CN115684828A - 5G leakage current fault monitoring method based on improved VMD and LMS - Google Patents
5G leakage current fault monitoring method based on improved VMD and LMS Download PDFInfo
- Publication number
- CN115684828A CN115684828A CN202211222093.3A CN202211222093A CN115684828A CN 115684828 A CN115684828 A CN 115684828A CN 202211222093 A CN202211222093 A CN 202211222093A CN 115684828 A CN115684828 A CN 115684828A
- Authority
- CN
- China
- Prior art keywords
- leakage current
- signal
- vmd
- imf
- lms
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Testing Of Short-Circuits, Discontinuities, Leakage, Or Incorrect Line Connections (AREA)
Abstract
The invention discloses a 5G leakage current fault monitoring method based on improved VMD and LMS, which comprises the following steps: acquiring load leakage current information and temperature information; optimizing through a leakage current signal by utilizing a firefly algorithm to obtain the optimal decomposition number M and a penalty factor beta of a signal VMD; performing VMD decomposition on a leakage current original signal by using M and beta values, and dividing the in-time-domain mode into an effective IMF, a mixed IMF and a noise IMF according to a ratio E of a cross-correlation coefficient R and a time-domain energy entropy; denoising the mixed IMF by using an LMS algorithm, and superposing a denoised mode and the effective IMF to obtain a recombined signal; and monitoring the recombination signals and the temperature signals, diagnosing faults, respectively alarming and acting by the protector when the faults occur, and transmitting information to the master station for analysis and processing through the 5G communication module in real time. The invention can improve the denoising effect of the leakage current signal and has high optimizing speed.
Description
Technical Field
The invention relates to the technical field of leakage current fault monitoring, in particular to a 5G leakage current fault monitoring method based on improved VMD and LMS.
Background
Fire conditions of national living places of nearly 10 years are released according to the fire rescue bureau of the emergency management department. Statistically, 132.4 million domestic fires occurred in the country in 2012 to 2021, causing 11634 people to be in distress, 6738 people to be injured and 77.7 million yuan direct property loss, wherein the electrical fire accounted for 42.7% (data source: department of emergency management fire rescue, 2022-02-18, https:// www.119.Gov.cn/article/46rcva01 Vzg). In the causes of electrical fire, the proportion of short circuit of the electric wire or grounding arc short circuit is the largest, the short circuit has the obvious characteristics of abruptness and concealment, and the increase of leakage current in the short circuit is a typical fault criterion. When the electrical equipment breaks down or the circuit is aged, the leakage current is increased slightly, and the leakage current is not changed much more than the leakage current under the normal condition and is not easy to detect. As the leakage current increases, the accumulation of its energy accelerates the aging of the line, which at some point can cause breakdown of the cable insulation (electrical breakdown) and create an arc short to ground at the breakdown. It is this high concealment of leakage current that leads to frequent electrical fire accidents. For electrical fire caused by short circuit of a line, an overcurrent protection device is arranged in the circuit, when the current of the line is overlarge, the circuit is actively cut off, although the occurrence of the fire can be effectively avoided, the probability of false alarm and missing is higher, and adverse effects are brought to smart power grids and economic development.
The leakage current signal is a weak nonlinear signal and has serious noise interference, so that the quick and accurate leakage current detection needs to be realized while the validity of the detection signal is ensured. The accurate extraction of effective components from the leakage current signal is the basis for ensuring the rapid and correct action of the leakage current protector. At present, various methods are proposed to improve the precision and speed of leakage current detection, and compared with the traditional leakage current protector which only acts according to peak current, false operation and refusal action are easy to cause, for example, wavelet analysis, neural network sum and other modern signal processing methods overcome the defects of the traditional leakage current protector, the reliability of the leakage current protector can be further improved, but the denoising effect and the identification precision cannot be ensured due to the restriction of environmental factors and real-time requirements. The Variational Modal Decomposition (VMD) is a recently proposed adaptive signal processing method, and is widely used for signal denoising and feature extraction, and how to apply the variational modal decomposition to a leakage current protector to improve the detection capability of the leakage current protector becomes a problem to be solved urgently.
Disclosure of Invention
It is an object of the present invention to address at least the above-mentioned deficiencies and to provide at least the advantages which will be described hereinafter.
