Signal denoising method for optimizing ICEEMD decomposition based on improved SSA algorithm
By improving the SSA algorithm to optimize ICEEMD decomposition and adaptively adjusting the noise standard deviation ratio and ensemble number, the problems of mode aliasing and noise residue in the ICEEMD method are solved, thereby improving the denoising effect and positioning accuracy of pipeline leakage signals.
Patent Information
- Application Number
- CN202510960938.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-12
- Publication Date
- 2025-11-11
AI Technical Summary
Existing ICEEMD methods suffer from modal aliasing and noise residue in pipeline leak detection. Traditional Sparrow Search Algorithm (SSA) suffers from premature convergence and blind chaotic perturbation, which affect the signal denoising effect.
An improved sparrow search algorithm (SSA) based on population diversity adaptive Tent chaotic mapping is adopted to optimize ICEEMD decomposition. The noise standard deviation ratio and ensemble number are adaptively adjusted, and the sparrow position update is optimized through adaptive Tent chaotic mapping, thereby improving parameter self-adaptability and reducing human intervention.
It achieves better signal denoising, improves the signal-to-noise ratio and reduces the mean square error, thereby enhancing the positioning accuracy of pipeline leak signals.
Smart Images

Figure CN120929720A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of signal denoising, and more specifically to a signal denoising method based on an improved SSA algorithm to optimize ICEEMD decomposition. Background Technology
[0002] Pipeline leaks not only cause resource waste and economic losses, but also impact people's lives and livelihoods. Therefore, mature pipeline leak detection technology is essential for minimizing these social losses. However, in actual detection processes, the acoustic signal of a leak is often mixed with a significant amount of noise due to the surrounding environment. This noise increases calculation errors, thereby reducing location accuracy. Therefore, when using acoustic signal methods for leak location, selecting an appropriate signal decomposition and noise reduction method is crucial.
[0003] Empirical Mode Decomposition (EMD) is commonly used to process nonlinear and non-stationary signals. However, EMD suffers from two serious problems: mode aliasing and endpoint effects, which reduce noise suppression effectiveness. ICEEMD significantly reduces mode aliasing by introducing white noise, but the ICEEMD method requires setting the noise standard deviation ratio (Nstd) and the number of integrations (NE) when decomposing the signal. Traditional methods rely on human experience and are difficult to adapt to complex noise environments. Fixed parameters may lead to mode aliasing or noise residue, affecting denoising accuracy.
[0004] Currently, among the relevant algorithms, for example, the patent application number 202410181976.7 entitled "A Shock Wave Signal Resonance Noise Reduction Method Based on Sparrow Search Algorithm to Optimize VMD" uses the Sparrow Search Algorithm to optimize the parameters of VMD decomposition. However, the traditional Sparrow Search Algorithm (SSA) has problems such as premature convergence and blind chaotic perturbation. Therefore, it is urgent to solve this problem. Summary of the Invention
[0005] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition. It is based on the sparrow search algorithm improved by adaptive Tent mapping of population diversity to optimize the noise parameters of ICEEMD, better achieve parameter self-adaptation, reduce manual intervention and improve the signal denoising effect.
[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is: a signal denoising method based on an improved SSA algorithm to optimize ICEEMD decomposition, comprising: Collect raw leak signals from the pipeline; The original leakage signal was decomposed using ICEEMD to obtain multiple IMF components. During the decomposition process, the SSA algorithm optimized by adaptive Tent chaotic mapping based on population diversity was used to optimize the number of integrations and the noise standard deviation ratio. Calculate the sample entropy of each IMF component, and select the effective IMF components based on the sample entropy magnitude; The effective IMF components are reconstructed to obtain the denoised pipeline leakage signal.
