Human body electric shock protection method and system based on multi-parameter fusion

Abstract

The invention belongs to the technical field of electrical safety, and particularly relates to a human body electric shock protection method and system based on multi-parameter fusion, multi-dimensional characteristic parameters are formed by extracting a high-frequency harmonic component, a voltage abrupt change slope and a magnetic field intensity fluctuation value of a residual current waveform, and the weight is dynamically adjusted according to the instantaneous change rate of each characteristic, so that the human body electric shock protection is realized. Generating a weighted feature vector; and a comprehensive criterion value is calculated in combination with a preset rule base and is compared with a dynamic threshold value, so that whether the circuit breaker is triggered to act or not is determined. According to the method, through fusion analysis of a plurality of physical characteristics, the identification capability of the system in a complex electromagnetic environment is improved, the probability of misoperation caused by external interference is effectively reduced, the action accuracy and stability of the circuit breaker are improved, and the power utilization safety and the power supply continuity are guaranteed.

Description

Technical Field

[0001] The present invention belongs to the field of electrical safety technology, and specifically relates to a method and system for protecting human body from electric shock based on multi-parameter fusion. Background Art

[0002] In the field of electrical safety protection, traditional residual current circuit breakers rely primarily on a single parameter, residual current, to determine whether a person is experiencing electric shock or a line leakage. When the residual current exceeds a set threshold, the circuit breaker triggers, shutting off the power supply and preventing electric shock accidents.

[0003] However, in practical applications, especially in environments with strong electromagnetic interference (such as industrial sites where variable-frequency equipment and high-power motors frequently start and stop), traditional leakage circuit breakers are easily affected by external magnetic fields and high-frequency noise, leading to misjudgment of leakage current and unnecessary tripping. This misoperation not only affects power supply continuity but can also cause production interruptions and safety hazards.

[0004] Although there are attempts in the existing technology to reduce the false operation rate by adding filtering circuits or adopting fixed threshold compensation mechanisms, since they only rely on a single characteristic parameter and lack the ability to adapt to complex interference environments, it is difficult to effectively improve the judgment accuracy. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for human electric shock protection based on multi-parameter fusion. By integrating and analyzing multiple physical characteristics, the system's recognition ability in complex electromagnetic environments is improved, the probability of false operation due to external interference is effectively reduced, the operation accuracy and stability of the circuit breaker are improved, and electricity safety and power supply continuity are guaranteed, so as to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention adopts the following technical solution: a method for protecting human body from electric shock based on multi-parameter fusion, comprising the following steps: The high-frequency harmonic components, voltage mutation slope, and magnetic field intensity fluctuation values ​​of the residual current waveform are extracted to form multidimensional characteristic parameters. The weight coefficients of each parameter are dynamically adjusted based on the instantaneous change rate of the multidimensional characteristic parameters to generate a weighted characteristic vector. The weighted characteristic vector is input into a preset fusion rule library to calculate a comprehensive criterion value. The comprehensive criterion value is compared with the dynamic threshold range. If the threshold is exceeded, the circuit breaker is triggered; otherwise, the power state is maintained.

[0007] Preferably, the high-frequency harmonic components, voltage mutation slope and magnetic field intensity fluctuation value of the residual current waveform are extracted to form multi-dimensional characteristic parameters, including: Perform band-pass filtering on the collected current signal to separate the high-frequency part with a frequency higher than 300 Hz, which is recorded as the first intermediate signal; Performing a time-domain difference operation on the first intermediate signal to obtain a sum of absolute values ​​of differences between adjacent periods, which is marked as a waveform mutation indicator; The spatial magnetic field variation is obtained through the Hall sensor, and the ratio of the difference between adjacent moments to the waveform mutation index is calculated as a reference index for the magnetic field interference intensity.

[0008] Preferably, the collected current signal is band-pass filtered to separate the high-frequency portion with a frequency higher than 300 Hz, including: The original current signal is converted into a preliminary high-frequency signal by a passive second-order high-pass filter circuit composed of an inductor and a capacitor, and the amplitude of the preliminary high-frequency signal is normalized to obtain a normalized high-frequency component; A fixed threshold is set, and the portion of the normalized high-frequency component whose amplitude exceeds the threshold is retained, and the rest is set to zero, thereby obtaining the high-frequency harmonic component.

