Electric arc hazard protection method for 0.4 kV low-voltage uninterruptible power operation

Abstract

The invention discloses a 0.4 kV low-voltage uninterruptible power operation electric arc hazard protection method, and relates to the technical field of electric power operation protection, and the method comprises the following steps: analyzing electric arc formation mechanisms in different working environments, identifying the potential risk factor combination of an electric arc to an operator, collecting typical electric arc parameters in the electric arc formation process, and calculating the electric arc hazard of the operator. Based on this, multiple hazard characteristic tests are carried out; inputting a test result into a pre-trained machine learning model, and outputting protection grade requirements of the protection appliance in different environments; according to protection grade requirements output by the model, protection appliance wearing specifications of the low-voltage uninterruptible power operation personnel in different working environments are formulated, so that targeted and efficient operation protection is realized, and the safety of the operation personnel is guaranteed; according to the method, the scientificity of protection measures is greatly improved, the overall safety efficiency of low-voltage non-power-cut operation is remarkably improved, the protection wearing decision process of an operator is simplified, and the operation efficiency and safety are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power operation protection, and in particular to a method for protecting arc hazards in 0.4kV low-voltage non-stop operation. Background Art

[0002] Low-voltage, non-stop work is a common and necessary method for the operation and maintenance of power systems, particularly important for ensuring power supply continuity and improving the operating efficiency of power equipment. However, in such environments, arcing can easily occur due to equipment short circuits, poor contact, or operational errors, posing a serious threat to the safety of operators and the stable operation of equipment.

[0003] An electric arc is a high-temperature, high-energy discharge phenomenon characterized by: high temperatures (reaching thousands to tens of thousands of degrees Celsius), which can cause equipment meltdowns, fires, or severe burns; high-intensity optical radiation, including ultraviolet and infrared rays, which can damage workers' eyesight; the expansion of high-pressure gas and the explosive shock wave that can cause mechanical damage to equipment and the human body; and the release of toxic gases (such as ozone and carbon monoxide), which can pose a long-term threat to workers' health. Existing low-voltage arc hazard protection measures are primarily focused on high-voltage or high-risk environments. However, in 0.4kV low-voltage, non-stop power operations, this hazard is often overlooked due to the lower arc energy. However, low-voltage arcs can also pose significant hazards to workers and equipment.

[0004] Furthermore, traditional arc protection methods are often based on static protection standards and are difficult to adapt to the dynamic changes in complex working environments. Therefore, as the demand for safety and intelligence in modern working environments continues to increase, this paper proposes a 0.4kV low-voltage, non-stop working arc hazard protection method to achieve more scientific safety protection standards. Summary of the Invention

[0005] To address the problems in the existing technology, the present invention provides a 0.4kV low-voltage non-stop operation arc hazard protection method. It can output the optimal protective equipment level requirements based on the real-time changes in different working environments, avoiding the potential shortcomings of traditional static protection measures and providing workers with a more flexible and accurate protection plan. The specific technical solution is as follows:

[0006] A method for protecting against arc hazards during 0.4kV low-voltage non-stop operation, comprising:

[0007] Analyze the arc formation mechanism in different working environments and identify the potential risk factors of arc to workers;

[0008] Based on the identified potential risk factors of arc to workers, collect typical arc parameters during the arc formation process;

[0009] Based on the collected typical arc parameters, multiple preset hazard characteristic tests are carried out. The results of multiple hazard characteristic tests are input into the pre-trained machine learning model to output multiple preset protection level requirements for protective equipment;

[0010] Based on the protection level requirements of multiple preset protective equipment output by the machine learning model, the wearing standards of protective equipment for workers performing low-voltage non-stop operations in different working environments are formulated.

[0011] Preferably, the combination of potential risk factors of arcing to operators is identified, including:

[0012] Based on the fault tree analysis method, a probability tree diagram of arc occurrence is constructed to identify all combinations of factors that lead to high-probability arc events.

[0013] Preferably, the fault tree analysis method includes:

[0014] Multiple failure mechanisms are preset and environmental factors are introduced;

[0015] The environmental factors include a temperature change index and a humidity change index.

[0016] Preferably, the method for obtaining the temperature change index includes:

[0017] In a fixed time window, obtain the temperature data collected at multiple time points, and then calculate the average value to get T avg , calculate the difference between the maximum temperature and the minimum temperature in the temperature data at multiple time points to obtain the temperature change amplitude ΔT;

[0018] The temperature change index calculation formula is:

[0019]

[0020] Where δ is the preset proportional coefficient, I T is the temperature change index, f(T) is the relationship function between temperature change and material deformation;

[0021] There are standard materials preset in the working environment, and the relationship function formula between temperature change and material deformation is:

[0022]

[0023] Where L0 represents the initial length of the standard material, ΔL represents the length change of the standard material in a fixed time window, and β represents the dynamic influence coefficient.