[0007] Furthermore, during the decomposition of the original leakage signal using ICEEMD, the g-1th residual component... Construct the g-th signal to be decomposed, and based on the g-th signal to be decomposed... The g-th intrinsic mode component is obtained. Specifically, it is expressed as:
[0008]
[0009]
[0010] In the formula, g = 3, ..., N N This indicates the number of ensemble iterations performed using the SSA algorithm optimized with an adaptive Tent chaotic map based on population diversity. This represents the (g-1)th noise figure. This represents the ratio of noise standard deviations obtained using the SSA algorithm optimized with an adaptive Tent chaotic map based on population diversity. This indicates the calculation of the standard deviation. This represents the g-th signal to be decomposed; This indicates the calculation of the local mean of the signal, where H represents the number of iterations.
[0011] Furthermore, the SSA algorithm is optimized using an adaptive Tent chaotic mapping based on population diversity; specifically:
[0012] Step a, define the adaptive Tent chaotic mapping parameters as follows:
[0013] in, Let the chaotic mapping parameters of the t-th iteration satisfy the following condition: D(t) is the normalized population diversity index for the t-th iteration, calculated using the following formula:
[0014] Where N is the number of individual sparrows, and L is the length of the diagonal of the solution space. Let be the mean of the j-th dimension variable in the t-th generation population, and d be the optimization dimension. Let be the parameter value of the i-th sparrow individual in the j-th generation of the population; Step b, generating an adaptive Tent chaotic sequence based on the adaptive Tent chaotic mapping parameters, specifically as follows:
[0015] in, for Interval chaotic sequence, initial value Generate N chaotic sequences
[0016] The value, N, represents the number of individual sparrows; Step c, the chaotic values are mapped to the parameter space, represented as:
[0017] in, This represents the i-th chaotic value. This represents the parameter position of the k-th individual sparrow.
[0018] Furthermore, during the decomposition process, the optimized SSA algorithm is used to optimize the ensemble number and noise standard deviation ratio; specifically including: Step A: Determine the noise standard deviation ratio Number of integrations ; Step B, determine the objective function as:
[0019] In the formula, Fitness is the fitness function, and SNR represents the signal-to-noise ratio. These are the weighting coefficients. The energy entropy of the IMF component after ICEEMD decomposition; Step C, set the parameter position of the i-th individual sparrow. Sparrow position updates incorporate adaptive chaotic perturbations, specifically including discoverer position updates, follower position updates, and vigilant position updates. Step D, Termination condition: Repeat step C until the maximum number of iterations is reached; Step E: After the iteration is complete, select the individual with the highest fitness. The optimal noise standard deviation ratio and optimal number of integrations Substitute this into the ICEEMD decomposition process.
[0020] Furthermore, during the sparrow location update process, The formula for updating the discoverer's location is:
[0021] In the formula, This represents the j-th dimension position of the i-th sparrow in the (t+1)-th generation. This represents the j-th position of the i-th sparrow in the t-th generation; Random values between Let be the adaptive Tent chaotic sequence values; T is the maximum number of iterations, Q is a random number following a normal distribution, and λ is a 1×d-dimensional matrix where each element is 1. , where is a random number, and ST is the safety threshold; The formula for updating the vigilant's position is:
[0022] In the formula, It is the globally optimal position in the j-th dimension when the current iteration number t is reached, and β and The step size control parameter is a random number that follows a standard normal distribution and is called the step size adjustment factor. N is the number of sparrows. The coefficient of intensity of chaotic disturbance. For the randomly selected j-th dimension sparrow position in the t-th generation, Let be the sparrow in the j-th dimension that is in the (t+1)-th generation population.
[0023] Furthermore, before using ICEEMD to decompose the original leakage signal, the following steps are also included: The collected raw pipeline leakage signals are filtered.
[0024] The present invention also relates to a signal denoising device based on an improved SSA algorithm to optimize ICEEMD decomposition, comprising: The acquisition module is used to acquire raw leakage signals from the pipeline. The decomposition module is used to decompose the original leakage signal using ICEEMD to obtain multiple IMF components. During the decomposition process, the SSA algorithm optimized by adaptive Tent chaotic mapping based on population diversity is used to optimize the number of integrations and the noise standard deviation ratio. The calculation and selection module is used to calculate the sample entropy of each IMF component and select the valid IMF components based on the sample entropy magnitude. The reconstruction module is used to reconstruct the effective IMF components to obtain the denoised pipeline leakage signal.