[0009] Preferably, performing a time domain difference operation on the first intermediate signal to obtain the sum of absolute values ​​of differences between adjacent periods includes: Dividing the high-frequency harmonic component according to the power frequency period, and extracting signal segments of two consecutive period segments; Subtracting the values ​​of the signal segments at corresponding time points and taking the absolute values ​​to obtain a difference sequence; The difference sequence is accumulated and summed within a period, and recorded as a waveform mutation index, which is used to indicate the degree of waveform mutation.

[0010] Preferably, obtaining the spatial magnetic field variation through the Hall sensor includes: A three-axis Hall sensor is installed inside the circuit breaker to collect the magnetic field intensity components in the X, Y, and Z directions respectively, and the magnetic field intensity components are synthesized to obtain the total magnetic field intensity; The total magnetic field strength is collected at fixed time intervals, and the difference between adjacent moments is calculated as the magnetic field change at the current moment.

[0011] Preferably, according to the instantaneous change rate of the multidimensional feature parameters, the weight coefficient of each parameter is dynamically adjusted to generate a weighted feature vector, including: Calculate the difference between the high-frequency harmonic component, voltage mutation slope and magnetic field interference intensity reference index at adjacent moments; Performing a ratio operation on the change difference and a set reference value to obtain a normalized change rate; The normalized change rate is used as a weight factor, multiplied by the original value of the corresponding parameter in sequence, and combined into a weighted feature vector.

[0012] Preferably, the normalized change rate is used as a weight factor, which is sequentially multiplied by the original value of the corresponding parameter to form a weighted feature vector, including: Obtain the real-time measurement values ​​of the current high-frequency harmonic components, voltage mutation slope, and magnetic field interference intensity reference indicators; Multiplying the weight factor by the original value of the corresponding parameter to obtain a weighted term; The weighted items are respectively subjected to clipping processing, and a maximum value is set. If an item exceeds the maximum value, it is set to the maximum value; The clipped weighted items are combined into triplets and output as the weighted feature vector at the current moment.

[0013] Preferably, the weighted feature vector is input into a preset fusion rule library to calculate the comprehensive criterion value, including: Set three sets of linear decision boundaries, corresponding to the three components in the weighted feature vector; Compare each component of the weighted feature vector with the corresponding boundary respectively, and if a component is greater than the corresponding boundary, the output flag is 1, otherwise it is 0; A logical code is generated according to the combination of flag bits, and a table lookup is performed based on the logical code to obtain the corresponding comprehensive criterion value for subsequent action judgment.

[0014] Preferably, comparing the comprehensive criterion value with the dynamic threshold range and triggering a circuit breaker action if the value exceeds the threshold value includes: Set a basic threshold and adjust the offset according to the current environmental interference level to generate a dynamic upper threshold limit; Comparing the comprehensive criterion value with the upper limit of the dynamic threshold, and determining an abnormal state if the value exceeds the upper limit; When it is determined to be in an abnormal state, the actuator is started with a delay to drive the circuit breaker to trip and cut off the line; Otherwise, maintain the original power-on state and continue to monitor the next cycle data.

[0015] On the other hand, the present invention proposes a human electric shock protection system based on multi-parameter fusion, comprising: Multi-dimensional feature parameter extraction module, used to extract the high-frequency harmonic components, voltage mutation slope and magnetic field intensity fluctuation value of the residual current waveform to form multi-dimensional feature parameters; A dynamic weight adjustment module, configured to dynamically adjust the weight coefficient of each parameter according to the instantaneous change rate of the multidimensional feature parameters to generate a weighted feature vector; A comprehensive criterion calculation module, used for inputting the weighted feature vector into a preset fusion rule library to calculate a comprehensive criterion value; The circuit breaker action control module is used to compare the comprehensive criterion value with the dynamic threshold range, and trigger the circuit breaker action if it exceeds the threshold, otherwise maintain the power-on state.