The invention also aims to provide a 5G leakage current fault monitoring method based on the improved VMD and the LMS, which solves the defects of poor current leakage current denoising effect and low recognition precision in summer high-temperature environment.
To achieve these objects and other advantages and in accordance with the purpose of the invention, as embodied and broadly described herein, the present invention provides a 5G leakage current fault monitoring method based on modified VMDs and LMS, comprising:
s1: acquiring load leakage current information and temperature information, and taking the load leakage current information and the temperature information as original signals;
s2: optimizing the leakage current original signal by using a firefly algorithm to obtain the optimal decomposition number M and a penalty factor beta of the signal VMD;
s3: performing VMD decomposition on a leakage current original signal by using the optimal M and beta values obtained by optimizing, taking the cross-correlation coefficient R and the time domain energy entropy ratio E of a recombined signal obtained by decomposition and the original signal as classification indexes, dividing the internal model of the time domain into three categories of effective IMF, mixed IMF and noise IMF, reserving the effective IMF and removing the noise IMF;
s4: denoising the mixed IMF by using an LMS algorithm, and superposing a denoised mode and an effective IMF to obtain a recombined leakage current signal;
s5: and monitoring the recombined leakage current signal and the temperature signal, diagnosing faults, respectively alarming and acting by the protector when the faults occur, and transmitting information to the main station for analysis and processing through the 5G communication module in real time during the process, so that the phenomenon of false alarm and missing report is avoided.
Preferably, in the 5G leakage current fault monitoring method based on the improved VMD and the LMS, the leakage current information is a vector sum of phase-to-ground leakage currents of the phases.
Preferably, in the 5G leakage current fault monitoring method based on the improved VMD and the LMS, specifically, the firefly algorithm uses the envelope entropy S g ＝f(M 0 ,β 0 ) As a fitness function, the leakage current signal is subjected to VMD decomposition and firefly algorithm optimization to obtain S g Also the optimal parameters M and β of VMD are obtained.
Preferably, in the 5G leakage current fault monitoring method based on the improved VMD and the LMS, the signal VMD is decomposed as follows:
1) Initialization u 1 m }、{ω 1 m }、τ 1 m And n =0;
2) Entering into circulation;
3) Introducing a secondary penalty factor beta and a Lagrange multiplier tau, and converting into an unconstrained variational problem;
wherein u is m Decomposing M single-component amplitude modulation frequency modulation signals; omega m Amplitude modulating the center frequency of the frequency modulated signal for each single component; n is the number of iterations; tau is a Lagrangian multiplier; χ is a noise margin parameter; t is time; j is an imaginary unit; δ (t) is a pulse function;
calculating a partial derivative for t for the function;
4) For { u } 1 m }、{ω 1 m }、τ 1 m Updating according to an updating formula, wherein the updating formula is as follows:
5) Giving precision sigma, if the precision sigma is met, stopping the condition, and outputting M modal components;
6) The loop is stopped, otherwise step 2) is returned to.
Preferably, in the 5G leakage current fault monitoring method based on the improved VMD and the LMS, the flow of the firefly algorithm is as follows:
s21: initializing parameters of a firefly algorithm, including the total number n of firefly individuals and the maximum iteration times t of the population max Step factor α of disturbance at t =1, attraction force β, and absorption coefficient θ of light;
s22: calculating the relative brightness I of the firefly, and enabling the firefly with lower brightness to move to higher brightness;
s23: calculating the attraction beta and updating the brightness of the firefly;
s24: recording the current solution and comparing with the optimal solution;
s25: and (4) whether the stopping condition is met or not, if so, exiting and outputting a result, and otherwise, repeatedly executing S22-S24.
Preferably, in the 5G leakage current fault monitoring method based on the improved VMD and the LMS, in step S3, a calculation formula of a cross-correlation system R between the recombined signal x' (t) decomposed by the VMD and the leakage current original signal x (t) is as follows:
ratio E of time domain energy entropy k The calculation formula of (2) is as follows:
wherein E is k Effective IMF is more than or equal to 8 percent and R is more than or equal to 0.9; e k Not less than 8% and R<0.9 is the mixed IMF; e k <8% is noise IMF.