[0025] The present invention also relates to an apparatus comprising: Memory, used to store computer programs; A processor is configured to execute the computer program, which, when executed by the processor, implements the steps of the signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition.
[0026] The present invention also relates to a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition.
[0027] The present invention also relates to a computer program product, comprising a computer program that, when executed by a processor, implements the steps of the signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition.
[0028] After adopting the above technical solution, the present invention seeks optimization through its adaptive parameter adjustment mechanism, iteratively calculates the optimal values of Nstd and NE required for ICEEMD decomposition, and then uses these optimal values to decompose the pipeline leakage signal. After the decomposition is completed, the sample entropy of each IMF component is calculated, the IMF component with the larger sample entropy is selected to reconstruct the signal, and finally the signal-to-noise ratio of the reconstructed signal is calculated. It is found that this method can achieve a good noise reduction effect. Attached Figure Description
[0029] Figure 1 This is a flowchart of the signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition according to the present invention; Figure 2 This is a diagram of the laboratory apparatus of the present invention; Figure 3 for Figure 2 Pipeline layout schematic diagram; Figure 4 The iterative graph of the optimized SSA algorithm; Figure 5 This is a time-domain diagram of the IMF components of the present invention; Figure 6 This is a spectrum diagram of the IMF components of the present invention; Figure 7 The diagram shows the original leakage signal and the reconstructed signal of the pipeline. Detailed Implementation
[0030] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings.
[0031] like Figure 1 As shown, a signal denoising method based on an improved SSA algorithm to optimize ICEEMD decomposition includes: Step S1, collect the original leakage signal of the pipeline; specifically including: Step S11, acquire the noisy signal s(m), sampling frequency m = 1, 2, ..., M, where M is the signal length; Step S12: Filter the acquired signal to obtain the processed signal x(m); Step S2 involves decomposing the signal x(m) using ICEEMD to obtain multiple IMF components. During the decomposition process, the SSA algorithm, optimized using adaptive Tent chaotic mapping based on population diversity, is employed to optimize the ensemble number and noise standard deviation ratio. Specifically: Step S21: Add zero-mean, unit-covariance Gaussian white noise to the filtered signal, expressed as:
[0032] Where x represents the filtered signal from the original signal. Represents the noise figure. This represents the first intrinsic mode component of the signal after EMD decomposition; Step S22: Calculate the local mean of the upper and lower envelopes of the signal during the EMD process, and obtain the first residual component. Subtract the residual from the noisy signal x to obtain the first intrinsic mode component (IMF), including the following process:
[0033]
[0034] in, Indicates the first residual component; This indicates the calculation of the local mean of the signal, where H represents the number of iterations. Indicates the first intrinsic mode component; Step S23: Calculate the second residual and the second eigenmode component.