[0024] Preferably, the dynamic influence coefficient satisfies the following formula:

[0025]

[0026] Where E represents the elastic modulus of the standard material, α represents the linear thermal expansion coefficient of the standard material, and v represents the Poisson's ratio of the standard material, which is the ratio of the lateral deformation to the longitudinal deformation of the standard material when subjected to force.

[0027] Preferably, the method for obtaining the humidity change index includes:

[0028] In a fixed time window, the humidity data collected at multiple time points are obtained, and then the absolute value of the difference between the humidity data at two adjacent time points is calculated to obtain the humidity change amplitude value ΔH(t), which represents the humidity change amplitude value at time point t;

[0029] Calculate the average value of all humidity change amplitudes ΔH(t) to obtain ΔH avg , calculate the standard deviation of all humidity change amplitudes ΔH(t) to obtain σ ΔH ;

[0030] The autocorrelation function is used to evaluate the autocorrelation of the humidity data time series at multiple preset lag time points. The calculation formula is as follows:

[0031]

[0032] H(t) represents the humidity data at time point t, represents the average value of humidity data in a fixed time window, N represents the total number of data points, ki represents the i-th preset lag time point, and ACF(ki) represents the autocorrelation at the i-th preset lag time point;

[0033] Perform weighted averaging on the autocorrelation ACF(ki) at multiple preset different lag time points to obtain the autocorrelation coefficient A;

[0034] The calculation formula of the humidity change index is as follows:

[0035] I H =w1·ΔH avg +w2·σ ΔH +w3·A;

[0036] w1, w2, w3 are all preset non-zero adjustment coefficients, and the sum of w1, w2, w3 is one, I H Indicates the humidity change index.

[0037] Preferably, the typical arc parameters refer to a plurality of common and representative preset physical parameters in the arc formation process.

[0038] Preferably, the machine learning model is a convolutional neural network model.

[0039] Preferably, the protective gear wearing specifications for workers performing low-voltage non-stop operations in different working environments are formulated based on the multiple preset protective gear protection level requirements output by the machine learning model, including:

[0040] Obtain the standard deviation of the temperature variation index and the standard deviation of the humidity variation index of the working environment in multiple fixed time windows. If the standard deviation of the temperature variation index and the standard deviation of the humidity variation index of the working environment do not exceed the corresponding preset safety fluctuation threshold, a normal signal is generated. If the standard deviation of the temperature variation index and the standard deviation of the humidity variation index of the working environment do not exceed the corresponding preset safety fluctuation threshold, an abnormal signal is generated.

[0041] When a normal signal is generated, the protective gear wearing specifications for workers performing low-voltage non-stop operations are the protection levels of multiple preset protective gear output by the machine learning model;

[0042] When an abnormal signal is generated, the protective equipment wearing specifications for workers performing low-voltage non-stop operations are the protection levels of multiple preset protective equipment output by the optimized machine learning model.

[0043] Preferably, the optimization principle of the multiple preset protection levels of protective equipment output by the optimized machine learning model is as follows:

[0044]

[0045] In the formula, B1 represents the standard deviation of the temperature change index, B2 represents the standard deviation of the humidity change index, Y1 represents the safety fluctuation threshold corresponding to the standard deviation of the temperature change index, Y2 represents the safety fluctuation threshold corresponding to the standard deviation of the humidity change index, r represents the preset tuning coefficient, Dj represents the protection level of the j-th preset protective equipment output by the machine learning model, and DYj represents the protection level of the j-th preset protective equipment after optimization.

[0046] A computer-readable storage medium includes a stored program, wherein when the program is executed, the method described in the invention is controlled by a device where the computer-readable storage medium is located.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] 1. In the present invention, the dynamic risk of the working environment is comprehensively evaluated by obtaining the temperature change index and humidity change index in the working environment and combining them with typical arc parameters. When the temperature and humidity fluctuations in the working environment exceed the preset safety threshold, an abnormal signal can be intelligently generated, and the protection level of protective equipment can be dynamically optimized, thereby ensuring the personal safety of workers in complex, unstable low-pressure working environments. By adopting a protection level assessment method based on a machine learning model (convolutional neural network), the present invention can output the optimal protective equipment level requirements according to the real-time changes of different working environments, avoiding the possible deficiencies of traditional static protection measures, and providing workers with more flexible and accurate protection solutions.