Preferably, in the 5G leakage current fault monitoring method based on the improved VMD and the LMS, the LMS algorithm has a simple structure and strong robustness, and in the step S4, the execution process of denoising the hybrid IMF by using the LMS algorithm is as follows:
s41: initialization of a filter: determining an initial weight vector w (0), an initial input signal p (0), η as a step factor and a filter order n
S42: for each new input sample p (n), calculating an output signal q (n);
s43: using the desired output d (n), the error signal e (n) is calculated, resulting in a gradient
S44: updating the weight vector using the LMS update formula:
s45: returning to S42 until finishing, an output sequence and an error sequence can be obtained.
Preferably, in the 5G leakage current fault monitoring method based on the improved VMD and the LMS, the leakage current may be monitored by performing shape extraction and recognition analysis on leakage current parameter information and environmental temperature information, in the step S5, the recombined leakage current signal information and the temperature information are combined, and a comprehensive trend recognition method is adopted to determine adaptive leakage current and temperature curve slopes respectively, and when the slope exceeds a predetermined value, it is determined as a sudden increase trend, and an alarm and an action are respectively performed, thereby finally realizing fault diagnosis and monitoring of the leakage current.
Preferably, in the 5G leakage current fault monitoring method based on the improved VMD and the LMS, a specific calculation process of fault diagnosis and monitoring of the leakage current is as follows:
s51: solving a current average value fc and a temperature average value ft according to the first S data uploaded by the S45;
s52: determining an initial current setting value oc and an initial temperature setting value ot, then respectively adding the next two sampling values into the sequence, and deleting the first two sampling values of the sequence;
s53: updating oc and ot again;
s54: fitting a leakage current and temperature curve and solving adaptive slopes Vc and Vt; and when the absolute value of Vc exceeds a preset value, transmitting the Vc to a master station through 5G communication for warning and analyzing and diagnosing faults in real time, and when the absolute value of Vt exceeds the preset value, the action of an executive element of the protector breaks a circuit to realize leakage current protection.
The 5G leakage current protector executes the 5G leakage current fault monitoring method based on the improved VMD and the LMS, and comprises the following steps:
the detection element is used for acquiring load leakage current information and temperature information;
the denoising element is used for the optimizing denoising step of the 5G leakage current fault monitoring method based on the improved VMD and LMS;
the amplifying element is used for amplifying the recombined leakage current signal and the temperature signal;
the threshold comparison element is used for monitoring and diagnosing the amplified recombined leakage current signal and the temperature signal;
the execution element is used for executing alarm and action according to the fault diagnosis result;
and the 5G communication element is used for communication between the leakage current protector and the main station.
The invention at least comprises the following beneficial effects:
1. the optimal decomposition number M and the quadratic penalty factor beta of the VMD are optimized by adopting a firefly algorithm, the modes in the VMD time domain are classified according to the ratio E of the cross correlation coefficient R and the time domain energy entropy, and finally the mixed modes are subjected to quadratic denoising by adopting an LMS algorithm, so that the denoising effect of the leakage current signal is improved, the optimization precision is high, the optimization speed is high, and the precision and the speed performance are very excellent.
2. The leakage current fault is identified according to the comprehensive trend identification method, the alarm and the action of the leakage current protector are realized according to the current and temperature self-adaptive slope Vc and Vt respectively, and the probability of the leakage current protector for false alarm and missing in a high-temperature environment is reduced.
3. Aiming at the problems that the detection capability of the original leakage current protector is insufficient and the communication module cannot transmit information in real time, the invention adopts the 5G communication module to transmit leakage current signals in real time, and is convenient for the main station to perform prediction analysis and real-time processing on the signals on the basis of reducing time delay.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention.
Drawings
FIG. 1 is a flow chart of the 5G leakage current fault monitoring method based on the improved VMD and LMS of the present invention;
FIG. 2 is a block diagram of the 5G leakage current protector of the present invention;
FIG. 3 is a flow chart of the firefly algorithm of the present invention;
FIG. 4 is a flow chart of the LMS algorithm of the present invention;
FIG. 5 is a flow chart of the integrated trend identification method of the present invention
FIG. 6 is a graph showing the convergence comparison of the firefly algorithm with the particle swarm algorithm and the gray wolf search algorithm of the present invention;
FIG. 7 is a graph of an optimization of FA-VMD based penalty factors of the present invention;
FIG. 8 is a graph of the optimization of the FA-VMD based decomposition mode number according to the present invention;
FIG. 9 is a graph comparing the modal components of FA-VMD of the present invention with the initial leakage current signal.