[0016] Technical effects and advantages of the present invention: Compared with the existing technology, the method and system for protecting human body from electric shock based on multi-parameter fusion proposed by the present invention have the following advantages: This method extracts the high-frequency harmonic components, voltage mutation slope, and magnetic field intensity fluctuations of the residual current waveform to form multidimensional characteristic parameters. The weights of each characteristic are dynamically adjusted based on its instantaneous rate of change to generate a weighted characteristic vector. This is then combined with a preset rule base to calculate a comprehensive criterion value, which is compared with a dynamic threshold to determine whether to trigger the circuit breaker. By integrating and analyzing multiple physical features, this method enhances the system's recognition capabilities in complex electromagnetic environments, effectively reduces the probability of false operation due to external interference, improves the accuracy and stability of the circuit breaker, and ensures power safety and continuity. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of the human body electric shock protection method based on multi-parameter fusion of the present invention; Figure 2 This is a block diagram of the human electric shock protection system based on multi-parameter fusion of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0019] The present invention provides a method for protecting human body from electric shock based on multi-parameter fusion. The method is applicable to residual current protection devices such as residual current operated circuit breakers and small leakage circuit breakers, which can include intelligent switches, intelligent circuit breakers and other products. In this embodiment, a circuit breaker is used as an example for description: The present invention provides Figure 1 A method for protecting human body from electric shock based on multi-parameter fusion is shown, comprising the following steps: Step 1: Extract the high-frequency harmonic components, voltage mutation slope, and magnetic field intensity fluctuation value of the residual current waveform to form multi-dimensional characteristic parameters; specifically, the following steps are involved: The collected current signal A is band-pass filtered to separate the high-frequency part with a frequency higher than 300Hz, which is recorded as C. Specifically, the original current signal A is passed through a passive second-order high-pass filter circuit composed of an inductor L and a capacitor C to output a preliminary high-frequency signal. ;right Perform amplitude normalization and obtain ; Set a fixed threshold Th, retain the part of C whose amplitude exceeds Th, and set the rest to zero to obtain the high-frequency harmonic component .

[0020] Indicates the initial high-frequency signal The purpose of amplitude normalization is to eliminate the influence caused by the difference in the original amplitude of the current signal, so that the feature extraction under different working conditions is comparable. Indicates taking The maximum absolute value within a cycle is used as the reference value for normalization to ensure that the peak value of the normalized signal is 1; In environments with strong electromagnetic interference, the amplitude of the current signal may fluctuate significantly due to external factors. Using the original amplitude directly for subsequent analysis can lead to misjudgment. Therefore, by dividing the signal by its maximum absolute value, its dynamic range is compressed to the interval [-1, 1]. This achieves amplitude normalization, preserves the signal waveform trend, and removes the influence of dimension and amplitude.

[0021] Perform time domain difference operation on C to obtain the sum of the absolute values ​​of the differences between adjacent cycles, which is marked as , used to reflect the severity of the mutation; specifically including: Divide by power frequency period T and extract signal segments of two consecutive period segments and ;right and Subtract the values ​​at the corresponding time points and take the absolute value to get the difference sequence ; This formula is used to extract the difference information between two adjacent periodic waveforms and obtain the absolute value of the difference at each moment by point-by-point comparison.

[0022] in, : Normalized high-frequency signal within the first cycle; : The normalized high-frequency signal in the second period immediately following it; : The absolute value of the waveform difference between two cycles at the same time point, used to measure local waveform mutation; When a person is electrocuted or a circuit leaks, the current waveform often exhibits sudden changes or asymmetry. By comparing the high-frequency harmonic signals of two consecutive power frequency cycles (e.g., 20ms at 50Hz), the waveform distortion characteristics caused by the fault can be captured. Absolute values ​​are used to eliminate directional effects and focus only on the magnitude of the change.

[0023] The sum of D in period T is recorded as , which is used to indicate the degree of waveform mutation. This formula is to accumulate the difference sequence D in a complete cycle to obtain a comprehensive index reflecting the degree of overall waveform mutation. Its purpose is to summarize the point-by-point difference information into a quantifiable evaluation value, which is convenient for subsequent comparison with the set threshold to determine whether an abnormality has occurred. a Indicates that the current waveform is significantly different from the previous cycle, which may be due to a sudden change caused by an electric shock or electromagnetic interference.

[0024] Acquiring spatial magnetic field changes through Hall sensors , calculate the difference between adjacent moments and The ratio of , as a reference indicator of magnetic field interference intensity; specifically, installing a three-axis Hall sensor element inside the circuit breaker to collect the magnetic field intensity components in the X, Y, and Z directions respectively ;right Perform synthesis operation to obtain the total magnetic field strength This formula is used to combine the magnetic field component signals collected by the three-axis Hall effect sensor into the total spatial magnetic field intensity. Because the magnetic field is a vector and directional, the magnetic field value in a single direction cannot fully reflect the true spatial magnetic field strength. Using this formula to calculate the vector modulus of the magnetic field components in three orthogonal directions (X, Y, and Z) yields the total magnetic field intensity in the space where the circuit breaker is located at the current moment.