[0035]
[0036] S24: Construct the g-th signal to be decomposed using the (g-1)-th residual component, and obtain the g-th intrinsic mode component based on the g-th signal to be decomposed:
[0037]
[0038] Where g = 3, ..., N N This represents the number of ensemble iterations used in the optimized SSA algorithm based on population diversity adaptive Tent chaotic mapping. Traditional ICEEMD selects the number of ensemble iterations NE based on experience, but an excessively large NE leads to excessive computation, while an excessively small NE results in residual noise. This represents the (g-1)th noise figure. , The noise standard deviation ratio is optimized using the SSA algorithm based on population diversity-adaptive Tent chaotic mapping. This indicates the calculation of the standard deviation. This represents the g-th signal to be decomposed. Let N represent the (g-1)th residual component. The traditional noise standard deviation ratio Nstd is empirically derived, and improper selection can lead to poor decomposition results. Therefore, this invention calculates the optimal Nstd using an optimized SSA algorithm. as well as The value is substituted into the ICEEMD decomposition; where the optimized SSA algorithm is used to optimize the number of integration steps and the noise standard deviation ratio, to obtain N. and Specifically, this includes: Step A, determine the optimization variables, which are: Noise standard deviation ratio Number of integrations ; Step B, determine the objective function (joint optimization of maximizing signal-to-noise ratio and minimizing energy entropy):
[0039] In the formula, Fitness is the fitness function, and SNR represents the signal-to-noise ratio. These are the weighting coefficients. , can be taken as 0.7; The energy entropy of the IMF component after ICEEMD decomposition is specifically expressed as:
[0040] in, Let K be the energy percentage of the i-th layer of the IMF, and K be the total number of layers in the IMF. These represent the energies of the i-th and j-th layer IMFs, respectively. Step C: Traditional Tent chaotic mapping uses fixed mapping parameters. This method is too sensitive to parameters and has poor periodicity. Therefore, in this embodiment, the adaptive Tent chaotic mapping parameters are defined as follows:
[0041] in, Let the chaotic mapping parameters of the t-th iteration satisfy the following condition: , ≥√2≈1.5, with the upper boundary of 1.8 as the threshold (determined by the Lyapunov exponent threshold). A value that is too large will result in an excessively strong perturbation, while a value that is too small will result in an excessively weak perturbation. In this embodiment, the value is set to 1.65 based on experience. - ≥0.05 (sensitivity threshold) ≥1.8 + 0.05 = 1.85 ≤1.99<2.0, The value of affects the convergence speed. In this embodiment, 1.98 is chosen based on experience, as this value results in faster convergence. D(t) is the normalized population diversity index for the t-th iteration, calculated using the following formula:
[0042] Where N is the sparrow population size, and L is the length of the diagonal of the solution space. Let be the mean of the j-th dimension variable in the t-th generation population, and d be the optimization dimension. Let be the parameter value of the i-th sparrow individual in the j-th generation of the population. With diversity index Dynamic adjustment breaks through the static limitations of fixed chaotic mapping; Step D: Generate an adaptive Tent chaotic sequence based on the adaptive Tent chaotic mapping parameters, as detailed in the table below. As shown:
[0043] in, for Interval chaotic sequence, initial value Generate N chaotic sequences
[0044] The value, N, is the population size. The traditional SSA algorithm uses a fixed value. Using values to generate chaotic sequences will result in insufficient perturbation strength, while adaptive... The value can be randomly adjusted according to the population situation to find the most suitable perturbation value for the population, so that the population can escape the local optimum. Step E: The chaotic values are mapped to the parameter space, represented as follows:
[0045] in, This represents the parameter position of the k-th individual sparrow; Step F introduces adaptive chaotic perturbation for sparrow position updates, including discoverer position updates, follower position updates, and vigilant position updates; Discoverer location update:
[0046] In the formula, This represents the j-th dimension position of the i-th sparrow in the (t+1)-th generation. This represents the j-th position of the i-th sparrow in the t-th generation; Random values between Let be the adaptive chaotic sequence values generated in step D; T is the maximum number of iterations; Q is a random number following a normal distribution; and λ is a 1×d-dimensional matrix where each element is 1. , where is a random number, representing the warning value issued by an individual sparrow in the population when it encounters danger; ST is the safety threshold. ;when This indicates that the current population is in a relatively safe location, and the discoverer can conduct a broad search to guide the population to obtain richer resources. When this occurs, it indicates that sparrows have spotted a predator and released a danger signal; the population immediately adjusts its search strategy and moves towards a safe area. The improved discoverer location update introduces chaos values. This can enhance population diversity and better balance the global search; Followers will accompany the discoverer sparrow in foraging to replenish its food supply. If the food has already been distributed when they arrive, they will have to leave the flock to find food again. The follower's position is updated as follows:
[0047] In the formula, Let j be the j-th dimension position of the i-th sparrow in the (t+1)-th generation. This represents the j-th position of the i-th sparrow in the t-th generation; Let represent the sparrow in the (t+1)th generation population that occupies the optimal position in the j-th dimension. Let A represent the position of the sparrow with the lowest fitness in generation t. Let A be a random matrix of 1×d dimensions, where each value is either 1 or -1. Let λ be a 1×d matrix of 1×d dimensions, where each element is 1. Let Q be random numbers following a normal distribution. N is the number of individual sparrows, when When , it means that the i-th follower sparrow has not obtained food, is in a state of hunger, and needs to fly to other places to find food.