[0049] 2. The present invention fully considers the characteristics of low-voltage arcs (such as short-term high temperature, high radiation, high pressure, etc.), quantifies the specific risks of arc generation through hazard characteristic tests, and provides reliable safety protection for operators in combination with the dynamic changes in the working environment. Under normal working conditions, the present invention directly uses the protection level output by the machine learning model, without the need to over-optimize the level of protective equipment, thus avoiding waste of resources; in abnormal environments, safety is improved by optimizing the protection level. A good balance between safety and cost control is achieved.

[0050] 3. The present invention combines convolutional neural network models and big data analysis technology to greatly improve the scientific nature of protective measures. This intelligent protection mechanism significantly improves the overall safety efficiency of low-voltage non-stop operations. Through the intelligent output of the machine learning model, the present invention converts complex environmental monitoring and risk assessment into clear protection level recommendations, simplifies the decision-making process of protective wear for operators, improves work efficiency, and reduces the uncertainty of human judgment. By conducting in-depth analysis of the arc hazard characteristics and dynamically optimizing protective measures, the present invention can effectively reduce personal injuries and equipment damage accidents caused by low-voltage arcs, and improve the overall safety and reliability of operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly describes the drawings required for the specific embodiments or the description of the prior art. Similar elements or parts are generally identified by similar reference numerals throughout the drawings. Elements or parts in the drawings are not necessarily drawn to scale.

[0052] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0054] It will be understood that when used in this specification and the appended claims, the terms “comprises” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0055] It should also be understood that the terms used in the present specification are only for the purpose of describing particular embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0056] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0057] Example 1

[0058] Arc hazard protection for 0.4kV low-voltage, non-stop power operations involves a series of measures to prevent arcing and the resulting hazards to personnel during maintenance or overhaul of 0.4kV low-voltage power equipment. Arcing is a common phenomenon when power equipment malfunctions and can cause burns, scalds, and even fires to personnel. Therefore, specific protective measures are required for non-stop power operations.

[0059] Specifically, this protection includes using specialized arc protection tools, wearing protective clothing, and installing arc protection devices. These protective measures can effectively isolate the arc generation area, preventing high-temperature arc damage to the human body while ensuring the safety of workers. This is especially true in low-voltage power systems, where arc generation and propagation are slow due to the relatively low voltage. However, the dangers of arcs are still significant, making protective measures essential.

[0060] The core purpose of this approach is to ensure that uninterrupted operations within low-voltage power systems can proceed smoothly without endangering personnel safety through the coordination of rational workflows, technical means, and protective equipment. This not only complies with relevant safety regulations but also provides maximum protection for operators, helping to improve the efficiency and safety of power system maintenance.

[0061] Please participate Figure 1 The present invention discloses a 0.4kV low-voltage non-stop operation arc hazard protection method, comprising the following steps:

[0062] S1. Analyze the arc formation mechanism in different working environments and identify the potential risk factor combinations of arc to operators.

[0063] It's understandable that arc formation mechanisms are closely tied to the working environment. Different electrical equipment, operating methods, and environmental conditions can all affect arc generation and characteristics. Analyzing arc formation mechanisms can identify potential operational risk factors, such as the conditions for arc flash, current intensity, and duration. This helps predict potential hazards and provides a basis for proactive protective design. Specifically, factors such as equipment aging and operating errors can contribute to arc formation. Arcing can cause safety issues such as fires, burns, and visual impairment, making analysis of these factors crucial.

[0064] S2. Based on the identified potential risk factors of the arc to the operator, collect typical arc parameters during the arc formation process.

[0065] It's understood that arc characteristic parameters (such as arc voltage, current, duration, and temperature) are important indicators for assessing the severity of the arc's hazard. By collecting typical arc parameters during the arc formation process, we can more accurately understand the specific threat the arc poses to workers and quantify the degree of danger. Specifically, the collected data can include arc current, voltage, explosive gas release, and thermal radiation. This data provides a quantitative basis for subsequent hazard analysis and the design of protective measures.

[0066] S3. Based on the collected typical arc parameters, multiple preset hazard characteristic tests are carried out, the results of the multiple hazard characteristic tests are input into the pre-trained machine learning model, and the protection level requirements of multiple preset protective equipment are output.