Detailed Description
The present invention is further described in detail below with reference to the drawings and examples so that those skilled in the art can practice the invention with reference to the description.
It should be noted that the experimental methods described in the following embodiments are all conventional methods unless otherwise specified, and the reagents and materials described therein are commercially available unless otherwise specified.
At present, the leakage current protector is widely used at home and abroad, and the use of the leakage current protector is effective for preventing electrical fire, electrical equipment damage and human body electric shock, and becomes an important component part in the safety protection of a low-voltage power grid. The leakage current protector is widely applied at home and abroad, and most developed countries put emphasis on the prevention of personal electric shock and fire. However, with the increasingly complex power system environment and the increasing public demand for daily power safety, the current leakage current protector cannot meet the demand. Many electric shock accidents and electrical fires are caused by insufficient detection capability of the leakage current protector.
The leakage current protector structure mainly comprises four basic parts, namely a detection element, a denoising element, an amplifying element, a threshold value comparison element and an execution element. As shown in the figure 1, the method specifically comprises the following steps: a leakage current and temperature detection element, an improved VMD and LMS denoising element, an amplification element, a leakage current and temperature threshold comparison element and an actuator. Wherein, the executive component comprises an alarm device, a switch and the like. The leakage current detection element can be a leakage current sensor, and the current collection frequency is set according to the requirement and is generally collected for 10 to 100 times/second; the temperature detection element can be a temperature sensor, and the frequency of the acquired signal is also set according to the requirement and is also 10-100 times/second.
When the leakage current protector executes the 5G leakage current fault monitoring method based on the improved VMD and the LMS:
when the trends of the leakage current are judged to be sudden increase, the alarm device alarms the master station through 5G communication and conducts real-time diagnosis and analysis, and when the trends of the temperature are judged to be sudden increase, the executive component acts, and the main switch disconnects the circuit. Compared with the existing 4G network, the 5G network can meet the requirements of ultrahigh bandwidth, ultralow time delay and ultra-large scale connection, and is used for further realizing the interconnection of everything and further serving various vertical industries. The 5G communication module is applied to the leakage current protector, so that high time delay between the leakage current protector and a main station can be effectively solved, and real-time monitoring and analysis decision of line leakage current are realized.
VMD is an adaptive signal processing method proposed in recent years, and is widely used for signal denoising and feature extraction.
The invention improves the VMD, mainly comprising the following two points:
1. and optimizing the optimal decomposition number M and the secondary punishment factor beta of the VMD by adopting a firefly algorithm.
2. And according to the ratio E of the cross correlation coefficient R and the time domain energy entropy as a classification index, classifying the mode in the time domain into three categories of effective IMF, mixed IMF and noise IMF, reserving the effective IMF and removing the noise IMF. And denoising the mixed IMF by using an LMS algorithm, and superposing a denoised mode and an effective mode to obtain a recombined leakage current signal.
The firefly algorithm is a group optimization algorithm simulating information exchange among fireflies and attraction and aggregation behaviors of the fireflies, and the principle is simple, but the proposal time is short.
In the present invention, the algorithm rule is idealized as the following three points:
(1) No distinction was made between the genders of all fireflies. Each firefly can be attracted by any other firefly;
(2) The brightness of the firefly is only related to the objective function. To solve the luminance optimization problem, the luminance is proportional to the value of the objective function. An alternative form of luminance is established, for example, using a method similar to the fitness function.
(3) The attraction of fireflies is only related to the brightness of fireflies. The darker firefly will move towards the brighter firefly. In addition, the relative brightness decreases as the distance between fireflies increases. If the situation arises that no brighter firefly can be found, the firefly will move randomly within the search space.
(4) The intensity I and the attraction force psi both vary with the distance r.
Specifically, it can be given by formulas (1) and (2):
I 0 and psi 0 The initial luminance and the attraction force when the distance is 0, respectively, θ is the absorption coefficient of light, and r is the distance between fireflies. The distance r between two fireflies i and j is given by the equation expressed in (3)
The location update formula of fireflies at each subsequent time is given by the following equation (4):
the first term in the formula (4) represents the current t-time position of the firefly, the second term represents the distance between the two fireflies generated due to the attraction of the fireflies, and the last term represents the random disturbance of the fireflies, so that the search area is favorably enlarged, and the premature stagnation of the algorithm is avoided. Where α is the step factor of the perturbation and is a constant between 0 and 1, G i Subject to the amount of change in the gaussian distribution. If the brightness of the fireflies is the same, the fireflies move randomly respectively, and through continuous updating of the fireflies, the group can finally gather at the fireflies position with the highest brightness, so that the target optimization is realized. The basic flow of the firefly algorithm is shown in fig. 3.