[0048] When a sparrow population encounters a predator, sparrows on the outer edges of the population will move closer to their companions in the middle or inner parts of the population. These sparrows that detect danger are called vigilant individuals, accounting for 10%-20% of the population. The vigilant position update in adaptive chaotic escape is specifically represented as follows:
[0049] In the formula, It is the globally optimal position in the j-th dimension when the current iteration number t is reached. This represents the sparrow in the (t+1)th generation population that occupies the optimal position in the j-th dimension; β and The step size control parameter is a random number that follows a standard normal distribution and is called the step size adjustment factor. N is the population size. The coefficient of intensity of chaotic disturbance. Let be the randomly selected position of the j-th dimension sparrow in the t-th generation.
[0050] Compared to the traditional SSA algorithm where the guard's position is updated, the introduction of adaptive chaos values allows the guard to search the surrounding space more meticulously, increasing the probability of finding the optimal solution. Step G, termination condition: Repeat step F until the maximum number of iterations is reached; Step H: After the iteration is complete, select the individual with the highest fitness. The optimal noise standard deviation ratio and optimal number of integrations Substitute into the ICEEMD decomposition process; Step S25: Repeat step S24 until the residual components can no longer be decomposed and the iteration stops, thus obtaining all intrinsic mode components and residual components. Step S3: Calculate the sample entropy of each IMF component and select the effective IMF component based on the sample entropy size. Step S4: Reconstruct the effective IMF components to obtain the denoised pipeline leakage signal.
[0051] In one embodiment, a signal denoising apparatus based on an improved SSA algorithm to optimize ICEEMD decomposition includes: The acquisition module is used to acquire raw leakage signals from the pipeline. The decomposition module is used to decompose the original leakage signal using ICEEMD to obtain multiple IMF components. During the decomposition process, the SSA algorithm optimized by adaptive Tent chaotic mapping based on population diversity is used to optimize the number of integrations and the noise standard deviation ratio. The calculation and selection module is used to calculate the sample entropy of each IMF component and select the valid IMF components based on the sample entropy magnitude. The reconstruction module is used to reconstruct the effective IMF components to obtain the denoised pipeline leakage signal.
[0052] In one embodiment, a device includes: Memory, used to store computer programs; A processor is configured to execute the computer program, which, when executed by the processor, implements the steps of the signal denoising method based on the improved SSA algorithm for optimizing ICEEMD decomposition as described in the above embodiments.
[0053] In one embodiment, a readable storage medium stores a computer program that, when executed by a processor, implements the steps of the signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition as described in the above embodiments.
[0054] In one embodiment, a computer program product includes a computer program that, when executed by a processor, implements the steps of the signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition as described in the above embodiments.
[0055] The solutions involved in the above embodiments will be described in detail below with reference to specific examples.
[0056] A signal denoising method based on an improved SSA algorithm to optimize ICEEMD decomposition, the specific steps of which are as follows: 1. Experimental parameters are as follows: The pipeline system in this experiment is a metal gas pipe, Φ50×5mm in diameter, with a total length of 6176cm. The medium is gas, the pressure is 0.2MPa, and the flow velocity is 1500m / s. Figure 2 As shown; the pipeline is equipped with a ball valve, pressure sensor, turbine flow meter, and uses a CASI-NGP-A infrasonic sensor with a leakage orifice diameter of 2mm and an initial pressure of 0.2MPa; there are a total of 5 leakage points upstream and downstream, see... Figure 3 Connect the digital acquisition instrument 02528 to the sensor placement port 1 as an upstream sensor, and connect the digital acquisition instrument 02536 to the sensor placement port 3 as a downstream sensor. Use the leak port 2 as the experimental leak point. Connect the digital network transmission instrument via cable, and then transmit the data to the analysis software on the PC via network cable.