[0067] It is understandable that through experiments, the hazard characteristics of different arc parameters to workers can be verified, covering multiple factors such as thermal radiation, gas jets, electric shock, arc duration, etc. After inputting these test results into the machine learning model, the protection level requirements of protective equipment for different working environments and arc characteristics can be obtained, thereby providing scientific protection recommendations for different work scenarios. Specific content: For example, under different arc intensities, currents and durations, different levels of burns, hearing damage and other hazards may be caused to workers. The machine learning model can automatically infer the appropriate protective equipment requirements based on test data and historical data, such as the heat resistance level of protective clothing, the insulation level of gloves, the impact resistance level of masks, etc.

[0068] S4. Based on the protection level requirements of multiple preset protective equipment output by the machine learning model, formulate the wearing standards of protective equipment for workers performing low-voltage non-stop operations in different working environments.

[0069] It's understandable that, based on the analysis results of the machine learning model, protective specifications can be developed that meet the requirements of different arc risk levels, thereby ensuring the most effective protection for workers during low-voltage, non-stop operations. Arc formation and characteristics may vary in different working environments, so protective measures should also vary. Specifically: For example, some working environments may require workers to wear specific levels of protective clothing, masks, or protective gloves, while other environments may only require lightweight protective equipment. By standardizing the protective equipment worn by workers based on the protection level requirements output by the model, the probability of accidents can be significantly reduced, ensuring worker safety.

[0070] As a further embodiment, the combination of identifying potential risk factors of arc to operators includes:

[0071] Based on the fault tree analysis method, a probability tree diagram of arc occurrence is constructed to identify all combinations of factors that lead to high-probability arc events. High probability means comparing the probability of an arc event with a preset probability threshold. If the probability of an arc event is greater than or equal to the preset probability threshold, the arc event is marked as a high-probability event.

[0072] Steps to construct the arc occurrence probability tree diagram FTA:

[0073] First, the goal of the analysis is to identify the main factors that cause arcing.

[0074] Arc events are usually caused by factors such as overload, short circuit, equipment damage, insulation failure, poor contact, etc. in the power system. Therefore, the target event of the system is "arc occurrence".

[0075] Identify the top event: The top event is the starting point of the fault tree analysis and is usually the final manifestation of a serious fault in the system. In this example, the top event is "arcing".

[0076] Construct a fault tree: In the fault tree, analyze downward from the top event to identify the direct cause, secondary causes, and basic events of the arc.

[0077] Possible secondary events (first-level events):

[0078] Excessive current (overload): The current in the equipment or circuit exceeds the designed value, causing overheating or arcing.

[0079] Equipment failure (switch failure, poor contact): The equipment itself has a fault and cannot work normally, resulting in abnormal current and possible arcing.

[0080] Insulation aging or damage: The insulation performance of power equipment is reduced or damaged, which increases the probability of arcing.

[0081] Short circuit or ground fault: When a short circuit or ground fault occurs in the circuit, the current increases sharply and arc is easily formed.

[0082] External environmental factors (e.g., humidity, temperature): Environmental factors such as humidity and excessive temperature may cause equipment failure or degradation of electrical insulation performance, leading to arcing.

[0083] Further analysis of secondary causes:

[0084] Each level 1 event can be further broken down: Excessive current may be caused by insufficient equipment capacity or abnormal load. Equipment failure can be caused by material defects, improper installation, or wear and tear from long-term use. Insulation aging is related to material quality and the operating environment. Short circuits or ground faults may be caused by damaged wires, switch failure, or improper circuit design. External environmental factors such as high humidity or low temperature may cause equipment to malfunction.

[0085] Selecting the appropriate logic gates: In FTA, logic gates (such as AND and OR gates) are used to connect events. OR gates: If any of the input events occurs, the top event occurs. For example, an arc event can occur if any equipment failure occurs. AND gates: Multiple events must occur simultaneously to cause the top event. For example, an arc can only occur if the load is too large and the insulation is damaged.

[0086] Quantifying the probability of an event: In FTA, the probability of occurrence can be estimated for each basic event (such as equipment failure and overload). For example, a 1% failure rate for a certain device can be used as the probability of occurrence of a basic event. By combining logic gates, the probability of occurrence of the top event can be calculated, helping to assess the risk of arcing.

[0087] In addition, when the fault tree analysis method is used, multiple fault mechanisms are preset, and environmental factors are introduced, including temperature change index and humidity change index, as external factors affecting the occurrence of arcing.

[0088] Ambient temperature and humidity variation indexes can be considered as external conditions that influence equipment or system failures. Their changes may cause fatigue, corrosion, and changes in electrical performance of equipment materials. Incorporating these factors into FTA helps to fully understand the root causes of failures. Specifically:

[0089] Temperature variation index: used to characterize the impact of temperature fluctuations on equipment. For example, a sudden rise or fall in temperature may cause thermal expansion or contraction, affecting the performance of electronic components and even causing electrical failures or arcing.