[0025] Collect B every fixed time interval Δt and calculate the difference between adjacent moments , as the magnetic field change at the current moment. This formula is used to extract the change in magnetic field strength between adjacent moments, that is, the rate of change of the magnetic field over time. During the operation of electrical equipment, electromagnetic interference or sudden load changes can cause fluctuations in the spatial magnetic field. By regularly sampling and calculating the difference in magnetic field strength between adjacent moments, these transient magnetic field disturbances can be effectively captured, providing a basis for subsequent determination of whether malfunctions are caused by external interference.

[0026] formula The vector synthesis from the three-axis magnetic field components to the total magnetic field intensity is realized. The two together form the basis for sensing the external electromagnetic environment and provide key input for subsequent multi-parameter fusion analysis, thereby improving the stability and accuracy of the human electric shock protection system.

[0027] Step 2: Dynamically adjust the weight coefficient of each parameter according to the instantaneous change rate of the multi-dimensional feature parameter to generate a weighted feature vector; specifically including: Calculate high-frequency harmonic components separately , voltage mutation slope , magnetic field fluctuations The difference in changes at adjacent moments is recorded as ; Will Set the benchmark value Perform ratio operation to obtain the normalized rate of change By normalizing the instantaneous changes in multidimensional features, they are converted into normalized rates of change with a uniform scale. These rates of change not only reflect the importance of each feature in its current state but also provide a basis for subsequent weighted fusion and comprehensive criterion calculations, thereby improving the intelligent recognition capabilities and operational reliability of the human electric shock protection system.

[0028] Will As a weight factor, it is multiplied by the original value of the corresponding parameter in turn to form a weighted feature vector ; This formula is used to multiply the multidimensional physical characteristics (high-frequency harmonic components, voltage mutation slope, magnetic field interference index) with their corresponding normalized change rates to form a weighted feature vector .

[0029] use (i.e. the ratio of the instantaneous change rate of each feature to the reference value) as the weight factor; The system then performs weighted processing to construct a weighted feature vector that reflects the importance of each feature in the current operating state. This allows the system to dynamically highlight features that are experiencing significant changes at the current moment, allowing for more accurate identification of electric shocks or other abnormal conditions.

[0030] Specifically including: obtaining the high-frequency harmonic components at the current moment , voltage mutation slope , magnetic field interference index Real-time measurement value of and Multiply them together to get the weighted term , similarly ;right Perform clipping processing separately, set the maximum value E, and if an item exceeds E, set it to E; combine the weighted items after clipping into triplets , which is output as the weighted feature vector at the current moment. This expression summarizes the results of the clipping of the weighted terms obtained in the previous step, representing the final output weighted feature vector. This clipping mechanism enhances the robustness of the system and prevents individual features from excessively influencing the overall judgment.

[0031] Step 3: Input the weighted feature vector into a preset fusion rule library to calculate the comprehensive criterion value; specifically including: Set three sets of linear decision boundaries, namely thresholds , used to correspond to the weighted eigenvector The three components in Each component is compared with the corresponding boundary, and if a component is greater than the corresponding threshold, the flag bit is output. ,otherwise ;according to Combinatorial Generative Logic Coding , as the identification mode; according to Q, the corresponding comprehensive judgment value J is obtained for subsequent action judgment. (each value is 0 or 1) are combined into a unique logical code Q, which serves as a discriminant mode identifier.

[0032] The first component (such as high-frequency harmonics ) is greater than the decision boundary , if yes, it is 1, otherwise it is 0; The second component (such as the voltage mutation term ) is greater than the decision boundary , if yes, it is 1, otherwise it is 0; The third component (such as magnetic field interference ) is greater than the decision boundary , if yes, it is 1, otherwise it is 0; This formula maps three binary digits to decimal numbers, with each digit representing whether a component in the weighted feature vector exceeds a set threshold. This encoding method compresses the multidimensional feature state into an integer value, making it easier to quickly look up the corresponding comprehensive criterion value J.