Firefly in the present invention has an envelope entropy value S g ＝f(M 0 ,β 0 ) As a fitness function, the signal is subjected to VMD and FA optimization processes to obtain S g The optimal parameter decomposition number M and the secondary penalty factor beta of the VMD decomposition are also obtained.
In the present invention, the specific process of VMD decomposition is as follows:
1) Initialization u 1 m }、{ω 1 m }、τ 1 m And n =0;
2) Entering into circulation;
3) Introducing a secondary penalty factor beta and a Lagrange multiplier tau, and converting into an unconstrained variational problem;
wherein u is m For the decomposed M single-component AM FM signals; omega m Amplitude modulating the center frequency of the frequency modulated signal for each single component; n is the number of iterations; τ is LagrangeA daily multiplier; χ is a noise margin parameter; t is time; j is an imaginary unit; δ (t) is a pulse function;
calculating a partial derivative for t for the function;
4) For { u } 1 m }、{ω 1 m }、τ 1 m Updating according to an updating formula, wherein the updating formula is as follows:
5) Giving precision sigma, if the precision sigma is met, stopping the condition, and outputting M modal components;
6) The loop is stopped, otherwise step 2) is returned to.
VMD decomposition is carried out on the leakage current original signal by utilizing the optimal M and beta values obtained by optimizing, and the mode in the time domain is divided into three types of effective IMF, mixed IMF and noise IMF according to the ratio E of the cross correlation coefficient R and the time domain energy entropy as a classification index. Wherein the R between the recombined signal x' (t) obtained by VMD decomposition and the original signal x (t) is calculated by the following formula:
further, the ratio E of the time domain energy entropy k The calculation formula of (c) is:
wherein x k (t) is K IMF components, and the model in the time domain can be divided into effective IMF (Ek is more than or equal to 10% and R is more than or equal to 0.8) and mixed IMF (Ek is more than or equal to 10% and R is more than or equal to 0.8) based on the analysis of the leakage current signal<0.8 ) sum noiseAcoustic IMF (Ek)<10%). And after classification, keeping the effective IMF, removing the noise IMF, and performing secondary denoising on the mixed IMF through an LMS algorithm.
The LMS algorithm is based on wiener filtering and is derived by the steepest descent algorithm. The wiener solution solved by wiener filtering must be determined knowing the a priori statistical information of the input signal and the desired signal and then performing an inversion operation on the autocorrelation matrix of the input signal. Based on the criterion of minimum mean square error, the LMS algorithm minimizes the mean square error between the output signal of the filter and the expected output signal, and the vector signal flow chart is shown in FIG. 4.
Where p (n) and w (n) are the input leakage current signal vector and weight vector, respectively, at time n, and d (n) is the desired output value. The LMS algorithm has simple structure and strong robustness, so that the LMS algorithm can be used for secondary denoising of hybrid IMF, and the execution process comprises the following steps:
1) Initialization of a filter: determining an initial weight vector w (0), an initial input signal p (0), a step factor η and a filter order n
2) For each new input sample p (n), the output signal q (n) is calculated
q(n)＝w T (n)p(n) (7)
3) Using the desired output d (n), the error signal e (n) is calculated, resulting in a gradient
4) Updating the weight vector by using an LMS updating formula:
w(n+1)＝w(n)+ηe(n)p(n) (9)
5) And returning to the step 2) until the end, and obtaining an output sequence and an error sequence.
And then, after a denoising current sequence is obtained, the fault monitoring of the leakage current is realized by combining the data of the temperature sensor. Under normal conditions, the leakage current and temperature trends are maintained in a steady state, and the leakage current fluctuates irregularly up and down only when a fault occurs. Therefore, as shown in fig. 5, a comprehensive trend recognition algorithm is selected, when the trends of the leakage current are judged to suddenly increase, the alarm device alarms to the main station through 5G communication and performs real-time diagnosis and analysis, and when the trends of the temperature are judged to suddenly increase, the execution element acts, and the main switch disconnects the circuit. The calculation process is as follows:
firstly, the average value f of the first s data uploaded by the denoising element is calculated c And f t ；
Determining initial current and temperature settings o c And o t
o c ＝f c +Δl (10)
o t ＝f t +Δl (11)
Where Δ l is a margin preset based on past fault data.