[0090] Humidity Variation Index: reflects the impact of humidity fluctuations on the internal electrical components of the equipment. Excessive humidity may cause electrical short circuits, corrosion, or poor contact, thereby increasing the risk of equipment failure.

[0091] These environmental factors can be used as input conditions for FTA and combined with other failure mechanisms of the system (such as design defects, operational errors, etc.) to further deduce possible failure modes.

[0092] The logic for obtaining the temperature change index is as follows:

[0093] In a fixed time window, obtain the temperature data collected at multiple time points, and then calculate the average value to get T avg , which is used to represent the overall temperature level within the time window and serves as the benchmark value for the subsequent calculation of the temperature variation amplitude; the temperature variation amplitude ΔT is obtained by calculating the difference between the maximum and minimum temperatures in the temperature data of multiple time points. It is used to measure the severity of temperature fluctuations and directly reflects the temperature variation range.

[0094] The temperature change index calculation formula is:

[0095] δ is a preset proportional coefficient used to adjust the calculation result of the temperature change index, reflecting the proportional adjustment factor under specific conditions. It can be adjusted appropriately according to different usage scenarios or experimental conditions. Tis the temperature change index, f(T) is the relationship function between temperature change and material deformation, which describes the deformation characteristics of the material when the temperature changes, converts the physical effects of temperature change (such as expansion or contraction) into the form of numerical calculation, and provides a basis for the calculation of the temperature change index; the temperature change index quantifies the impact of temperature fluctuations by combining the temperature change amplitude and the ambient temperature change-material deformation relationship function.

[0096] There are standard materials preset in the working environment, and the relationship function formula between temperature change and material deformation is: L0 represents the initial length of the standard material, ΔL represents the length change of the standard material in a fixed time window, and β represents the dynamic influence coefficient, which describes the dynamic response intensity of the material's length change when the temperature changes. It is used to measure the degree of dynamic response of temperature change to the material's length change.

[0097] The dynamic influence coefficient β describes the comprehensive dynamic performance of the material under temperature fluctuations, combining elastic properties, thermal expansion properties, and mechanical properties. The acquisition logic is:

[0098] Satisfies the following formula: E represents the elastic modulus of a standard material, reflecting its ability to resist elastic deformation and a key parameter in describing its mechanical properties. α represents the linear thermal expansion coefficient of a standard material, describing the proportional change in length per unit temperature change and an important parameter in its thermal performance. v represents the Poisson's ratio of a standard material, the ratio of its lateral to longitudinal deformation under stress. 1-v represents a correction factor related to the coupling of the material's lateral and longitudinal deformations. This formula integrates the material's thermodynamic behavior under temperature changes and stress conditions, derived from the thermal expansion effect and elasticity theory. Temperature changes cause thermal expansion of the material, while the material's elastic deformation affects the degree of expansion. The Poisson's ratio corrects for the coupling between lateral and longitudinal deformations.

[0099] In addition, the logic for obtaining the humidity change index is:

[0100] In a fixed time window, the humidity data collected at multiple time points are obtained, and then the absolute value of the difference between the humidity data at two adjacent time points is calculated to obtain the humidity change amplitude value ΔH(t), which represents the humidity change amplitude value at time point t and is used to measure the strength of the temperature fluctuation between two adjacent time points;

[0101] Calculate the average value of all humidity change amplitudes ΔH(t) to obtain ΔH avg , which is used to measure the average level of overall temperature change within the time window. It is an important part of calculating the temperature change index. The standard deviation of all humidity change amplitude values ​​ΔH(t) is calculated to obtain σ ΔH, which is used to measure the discrete degree of temperature change and reflect the intensity of temperature data fluctuation within the time window.

[0102] The autocorrelation function is used to evaluate the autocorrelation of the humidity data time series at multiple preset lag time points. The formula is:

[0103]

[0104] H(t) represents the humidity data at time point t, which is used to describe the data sequence of temperature changes over time and is the basis for the entire temperature change analysis. It represents the average value of humidity data in a fixed time window, reflects the overall level of temperature data in the time window, and is used as a reference point for relative changes and deviations. N represents the total number of data points, ki represents the i-th preset lag time point, and is used to determine the calculation range of the autocorrelation function. ACF(ki) represents the autocorrelation at the i-th preset lag time point, and is used to describe the time series characteristics of temperature data, that is, the degree of correlation between a certain moment in the time series and the lag time point.