[0033] This approach has the following advantages: Clear structure: three sets of comparison results correspond to three binary codes; Unique encoding: can be represented Different state combinations; Facilitates rule matching: Each code corresponds to a fault / interference mode, and the corresponding criterion value or action strategy can be pre-set; Improve judgment efficiency: avoid complex calculations and directly obtain judgment basis through coding table lookup.

[0034] Step 4: Compare the comprehensive criterion value with the dynamic threshold range. If the value exceeds the threshold, the circuit breaker is triggered; otherwise, the power state is maintained. Specifically, the following steps are performed: Setting a base threshold , and adjust the offset G according to the current environmental interference level to generate a dynamic threshold upper limit ; Combine the comprehensive criterion value J with For comparison, if J> It is determined to be an abnormal state; this formula is used to generate a dynamic upper threshold , as the basis for judging whether to trigger the circuit breaker action.

[0035] In electrical protection systems, if fixed thresholds are used for discrimination, false action may occur in complex electromagnetic interference environments. To solve this problem, a basic threshold is introduced. And offset G, dynamically adjust the upper threshold value according to the current environmental interference level, making the system more adaptable and stable.

[0036] When the abnormal state is determined, the actuator is started with a delay, the circuit breaker is driven to trip, and the line is cut off; specifically, when the abnormal state is determined, the timing counter is started and the delay time H is set; during the H period, the comprehensive judgment value J of the subsequent cycle is continuously monitored, and if J> , confirming that the action conditions are met; a control signal is output to the electromagnetic drive device, causing the moving iron core to move, driving the mechanical contacts of the circuit breaker to separate; after the trip is completed, the actuator is locked and can only be put into operation again after manual reset. Otherwise, the original power-on state is maintained and the next cycle data is monitored.

[0037] On the other hand, the present invention proposes a human electric shock protection system based on multi-parameter fusion, such as Figure 2 As shown, including: Multi-dimensional feature parameter extraction module, used to extract the high-frequency harmonic components, voltage mutation slope and magnetic field intensity fluctuation value of the residual current waveform to form multi-dimensional feature parameters; A dynamic weight adjustment module, configured to dynamically adjust the weight coefficient of each parameter according to the instantaneous change rate of the multidimensional feature parameters to generate a weighted feature vector; A comprehensive criterion calculation module, used for inputting the weighted feature vector into a preset fusion rule library to calculate a comprehensive criterion value; The circuit breaker action control module is used to compare the comprehensive criterion value with the dynamic threshold range, and trigger the circuit breaker action if it exceeds the threshold, otherwise maintain the power-on state.

[0038] In addition, the above modules are also used to implement other steps of the above-mentioned human electric shock protection method based on multi-parameter fusion when executed, as shown below: Imagine a distribution box equipped with a circuit breaker equipped with intelligent recognition capabilities to detect electric shock in real time. The system continuously collects current signals, voltage changes, and spatial magnetic field information, and uses multi-parameter fusion analysis to determine whether to trigger a trip.

[0039] Step 1: Extract multidimensional feature parameters Collect the current signal A. Assume that the collected original current signal is: A=[0.5, 0.6, 0.8, 1.2, 0.9, 0.4, 0.3, 0.2]; Bandpass filtering is performed to obtain the high-frequency component C_p. A passive second-order high-pass filter circuit composed of an inductor L and a capacitor C is used to retain components with frequencies higher than 300 Hz.

[0040] Get the preliminary high-frequency signal: =[0.2,0.3,0.5,0.7,0.6,0.2,0.1,0.05]; Normalization processing: , calculate the maximum absolute value: =0.7, normalized result: C=[0.286,0.429,0.714,1.0,0.857,0.286,0.143,0.071].

[0041] Set the threshold Th=0.5 to extract high-frequency harmonic components , set the points below Th to zero: =[0,0,0.714,1.0,0.857,0,0,0].

[0042] Time domain difference operation to obtain waveform mutation index ,Will Divide by power frequency period T and take two consecutive period segments: =[0,0,0.714,1.0], =[0.857,0,0,0]; Difference Series :D=[0.857,0,0.714,1.0];accumulate and sum to get :D_a=0.857+0+0.714+1.0=2.571.

[0043] Acquiring magnetic field strength fluctuations , the three-axis Hall sensor measures: ; Synthetic total magnetic field intensity B: =0.3.