[0057] 2. The signal acquired by the upstream sensor is used as the input signal s(n). The filtered signal is x(n). Gaussian white noise is added, and ICEEMD decomposition is performed to obtain the expression of the k-th IMF component. The result of this expression is affected by the number of integrations NE and the noise standard deviation Nstd. Traditionally, the values of NE and Nstd are selected based on experience. Improper selection will lead to poor signal decomposition results. Therefore, this embodiment uses the optimized SSA algorithm to search for the optimal values of NE and Nstd, which are defined as N... 、Nst .
[0058] The optimized SSA algorithm has a population size of 30 and a sampling frequency of 40. The optimization iteration graph of the optimized SSA algorithm is shown below. Figure 4 As shown in the figure, with the increase of the number of iterations, the iteration curve quickly escapes the local optimum after reaching two local optima and arrives at the global optimum. The fitness value decreases rapidly, and convergence is achieved quickly. The optimized SSA algorithm yields an optimal fitness value of 6.2963, and the optimal N... For 87, Nst The value is 0.255. Using these two optimal values, ICEEMD decomposition is performed, and the time-domain and frequency-domain plots of the IMF components are shown below. Figure 5 , Figure 6 As shown in the figure, ICEEMD decomposition decomposes the signal into 8 IMF components, the last of which is the residual component.
[0059] 3. Calculate the sample entropy for each IMF component, as shown in Table 1:
[0060] Table 1. Sample entropy of each IMF component As shown in the table above, the sample entropy of IMF4-6 is significantly higher than the others. Therefore, IMF4-6 were chosen for signal reconstruction. The signals before and after reconstruction are as follows: Figure 7 As shown, Figure 7 In the image, the upper signal is the original signal, and the lower signal is the reconstructed signal. It is easy to see that, after the improvement, the waveform of the reconstructed signal is smoother than that of the original signal, indicating that some noise has been removed.
[0061] 4. The SNR and MSE of the signal before and after the SSA algorithm improvement are shown in Table 2:
[0062] The table above shows that before the improvement refers to using the conventional SSA algorithm to optimize the ICEEMD parameters, while after the improvement refers to using the optimized SSA algorithm in this embodiment to optimize the ICEEMD parameters.
[0063] As shown in the table above, after the improvement in this embodiment, the signal-to-noise ratio increased by 1.528 and the mean square error decreased by 0.346, indicating that the optimized ICEEMD decomposition after the improvement of SSA can achieve a good denoising effect and eliminate the interference caused by the need to manually set parameters.
[0064] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.
Claims
1. A signal denoising method based on an improved SSA algorithm to optimize ICEEMD decomposition, characterized in that, include: Collect raw leak signals from the pipeline; The original leakage signal was decomposed using ICEEMD to obtain multiple IMF components. During the decomposition process, the SSA algorithm optimized by adaptive Tent chaotic mapping based on population diversity was used to optimize the number of integrations and the noise standard deviation ratio. Calculate the sample entropy of each IMF component, and select the effective IMF components based on the sample entropy magnitude; The effective IMF components are reconstructed to obtain the denoised pipeline leakage signal.
2. The signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition according to claim 1, characterized in that, During the decomposition of the original leakage signal using ICEEMD, the g-1th residual component is used... Construct the g-th signal to be decomposed, and based on the g-th signal to be decomposed... The g-th intrinsic mode component is obtained. Specifically, it is expressed as: In the formula, g = 3, ..., N N This indicates the number of ensemble iterations performed using the SSA algorithm optimized with an adaptive Tent chaotic map based on population diversity. This represents the (g-1)th noise figure. This represents the ratio of noise standard deviations obtained using the SSA algorithm optimized with an adaptive Tent chaotic map based on population diversity. This indicates the calculation of the standard deviation. This represents the g-th signal to be decomposed; This indicates the calculation of the local mean of the signal, where H represents the number of iterations.