[0105] The autocorrelation coefficient A is obtained by taking a weighted average of the autocorrelation values ​​ACF(ki) at multiple preset lag time points. If the autocorrelation ACF(ki) is high, it means that the temperature fluctuation has a certain regularity or periodicity. The dynamic trend of temperature change can be evaluated through the autocorrelation values ​​at multiple lag time points. In industrial or environmental monitoring, the autocorrelation coefficient can be used to determine whether there is an abnormality in the power system. If the autocorrelation coefficient is significantly reduced, it may mean that the regularity of the low-voltage power system is destroyed and the degree of influence by environmental changes may be more sensitive.

[0106] The calculation formula of humidity change index is:

[0107] I H =w1·ΔH avg +w2·σ ΔH +w3·A; w1, w2, and w3 are all preset non-zero adjustment coefficients, and the sum of w1, w2, and w3 is one, which is used to adjust the contribution ratio of each part to the temperature change index. In different scenarios, different emphasis can be placed on the factors affecting temperature change by adjusting the value of the non-zero adjustment coefficient. H Represents the humidity variability index. A high humidity variability index is often associated with a dynamic and unstable humidity environment, indicating that external conditions may change significantly over a short period of time. Unstable humidity may have more negative impacts on equipment or personnel in low-voltage power systems.

[0108] Typical arc parameters refer to a number of common and representative preset physical parameters during the arc formation process. These parameters can reflect the arc's characteristics, energy, impact range, and potential threats to the surrounding environment and equipment. They include but are not limited to the following common typical arc parameters, for example:

[0109] Arc voltage: The potential difference between the two ends of the arc channel, reflecting the electrical characteristics of the arc during operation, usually related to the arc length and gas medium.

[0110] Arc current: The magnitude of the current passing through the arc channel reflects the intensity of the arc and the energy generated, and is directly related to the arc temperature and light radiation intensity.

[0111] Arc temperature: The temperature inside the arc channel. Arc temperature is an important factor affecting the degree of damage caused by the arc to the surrounding environment. It can reach a very high degree Celsius that affects the work of the workers.

[0112] Arc duration: The length of time from the formation to the extinction of the arc. The duration determines the total amount of energy released by the arc and affects the degree of damage to equipment or operators.

[0113] Arc power: The power released by the arc per unit time. The power directly affects the thermal effect and destructive power of the arc on surrounding objects (such as materials, air, etc.).

[0114] The machine learning model is a convolutional neural network model. The convolutional neural network converts complex multi-dimensional input data (such as arc parameters and hazard characteristics) into structured outputs (such as protection level requirements) through layer-by-layer feature extraction. The CNN model mainly includes the following key parts:

[0115] Input layer: Input data: Organize the typical arc parameters and hazard characteristics test results into tensor form and input them into CNN.

[0116] Input data may include: Arc parameters: voltage, current, temperature, duration, energy, etc.

[0117] Hazard test results: such as burn depth, heat distribution, equipment damage, pressure wave impact, etc.

[0118] Environmental conditions: such as humidity, temperature, gas composition, etc.

[0119] The input data can be organized into matrix or image form (for example, mapping the experimental data into a two-dimensional heat map) for CNN processing.

[0120] Convolutional layer: The convolution kernel (filter) scans the input data to extract key features. The convolution kernel can identify patterns, trends, and local features in the arc parameters and hazard test data. Process: The filter is multiplied by the input data point by point and summed, outputting a feature map. The filter can capture information such as the burn risk corresponding to the arc temperature or the pressure wave intensity corresponding to the arc current. The convolution operation reduces the dimensionality of the data while retaining the most important features, reducing computational complexity.

[0121] The pooling layer downsamples the feature maps generated by the convolutional layer, further compressing the data while preserving the key information of the features. Process: Common pooling methods include max pooling and average pooling. For example, max pooling can retain the most significant values ​​in the feature map (such as the maximum arc temperature or current).

[0122] The fully connected layer flattens the feature maps extracted by the pooling layer into a one-dimensional vector and inputs it into a fully connected neural network for classification or regression. Process: Each neuron is connected to all neurons in the previous layer. Combining weight parameters and activation functions, this completes the nonlinear mapping of high-dimensional features. It outputs multiple protection level requirements for pre-set protective equipment, such as: mask rating (e.g., arc radiation resistance); glove rating (e.g., insulation and high-temperature resistance); and protective clothing rating (e.g., resistance to high temperatures and pressure waves).