[0044] The magnetic field at the previous moment is =0.25, then the magnetic field change : =0.3-0.25=0.05, and the reference index of magnetic field interference intensity is obtained : =0.05 / 2.571=0.0194.

[0045] Step 2: Generate weighted feature vector P_f Get the current feature value: ; Calculate the instantaneous rate of change: ; Characteristics of the previous cycle: ; Change difference: ; Normalized rate of change: Set the baseline value: ; Normalization: =0.143 / 0.2=0.715; =0.05 / 0.1=0.5; =0.0044 / 0.01=0.44; Weighted combination, generating weighted feature vector P_f, weighted terms: =0.715*0.857=0.613; =0.5*0.3=0.15; =0.44*0.0194=0.0085; Limiting processing (E=1): =0.613(≤1→unchanged); =0.15; =0.0085; Output weighted feature vector: =(0.613,0.15,0.0085).

[0046] Step 3: Calculate the comprehensive criterion value J Set the decision boundary: ; Compare whether each component is out of bounds: ; ; ; Generate logical code Q: ; Look up the table to obtain the comprehensive criterion value J: Look up the preset rule table, Q=6 corresponds to: J=0.82.

[0047] Step 4: Dynamic Threshold Comparison and Action Control Set the base threshold and offset: =0.75, G=0.05 (according to the current environmental interference level); Calculating the dynamic upper threshold : =0.75+0.05=0.80; Compare J with :J=0.82> =0.80→determined as abnormal state; Start the delayed confirmation mechanism: set the delay time H = 3 cycles, N = 2 consecutive times exceeding the limit; If the current cycle meets the conditions, continue monitoring the next cycle.

[0048] If J> appears again in subsequent cycles When it reaches N times, the tripping action is executed: Output control signals to the electromagnetic drive device to push the core to move and disconnect the mechanical contacts; After the trip is completed, the actuator is locked and a manual reset is required to restore power.

[0049] This example fully simulates the electric shock protection process in a real-world application scenario, including signal acquisition and filtering, feature extraction and normalization, dynamic weight adjustment and weighted fusion, logic encoding and rule matching, dynamic threshold comparison, and delayed confirmation. This entire process demonstrates the advantages of multi-parameter fusion judgment, improving the system's stability and accuracy in complex electromagnetic interference environments, effectively preventing false trips and ensuring electrical safety.

[0050] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for protecting human body from electric shock based on multi-parameter fusion, characterized in that: The following steps are involved: Extract high-frequency harmonic components, voltage mutation slope, and magnetic field intensity fluctuation values ​​of the residual current waveform to form multi-dimensional characteristic parameters; According to the instantaneous change rate of the multi-dimensional characteristic parameters, the weight coefficient of each parameter is dynamically adjusted to generate a weighted characteristic vector; Inputting the weighted feature vector into a preset fusion rule library to calculate a comprehensive criterion value; The comprehensive criterion value is compared with the dynamic threshold range. If the value exceeds the threshold, the circuit breaker is triggered to operate; otherwise, the power-on state is maintained.

2. The method for protecting human body from electric shock based on multi-parameter fusion according to claim 1, characterized in that: The high-frequency harmonic components, voltage mutation slope, and magnetic field intensity fluctuation values ​​of the residual current waveform are extracted to form multi-dimensional characteristic parameters, including: Perform band-pass filtering on the collected current signal to separate the high-frequency part with a frequency higher than 300 Hz, which is recorded as the first intermediate signal; Performing a time-domain difference operation on the first intermediate signal to obtain a sum of absolute values ​​of differences between adjacent periods, which is marked as a waveform mutation indicator; The spatial magnetic field variation is obtained through the Hall sensor, and the ratio of the difference between adjacent moments to the waveform mutation index is calculated as a reference index for the magnetic field interference intensity.

3. The method for protecting human body from electric shock based on multi-parameter fusion according to claim 2, characterized in that: The collected current signal is band-pass filtered to separate the high-frequency part with a frequency higher than 300 Hz, including: The original current signal is converted into a preliminary high-frequency signal by a passive second-order high-pass filter circuit composed of an inductor and a capacitor, and the amplitude of the preliminary high-frequency signal is normalized to obtain a normalized high-frequency component; A fixed threshold is set, and the portion of the normalized high-frequency component whose amplitude exceeds the threshold is retained, and the rest is set to zero, thereby obtaining the high-frequency harmonic component.