Then, the next two sampling values are added into the sequence respectively, and the first two sampling values of the sequence are deleted. Update o again c And o t Fitting the current and temperature curves and calculating the adaptive slope V thereof c And V t When V is c When the absolute value of the voltage exceeds a preset value, the voltage is transmitted to a main station through 5G communication to alarm and analyze and diagnose the fault in real time, and when V is greater than a preset value t When the absolute value of the protector exceeds a preset value, the protector execution element acts to disconnect the circuit to realize leakage current protection.
Comparison of Performance tests
The FA firefly algorithm, PSO (particle swarm optimization) and GWOL (wolf search algorithm) are processed by a fitness function S g ＝f(M 0 ,β 0 ) In contrast, the convergence ratio of the algorithm is shown in fig. 6.
According to analysis and comparison with other two traditional group intelligent optimization algorithms, the firefly algorithm has higher performance in the aspect of local search, and is high in optimization precision, high in optimization speed, and excellent in precision and optimization speed.
Through data acquired by the leakage current protector, optimization curves of the penalty factor beta and the modal number M obtained through the FA-VMD process are shown in FIGS. 7 and 8, and it can be seen that the optimal parameter =7 of M and the optimal parameter =95.38 of beta are finally determined after 10 iterations and 2 iterations respectively.
Substituting the optimal parameters M and beta into the VMD decomposition process to obtain a comparison graph of each modal component of the FA-VMD and the initial leakage current signal, as shown in FIG. 9, wherein the sampling frequency is 10 times/second;
the IMF1 is a leakage current signal after FA-VMD denoising, the variation trend of the leakage current signal on different details is reflected, the denoising effect is obvious by comparing with an initial leakage current signal, and the leakage current information of a line can be more effectively reflected. And the other components IMF 2-IMF 7 reflect the randomness of the initial leakage current signal, and all the components are stably distributed on two sides, so that the characteristics of the leakage current signal are more clearly displayed.
In conclusion, the 5G leakage current fault monitoring method based on the improved VMD and the LMS can overcome the defects of poor current leakage current denoising effect and low identification precision, denoising is carried out on the leakage current signals through the improved VMD and the improved LMS, meanwhile, leakage current fault warning, diagnosis and real-time communication are carried out by combining the 5G communication module and the comprehensive trend identification method, the leakage current signals can be identified efficiently and accurately, and guarantee is provided for reducing the occurrence of electric fire.
While embodiments of the invention have been disclosed above, it is not intended that they be limited to the applications set forth in the specification and examples. It can be applied to all kinds of fields suitable for the present invention. Additional modifications will readily occur to those skilled in the art.
Claims (10)
1. 5G leakage current fault monitoring method based on improved VMD and LMS is characterized by comprising the following steps:
s1: acquiring load leakage current information and temperature information as original signals;
s2: optimizing the leakage current original signal by using a firefly algorithm to obtain the optimal decomposition number M and a penalty factor beta of the signal VMD;
s3: performing VMD decomposition on a leakage current original signal by using the optimal M and beta values obtained by optimization, taking the cross-correlation coefficient R and the time domain energy entropy ratio E of a recombined signal obtained by decomposition and the original signal as classification indexes, dividing the intra-time domain mode into three categories of effective IMF, mixed IMF and noise IMF, reserving the effective IMF and removing the noise IMF;
s4: denoising the mixed IMF by using an LMS algorithm, and superposing a denoised mode and the effective IMF to obtain a recombined leakage current signal;
s5: and monitoring the recombined leakage current signal and the temperature signal, diagnosing faults, respectively alarming and acting by the protector when the faults occur, and transmitting information to the main station for analysis and processing through the 5G communication module in real time during the process, so that the phenomenon of false alarm and missing report is avoided.
2. The improved VMD and LMS based 5G leakage current fault monitoring method according to claim 1, wherein the leakage current information is a vector sum of phase-to-ground leakage currents.