[0123] The output layer outputs the protection level requirements for multiple protective gear, which are used to develop wear standards. Result: The output protection level is a discrete classification (such as level 1, level 2, level 3). For example: a protective mask needs to be protection level 3, protective gloves need to be protection level 2, and protective clothing needs to be protection level 3.

[0124] When formulating protective equipment wearing specifications for workers performing low-voltage, non-stop operations in different working environments based on the protection level requirements of multiple preset protective equipment output by the machine learning model, obtain the standard deviation of the temperature change index and the standard deviation of the humidity change index of the working environment in multiple fixed time windows. If the standard deviation of the temperature change index and the standard deviation of the humidity change index of the working environment do not exceed the corresponding preset safety fluctuation thresholds, a normal signal is generated. If the standard deviation of the temperature change index and the standard deviation of the humidity change index of the working environment do not exceed the corresponding preset safety fluctuation thresholds, an abnormal signal is generated.

[0125] When a normal signal is generated, the protective gear wearing specifications for workers performing low-voltage non-stop operations are the protection levels of multiple preset protective gear output by the machine learning model;

[0126] When an abnormal signal is generated, the protective equipment wearing specifications for workers performing low-voltage non-stop operations are the protection levels of multiple preset protective equipment output by the optimized machine learning model.

[0127] The optimization principle is: B1 represents the standard deviation of the temperature change index, B2 represents the standard deviation of the humidity change index, Y1 represents the safety fluctuation threshold corresponding to the standard deviation of the temperature change index, Y2 represents the safety fluctuation threshold corresponding to the standard deviation of the humidity change index, r represents the preset tuning coefficient, Dj represents the protection level of the j-th preset protective equipment output by the machine learning model, and DYj represents the protection level of the j-th preset protective equipment after optimization.

[0128] The standard deviation of the temperature variation index quantifies the degree of temperature variation index fluctuation within multiple fixed time windows. A higher standard deviation indicates more severe temperature fluctuations and a more unstable working environment. The standard deviation of the humidity variation index quantifies the degree of humidity variation index fluctuation within multiple fixed time windows. A higher standard deviation indicates more severe humidity fluctuations and a more unstable working environment. When an abnormal signal is generated, it is used to optimize the protection level of protective equipment, increasing the protection level to ensure worker safety. When an abnormal signal is generated, the protection level is optimized and adjusted to cope with abnormal environmental conditions.

[0129] When normal signals are generated, environmental fluctuations are small and the risk is low, so there is no need for additional protection level upgrades. Directly using the protection level output by the model can avoid over-protection, reduce costs and unnecessary operational complexity. When positive abnormal signals are generated, drastic temperature fluctuations may lead to higher arc temperatures or energy release, increasing hazards to humans and equipment. Severe humidity fluctuations may reduce insulation performance and increase the risk of arc propagation. The original protection level is based on the prediction results under normal conditions and cannot cope with extreme situations beyond the preset fluctuation range. Increasing the protection level (such as higher-level masks and protective clothing) can effectively reduce operational risks and protect personnel safety. Optimizing operations not only improves flexibility, but also adapts to the actual needs of different working environments.

[0130] Example 2

[0131] A computer-readable storage medium includes a stored program, wherein when the program is executed, the method described in the invention is controlled by a device where the computer-readable storage medium is located.

[0132] Those skilled in the art will appreciate that the units of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0133] In the embodiments provided by the present invention, it should be understood that the division of units is merely a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0134] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0135] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nly Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc., various media that can store program code.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and description of the present invention.

Claims

1. A 0.4kV low voltage non-stop operation arc hazard protection method, characterized in that: include: Analyze the arc formation mechanism in different working environments and identify the potential risk factors of arc to workers; Based on the identified potential risk factors of arc to workers, collect typical arc parameters during the arc formation process; Based on the collected typical arc parameters, multiple preset hazard characteristic tests are carried out. The results of multiple hazard characteristic tests are input into the pre-trained machine learning model to output multiple preset protection level requirements for protective equipment; Based on the protection level requirements of multiple preset protective equipment output by the machine learning model, the wearing standards of protective equipment for workers performing low-voltage non-stop operations in different working environments are formulated.

2. The arc hazard protection method for 0.4kV low-voltage non-stop operation according to claim 1 is characterized in that: The combination of potential risk factors of arcing to workers includes: Based on the fault tree analysis method, a probability tree diagram of arc occurrence is constructed to identify all combinations of factors that lead to high-probability arc events.