4. The method for protecting human body from electric shock based on multi-parameter fusion according to claim 2, characterized in that: Performing a time domain difference operation on the first intermediate signal to obtain a sum of absolute values ​​of differences between adjacent periods includes: Dividing the high-frequency harmonic component according to the power frequency period, and extracting signal segments of two consecutive period segments; Subtracting the values ​​of the signal segments at corresponding time points and taking the absolute values ​​to obtain a difference sequence; The difference sequence is accumulated and summed within a period, and recorded as a waveform mutation index, which is used to indicate the degree of waveform mutation.

5. The method for protecting human body from electric shock based on multi-parameter fusion according to claim 2, characterized in that: The Hall sensor is used to obtain the spatial magnetic field variation, including: A three-axis Hall sensor is installed inside the circuit breaker to collect the magnetic field intensity components in the X, Y, and Z directions respectively, and the magnetic field intensity components are synthesized to obtain the total magnetic field intensity; The total magnetic field strength is collected at fixed time intervals, and the difference between adjacent moments is calculated as the magnetic field change at the current moment.

6. The method for protecting human body from electric shock based on multi-parameter fusion according to claim 1, characterized in that: According to the instantaneous change rate of the multi-dimensional feature parameters, the weight coefficient of each parameter is dynamically adjusted to generate a weighted feature vector, including: Calculate the difference between the high-frequency harmonic component, voltage mutation slope and magnetic field interference intensity reference index at adjacent moments; Performing a ratio operation on the change difference and a set reference value to obtain a normalized change rate; The normalized change rate is used as a weight factor, multiplied by the original value of the corresponding parameter in sequence, and combined into a weighted feature vector.

7. The method for protecting human body from electric shock based on multi-parameter fusion according to claim 6, characterized in that: The normalized change rate is used as a weight factor and multiplied by the original value of the corresponding parameter in sequence to form a weighted feature vector, including: Obtain the real-time measurement values ​​of the current high-frequency harmonic components, voltage mutation slope, and magnetic field interference intensity reference indicators; Multiplying the weight factor by the original value of the corresponding parameter to obtain a weighted term; The weighted items are respectively subjected to clipping processing, and a maximum value is set. If an item exceeds the maximum value, it is set to the maximum value; The clipped weighted items are combined into triplets and output as the weighted feature vector at the current moment.

8. The method for protecting human body from electric shock based on multi-parameter fusion according to claim 1, characterized in that: The weighted feature vector is input into a preset fusion rule library to calculate the comprehensive criterion value, including: Set three sets of linear decision boundaries, corresponding to the three components in the weighted feature vector; Compare each component of the weighted feature vector with the corresponding boundary respectively, and if a component is greater than the corresponding boundary, the output flag is 1, otherwise it is 0; A logical code is generated according to the combination of flag bits, and a table lookup is performed based on the logical code to obtain the corresponding comprehensive criterion value for subsequent action judgment.

9. The method for protecting human body from electric shock based on multi-parameter fusion according to claim 1, characterized in that: Comparing the comprehensive criterion value with the dynamic threshold range, and triggering a circuit breaker action if the threshold is exceeded, including: Set a basic threshold and adjust the offset according to the current environmental interference level to generate a dynamic upper threshold limit; Comparing the comprehensive criterion value with the upper limit of the dynamic threshold, and determining an abnormal state if the value exceeds the upper limit; When it is determined to be in an abnormal state, the actuator is started with a delay to drive the circuit breaker to trip and cut off the line; Otherwise, maintain the original power-on state and continue to monitor the next cycle data.

10. A human electric shock protection system based on multi-parameter fusion for implementing the method according to any one of claims 1 to 9, characterized in that: include: Multi-dimensional feature parameter extraction module, used to extract the high-frequency harmonic components, voltage mutation slope and magnetic field intensity fluctuation value of the residual current waveform to form multi-dimensional feature parameters; A dynamic weight adjustment module, configured to dynamically adjust the weight coefficient of each parameter according to the instantaneous change rate of the multidimensional feature parameters to generate a weighted feature vector; A comprehensive criterion calculation module, used for inputting the weighted feature vector into a preset fusion rule library to calculate a comprehensive criterion value; The circuit breaker action control module is used to compare the comprehensive criterion value with the dynamic threshold range, and trigger the circuit breaker action if it exceeds the threshold, otherwise maintain the power-on state.

Images