3. The improved VMD and LMS based 5G leakage current fault monitoring method of claim 1, wherein specifically, said firefly algorithm uses an envelope entropy S g ＝f(M 0 ,β 0 ) As a fitness function, the leakage current signal is subjected to VMD decomposition and firefly algorithm optimization to obtain S g Also the optimal parameters M and β of VMD are obtained.
4. The improved VMD and LMS based 5G leakage current fault monitoring method of claim 3, wherein the VMD decomposition process is as follows:
1) Initialization u 1 m }、{ω 1 m }、τ 1 m And n =0;
2) Entering into circulation;
3) Introducing a secondary penalty factor beta and a Lagrange multiplier tau, and converting into an unconstrained variational problem;
wherein u is m For the decomposed M single-component AM FM signals; omega m Amplitude modulating the center frequency of the frequency modulated signal for each single component; n is the number of iterations; τ is the Lagrangian multiplier; χ is a noise margin parameter; t is time; j is an imaginary unit; δ (t) is a pulse function;
calculating a partial derivative for t for the function;
4) For { u 1 m }、{ω 1 m }、τ 1 m Updating according to an updating formula, wherein the updating formula is as follows:
5) Giving precision sigma, if the precision sigma is met, stopping the condition, and outputting M modal components;
6) The loop is stopped, otherwise step 2) is returned to.
5. The improved VMD and LMS based 5G leakage current fault monitoring method of claim 4, wherein the flow of said firefly algorithm is:
s21: initializing parameters of a firefly algorithm, including the total number n of firefly individuals and the maximum iteration times t of a population max Step factor α of disturbance at t =1, attraction ψ, absorption coefficient θ of light;
s22: calculating the relative brightness I of the firefly, and enabling the firefly with lower brightness to move to higher brightness;
s23: calculating the attraction psi and updating the firefly brightness;
s24: recording the current solution and comparing with the optimal solution;
s25: and (4) judging whether the stopping condition is met, if so, exiting and outputting the result, and otherwise, repeatedly executing S22-S24.
6. The improved VMD and LMS based 5G leakage current fault monitoring method according to claim 1, wherein in step S3, the cross-correlation R between the reconstructed signal x' (t) decomposed by the VMD and the leakage current original signal x (t) is calculated as follows:
ratio E of time domain energy entropy k The calculation formula of (2) is as follows:
wherein E is k Effective IMF is more than or equal to 8 percent and R is more than or equal to 0.9; e k Not less than 8% and R<0.9 is mixed IMF; e k <8% is noise IMF.
7. The improved VMD and LMS based 5G leakage current fault monitoring method of claim 1, wherein the step S4, the performing of denoising the hybrid IMF by using the LMS algorithm comprises:
s41: initialization of a filter: determining an initial weight vector w (0), an initial input signal p (0), a step factor η and a filter order n;
s42: for each new input sample p (n), computing an output signal q (n);
s43: using the desired output d (n), the error signal e (n) is calculated, resulting in a gradient
S44: updating the weight vector using the LMS update formula:
s45: returning to S42 until finishing, an output sequence and an error sequence can be obtained.
8. The improved VMD and LMS-based 5G leakage current fault monitoring method according to claim 7, wherein in the step S5, the recombined leakage current signal information and temperature information are combined, and a comprehensive trend recognition method is adopted to determine adaptive leakage current and temperature curve slopes respectively, and when the slope exceeds a predetermined value, it is determined as a sudden increase trend, and an alarm and an action are respectively performed, thereby finally realizing fault diagnosis and monitoring of leakage current.
9. The improved VMD and LMS based 5G leakage current fault monitoring method of claim 8, wherein the fault diagnosis and monitoring of the leakage current is calculated as follows:
s51: solving a current average value fc and a temperature average value ft according to the previous S data uploaded in the S45;
s52: determining an initial current setting value oc and an initial temperature setting value ot, then respectively adding the next two sampling values into the sequence, and deleting the first two sampling values of the sequence;
s53: updating oc and ot again;
s54: fitting a leakage current and temperature curve and solving adaptive slopes Vc and Vt; when the absolute value of Vc exceeds a preset value, the Vc is transmitted to a main station through 5G communication to alarm and analyze and diagnose faults in real time, and when the absolute value of Vt exceeds the preset value, the action of an execution element of the protector breaks a circuit to realize leakage current protection.