3. The arc hazard protection method for 0.4kV low-voltage non-stop operation according to claim 2 is characterized in that: The fault tree analysis method includes: Multiple failure mechanisms are preset and environmental factors are introduced; The environmental factors include a temperature change index and a humidity change index.

4. The arc hazard protection method for 0.4kV low-voltage non-stop operation according to claim 3 is characterized in that: The method for obtaining the temperature change index includes: In a fixed time window, obtain the temperature data collected at multiple time points, and then calculate the average value to get T avg , calculate the difference between the maximum temperature and the minimum temperature in the temperature data at multiple time points to obtain the temperature change amplitude ΔT; The temperature change index calculation formula is: Where δ is the preset proportional coefficient, I T is the temperature change index, f(T) is the relationship function between temperature change and material deformation; There are standard materials preset in the working environment, and the relationship function formula between temperature change and material deformation is: Where L0 represents the initial length of the standard material, ΔL represents the length change of the standard material in a fixed time window, and β represents the dynamic influence coefficient.

5. The arc hazard protection method for 0.4kV low-voltage non-stop operation according to claim 4 is characterized in that: The dynamic influence coefficient satisfies the following formula: Wherein, E represents the elastic modulus of the standard material, α represents the linear thermal expansion coefficient of the standard material, and v represents the Poisson's ratio of the standard material, which is the ratio of the lateral deformation to the longitudinal deformation of the standard material when subjected to force.

6. A 0.4kV low voltage non-stop operation arc hazard protection method according to claim 5, characterized in that: The method for obtaining the humidity change index includes: In a fixed time window, the humidity data collected at multiple time points are obtained, and then the absolute value of the difference between the humidity data at two adjacent time points is calculated to obtain the humidity change amplitude value ΔH(t), which represents the humidity change amplitude value at time point t; Calculate the average value of all humidity change amplitudes ΔH(t) to obtain ΔH avg , calculate the standard deviation of all humidity change amplitudes ΔH(t) to obtain σ ΔH ; The autocorrelation function is used to evaluate the autocorrelation of the humidity data time series at multiple preset lag time points. The calculation formula is as follows: H(t) represents the humidity data at time point t, represents the average value of humidity data in a fixed time window, N represents the total number of data points, ki represents the i-th preset lag time point, and ACF(ki) represents the autocorrelation at the i-th preset lag time point; Perform weighted averaging on the autocorrelation ACF(ki) at multiple preset different lag time points to obtain the autocorrelation coefficient A; The calculation formula of the humidity change index is as follows: I H =w1·ΔH avg +w2·σ ΔH +w3·A; w1, w2, w3 are all preset non-zero adjustment coefficients, and the sum of w1, w2, w3 is one, I H Indicates the humidity change index.

7. A 0.4kV low voltage non-stop operation arc hazard protection method according to claim 6, characterized in that: The typical arc parameters refer to multiple common and representative preset physical parameters in the arc formation process.

8. The arc hazard protection method for 0.4kV low-voltage non-stop operation according to claim 1 is characterized in that: The protective gear wearing specifications for workers performing low-voltage non-stop operations in different working environments are formulated based on the protection level requirements of multiple preset protective gear output by the machine learning model, including: Obtain the standard deviation of the temperature variation index and the standard deviation of the humidity variation index of the working environment in multiple fixed time windows. If the standard deviation of the temperature variation index and the standard deviation of the humidity variation index of the working environment do not exceed the corresponding preset safety fluctuation threshold, a normal signal is generated. If the standard deviation of the temperature variation index and the standard deviation of the humidity variation index of the working environment do not exceed the corresponding preset safety fluctuation threshold, an abnormal signal is generated. When a normal signal is generated, the protective gear wearing specifications for workers performing low-voltage non-stop operations are the protection levels of multiple preset protective gear output by the machine learning model; When an abnormal signal is generated, the protective equipment wearing specifications for workers performing low-voltage non-stop operations are the protection levels of multiple preset protective equipment output by the optimized machine learning model.

9. The arc hazard protection method for 0.4kV low-voltage non-stop operation according to claim 6, characterized in that: The optimization principle of the protection levels of multiple preset protective equipment output by the optimized machine learning model is as follows: In the formula, B1 represents the standard deviation of the temperature change index, B2 represents the standard deviation of the humidity change index, Y1 represents the safety fluctuation threshold corresponding to the standard deviation of the temperature change index, Y2 represents the safety fluctuation threshold corresponding to the standard deviation of the humidity change index, r represents the preset tuning coefficient, Dj represents the protection level of the j-th preset protective equipment output by the machine learning model, and DYj represents the protection level of the j-th preset protective equipment after optimization.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 9.

Images