3. The signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition according to claim 1, characterized in that, The SSA algorithm is optimized using an adaptive Tent chaotic map based on population diversity; specifically: Step a, define the adaptive Tent chaotic mapping parameters as follows: in, Let the chaotic mapping parameters of the t-th iteration satisfy the following condition: D(t) is the normalized population diversity index for the t-th iteration, calculated using the following formula: Where N is the number of individual sparrows, and L is the length of the diagonal of the solution space. Let be the mean of the j-th dimension variable in the t-th generation population, and d be the optimization dimension. Let be the parameter value of the i-th sparrow individual in the j-th generation of the population; Step b, generate an adaptive Tent chaotic sequence based on the adaptive Tent chaotic mapping parameters, as detailed in the table. As shown: in, for Interval chaotic sequence, initial value Generate N chaotic sequences The value, N, represents the number of individual sparrows; Step c, the chaotic values are mapped to the parameter space, represented as: in, This represents the k-th chaotic value. This represents the parameter position of the k-th individual sparrow.
4. The signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition according to claim 3, characterized in that, During the decomposition process, the optimized SSA algorithm is used to optimize the ensemble number and the noise standard deviation ratio; specifically, this includes: Step A: Determine the noise standard deviation ratio Number of integrations ; Step B, determine the objective function as: In the formula, Fitness is the fitness function, and SNR represents the signal-to-noise ratio. These are the weighting coefficients; The energy entropy of the IMF component after ICEEMD decomposition; Step C introduces adaptive chaotic perturbation for sparrow position updates, specifically including discoverer position updates, follower position updates, and vigilant position updates; Step D, Termination condition: Repeat step C until the maximum number of iterations is reached; Step E: After the iteration is complete, select the individual with the highest fitness. The optimal noise standard deviation ratio and optimal number of integrations Substitute this into the ICEEMD decomposition process.
5. The signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition according to claim 4, characterized in that, During the sparrow location update process, The formula for updating the discoverer's location is: In the formula, This represents the j-th dimension position of the i-th sparrow in the (t+1)-th generation; This represents the j-th position of the i-th sparrow in the t-th generation; Random values between For adaptive Tent chaotic sequence values; T is the maximum number of iterations, and Q is a random number that follows a normal distribution. Given a 1×d dimensional matrix where each element is 1, , where is a random number, and ST is the safety threshold; The formula for updating the vigilant's position is: In the formula, It is the globally optimal position in the j-th dimension when the current iteration number t is reached, and β and The step size control parameter is a random number that follows a standard normal distribution and is called the step size adjustment factor. N is the number of sparrows. The coefficient of intensity of chaotic disturbance. The j-th dimension sparrow position is randomly selected in the t-th generation. This represents the optimal position of the sparrow in the j-th dimension within the (t+1)-th generation population.
6. The signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition according to claim 1, characterized in that, Before using ICEEMD to decompose the original leakage signal, the following steps are also included: The collected raw pipeline leakage signals are filtered.
7. A signal denoising device based on an improved SSA algorithm to optimize ICEEMD decomposition, characterized in that, include: The acquisition module is used to acquire raw leakage signals from the pipeline. The decomposition module is used to decompose the original leakage signal using ICEEMD to obtain multiple IMF components. During the decomposition process, the SSA algorithm optimized by adaptive Tent chaotic mapping based on population diversity is used to optimize the number of integrations and the noise standard deviation ratio. The calculation and selection module is used to calculate the sample entropy of each IMF component and select the valid IMF components based on the sample entropy magnitude. The reconstruction module is used to reconstruct the effective IMF components to obtain the denoised pipeline leakage signal.
8. A device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program, wherein the computer program, when executed by the processor, implements the steps of the signal denoising method based on the improved SSA algorithm for optimizing ICEEMD decomposition as described in any one of claims 1-6.
9. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition as described in any one of claims 1-6.
10. A computer program product, comprising a computer program, characterized in that, When a computer program is executed by a processor, it implements the steps of the signal denoising method based on the improved SSA algorithm to optimize ICEEMD decomposition as described in any one of claims 1-6.
Citation Information
Patent Citations
Shock wave signal resonance noise reduction method for optimizing VMD based on sparrow search algorithm
CN118013262A