage current protection device according to claim 6, characterized in that, when The process for obtaining the arc damage index is as follows: S10: By analyzing the thermal infrared image data within a preset time period after the characteristic time point of the electric arc generation, the location with the greatest temperature change in the thermal infrared image data within the preset time period is obtained as the characteristic location. S20: Determine the preset light radiation intensity and the preset structural complexity within the preset range of the feature location based on the feature location; S30: Based on the preset light radiation intensity, the structural complexity within the preset range of the characteristic location, the light radiation intensity at the characteristic time point, and the characteristic audio risk index, the arc damage index is obtained through analysis.

8. A multi-dimensional inductive leakage current protection device according to claim 7, characterized in that, Through the formula: ； Calculating the arc damage index ; in, The preset light radiation intensity at the characteristic location, The structural complexity is preset within a certain range for the feature locations; This is the first weighting coefficient; This is the second weighting coefficient; This is the third weighting coefficient; This is the first preset constant; This is the second preset constant; This is the third preset constant.

Images