10.5G earth leakage protector, characterized by comprising the following components required for executing the 5G leakage current fault monitoring method based on modified VMD and LMS of claim 1:
the detection element is used for acquiring load leakage current information and temperature information;
a denoising element for optimizing denoising;
the amplifying element is used for amplifying the signal;
the threshold comparison element is used for monitoring and diagnosing the amplified signals;
the execution element is used for executing alarm and action according to the fault diagnosis result;
and the 5G communication element is used for communication between the 5G leakage protector and the main station.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211222093.3A CN115684828A (en) | 2022-10-08 | 2022-10-08 | 5G leakage current fault monitoring method based on improved VMD and LMS |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211222093.3A CN115684828A (en) | 2022-10-08 | 2022-10-08 | 5G leakage current fault monitoring method based on improved VMD and LMS |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115684828A true CN115684828A (en) | 2023-02-03 |
Family
ID=85065254
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211222093.3A Pending CN115684828A (en) | 2022-10-08 | 2022-10-08 | 5G leakage current fault monitoring method based on improved VMD and LMS |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115684828A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117368648A (en) * | 2023-11-08 | 2024-01-09 | 国网四川省电力公司电力科学研究院 | Power distribution network single-phase earth fault detection method, system, equipment and storage medium |
| CN117368648B (en) * | 2023-11-08 | 2024-06-04 | 国网四川省电力公司电力科学研究院 | Power distribution network single-phase earth fault detection method, system, equipment and storage medium |
-
2022
- 2022-10-08 CN CN202211222093.3A patent/CN115684828A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117368648A (en) * | 2023-11-08 | 2024-01-09 | 国网四川省电力公司电力科学研究院 | Power distribution network single-phase earth fault detection method, system, equipment and storage medium |
| CN117368648B (en) * | 2023-11-08 | 2024-06-04 | 国网四川省电力公司电力科学研究院 | Power distribution network single-phase earth fault detection method, system, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107204876B (en) | Network security risk assessment method | |
| Huang et al. | High-impedance fault detection utilizing a Morlet wavelet transform approach | |
| CN109633368B (en) | VMD and DFA-based power quality disturbance detection method for power distribution network with distributed power supplies | |
| US7069116B2 (en) | High impedance fault detection | |
| CN111585948B (en) | Intelligent network security situation prediction method based on power grid big data | |
| CN107340459A (en) | A kind of DC Line Fault arc method for measuring and system | |
| CN109739214B (en) | Method for detecting intermittent faults in industrial process | |
| CN105510760A (en) | Method for detecting short circuit fault data based on wavelet analysis | |
| CN109241989A (en) | A kind of method and system of the intelligent substation intrusion scenario reduction based on space-time similarity mode | |
| CN110852441B (en) | Fire disaster early warning method based on improved naive Bayes algorithm | |
| CN108508297B (en) | Fault arc detection method based on mutation coefficient and SVM | |
| CN116819250A (en) | Method for detecting fault arc in photovoltaic system | |
| CN112183628A (en) | Alternating current arc fault detection method and system based on multiple linear time-frequency transformations | |
| Sia et al. | PMU change point detection of imminent voltage collapse and stealthy attacks | |
| CN115684828A (en) | 5G leakage current fault monitoring method based on improved VMD and LMS | |
| CN113466593A (en) | Non-invasive load monitoring safety protection method and sample library establishment method | |
| CN113376474A (en) | Neural network fault arc identification system and method based on generalized S transformation | |
| CN111693829A (en) | Partial discharge noise and discharge distinguishing method for non-contact ultrasonic detection | |
| CN109581211A (en) | A kind of load ratio bridging switch mechanical breakdown on-line monitoring method based on current of electric | |
| CN115327325A (en) | Charging pile series arc fault detection method and device and electronic equipment | |
| CN213365710U (en) | Thunder and lightning fire early warning system | |
| Chen et al. | A data preparation method for machine-learning-based power system cyber-attack detection | |
| Tong et al. | A novel acceleration criterion for remote-end grounding-fault in MMC-MTDC under communication anomalies | |
| Zhou et al. | Method of Series Arc Fault Detection Based on Phase Space Reconstruction and Convolutional Neural Network | |
| CN113269938B (en) | On-line monitoring system and method for pole tower burglary prevention based on OPGW of power transmission line |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |