Grounding grid corrosion prediction method based on fruit fly algorithm and grey model

Abstract

The present invention relates to the technical field of power grounding corrosion detection, and in particular to a grounding grid corrosion prediction method based on a fruit fly algorithm and a gray model. The method comprises the following steps: obtaining the grounding resistance of the grounding grid according to a fixed time series; processing the grounding resistance using a gray model to establish a prediction model for grounding grid corrosion diagnosis; introducing a logistic chaos map to obtain an optimized fruit fly algorithm; optimizing the background value of the prediction model for grounding grid corrosion diagnosis using the optimized fruit fly algorithm to obtain an optimized prediction model; updating the grounding resistance based on the optimized prediction model, and predicting subsequent changes in the grounding resistance to obtain the corrosion rate of the grounding grid. The method provides a more accurate prediction of the corrosion status of the grounding grid, effectively resolving the problem of poor prediction accuracy of the grounding grid corrosion status.

Description

Technical Field

[0001] The present invention relates to the technical field of power grounding corrosion detection, and in particular to a grounding grid corrosion prediction method based on a fruit fly algorithm and a grey model. Background Art

[0002] The grounding grid is a large-scale concealed project in the power system. It protects and shields power facilities by burying a large number of metal conductors deep in the soil. However, due to the complex environment in the soil, the grounding grid is affected by various salts, water, oxygen, microorganisms and the current flowing during operation, which gradually corrodes the grounding performance, deteriorates, and eventually loses the grounding function, posing a hidden danger to the safe and stable operation of the power system.

[0003] Regulations require regular inspections of the grounding grid's grounding resistance to assess its electrical performance. However, inspection results do not accurately reflect the corrosion status of the grounding grid. This is because the grounding resistance may remain normal even when the grounding grid conductor is corroded or even broken. Waiting until the grounding grid's grounding resistance is found to be substandard or has caused an accident before conducting large-scale excavations to locate broken points and corroded sections is both blind and labor-intensive.

[0004] To address this issue, the current method for predicting the corrosion rate of the grounding grid is to establish a corrosion parameter sample library. This method uses soil parameters collected on-site as input parameters of a grey model to establish a grounding grid prediction model. This method requires less sample data and can also roughly predict the grounding grid corrosion situation even when some data samples are missing or there is uncertainty among the grounding grid corrosion factors, which can greatly reduce the prediction cost. However, in actual applications, due to the low detection accuracy of the grey model itself, the prediction accuracy of the grounding grid corrosion status is still poor. Summary of the Invention

[0005] The present invention provides a grounding grid corrosion prediction method based on a fruit fly algorithm and a grey model, which is used to solve the problem of poor accuracy in predicting the corrosion state of the grounding grid.

[0006] The present invention provides a grounding grid corrosion prediction method based on a fruit fly algorithm and a grey model, comprising the following steps:

[0007] S1: Obtain the grounding resistance of the grounding grid according to a fixed time series;

[0008] S2: Using a grey model to process the grounding resistance, and establishing a prediction model for grounding grid corrosion diagnosis;

[0009] S3: Introducing the Logistic Chaos Map to generate the initial position of the fruit fly population, redesigning the dynamic search radius and expanding the search range to obtain the optimized fruit fly algorithm;

[0010] S4: optimizing the background value of the prediction model for grounding grid corrosion diagnosis using an optimized fruit fly algorithm to obtain an optimized prediction model;

[0011] S5: The grounding resistance is updated according to the optimized prediction model, and subsequent changes in the grounding resistance are predicted to obtain the corrosion rate of the grounding grid.

[0012] Specifically, in step S2, the grounding resistance is processed using a grey model to establish a prediction model for grounding grid corrosion diagnosis, specifically:

[0013] S21: Introduce the grey model in grey theory and construct the original sequence;

[0014] S22: performing cumulative calculation according to the original sequence to obtain a cumulative sequence;

[0015] S23: Calculating the adjacent mean according to the accumulated sequence to obtain an adjacent mean sequence;

[0016] S24: establishing a whitened differential equation according to the original sequence, and solving the whitened differential equation to obtain a prediction model of the accumulated sequence;

[0017] S25: Obtaining a time response sequence of a grey model according to the prediction model of the cumulative sequence;

[0018] S26: The time response sequence of the grey model is cumulatively reduced to obtain a prediction model for grounding grid corrosion diagnosis.

[0019] Specifically, in step S3, the Logistic chaotic map is introduced to generate the initial position of the fruit fly population, the dynamic search radius is redesigned, and the search range is expanded to obtain the optimized fruit fly algorithm, which is specifically as follows:

[0020] S31: Introduce the Logistic chaotic map to initialize the positions of all fruit flies and generate the initial position of the fruit fly population;

[0021] S32: Introduce dynamic search radius strategy and redesign the dynamic search radius generation method;

[0022] S33: The odor concentration determination formula of the fruit fly algorithm is improved by the sgn function, and the search range is expanded by the improved odor concentration determination formula of the fruit fly algorithm to obtain the optimized fruit fly algorithm.

[0023] Specifically, in step S4, the background value of the prediction model for grounding grid corrosion diagnosis is optimized using the optimized fruit fly algorithm to obtain an optimized prediction model, specifically:

[0024] The fixed value in the background value of the grey model is replaced by the dynamically generated coefficient to obtain the optimized grey model.

[0025] Specifically, in step S5, the grounding resistance is updated according to the optimized prediction model, and subsequent changes in the grounding resistance are predicted to obtain the corrosion rate of the grounding grid, specifically:

[0026] S51: Perform exponential transformation on the data in the original sequence to obtain a new original sequence;

[0027] S52: Substitute the new original sequence into the optimized grey model to obtain a new response sequence;

[0028] S53: perform cumulative reduction on the new response sequence;

[0029] S54: Obtaining a restored value of the original sequence according to the restored new response sequence and the new original sequence.

[0030] Specifically, step S1 further includes: acquiring the electromagnetic field, contact potential, and step potential of the buried position of the grounding grid according to a fixed time sequence.

[0031] The beneficial effect of the present invention is that the present invention provides a grounding grid corrosion prediction method based on a fruit fly algorithm and a gray model, comprising the following steps: S1: obtaining the grounding resistance of the grounding grid according to a fixed time series; S2: using a gray model to process the grounding resistance and establish a prediction model for grounding grid corrosion diagnosis; S3: introducing a logistic chaotic map to generate an initial position of a fruit fly population, redesigning a dynamic search radius and expanding the search range to obtain an optimized fruit fly algorithm; S4: using the optimized fruit fly algorithm to optimize the background value of the prediction model for grounding grid corrosion diagnosis to obtain an optimized prediction model; S5: updating the grounding resistance according to the optimized prediction model, and predicting subsequent changes in the grounding resistance to obtain the corrosion rate of the grounding grid.

[0032] The present invention can predict the corrosion trend of the grounding grid by introducing a grey model into the grounding resistance of a fixed time series, and then optimize the fruit fly algorithm by introducing logistic chaos mapping to improve the convergence accuracy of the traditional fruit fly algorithm and expand the search range. The optimized fruit fly algorithm improves the accuracy and stability of the grey model, and obtains a more accurate corrosion state of the grounding grid, effectively solving the problem of poor prediction accuracy of the corrosion state of the grounding grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 This is a flow chart of the grounding grid corrosion prediction method based on the fruit fly algorithm and grey model; DETAILED DESCRIPTION

[0035] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0036] The present invention provides a grounding grid corrosion prediction method based on the fruit fly algorithm and the grey model, which specifically comprises the following steps:

[0037] S1: Obtain the grounding resistance u1(i) of the grounding grid according to a fixed time series;

[0038] Where: i is the preset time series, i = 1, 2, 3...n, the unit is week;

[0039] S2: Establish a prediction model for grounding grid corrosion diagnosis based on the grey model GM(1,1);

[0040] S3: The optimized fruit fly algorithm is obtained by introducing the Logistic chaos map to improve the convergence accuracy of the traditional fruit fly algorithm and expand the search range;

[0041] S4: Using the optimized fruit fly algorithm to optimize the background value of the grey model GM (1, 1) to obtain the optimized grey model;

[0042] S5: Update the value of the grounding resistance u1(i) according to the optimized grey model, deduce the n+1 sequence, predict the subsequent change of the grounding resistance value, and obtain the grounding grid corrosion rate.

[0043] The present invention also provides a specific embodiment. Based on the above embodiment, step S2 includes:

[0044] S21: Introduce the grey model GM(1,1) in grey theory and construct the original sequence U (0) ;

[0045] S22: Substitute the ground resistance u1(i) obtained in step S1 into the original sequence U (0) , then:

[0046] U (0) =(u1 (0) (1), u1 (0) (2),…,u1 (0) (n))

[0047] Where: u (0) (k)≥0, and the corresponding time series is t={t1, t2,…, t n}, k is an ordinal number, k = 1, 2, ..., n;

[0048] S23: According to the original sequence U (0) Perform cumulative calculation to obtain the cumulative sequence U (1) , then:

[0049] U( 1 )=(u1 (1) (1), u1( 1 )(2),…,u1( 1 )(n))

[0050] The cumulative formula is:

[0051]

[0052] Where: k = 1, 2, ..., n;

[0053] S24: According to the cumulative sequence U (1) Get the adjacent mean sequence V (1) , then:

[0054] V (1) =v1 (1) (2), v1 (1) (3),…,v1 (1) (n))

[0055] The calculation formula for the adjacent mean is:

[0056] v (1) (k)=[u (1) (k)+u (1) (k-1)] / 2;

[0057] Where: k = 2, 3, ..., n;

[0058] S25: Accumulate sequence U (1) and the adjacent mean sequence V (1) The grey differential equation for establishing the GM(1,1) model is:

[0059] u1 (0) (k)+av1 (1) (k) = b

[0060] S26: Based on the grey differential equation of the GM(1,1) model, a whitening differential equation is established and solved. The formula is as follows:

[0061] du1 (1) / dt+au1 (1) =b

[0062] In the formula, a is the development coefficient, b is the grey action;

[0063] The values ​​of a and b are obtained by the least square method, that is, The least squares method can be used to estimate:

[0064]

[0065] Where:

[0066] The prediction model of the cumulative sequence is solved as follows:

[0067]

[0068] S27: The time response sequence of the GM(1,1) model is obtained based on the prediction model of the cumulative sequence as follows:

[0069]

[0070] S28: The time response sequence of the GM (1, 1) model is cumulatively reduced to obtain the grounding grid corrosion prediction model, which is as follows:

[0071]

[0072] The above formula is the original sequence U (0) The prediction model for grounding grid corrosion diagnosis obtained using the GM(1,1) model.

[0073] In a more specific embodiment of the present invention, step S3 specifically includes the following steps:

[0074] S31: By introducing the Logistic Chaos Map to generate the initial position of the fruit fly population, the problem of uneven distribution of the randomly initialized population is solved, the diversity and uniformity of the search population are improved, and the stability of the algorithm is enhanced;

[0075] The simplest and most effective chaos mapping solution, Logistic chaos mapping, is used to initialize the positions of all fruit flies, so that the optimized positions of fruit flies have the characteristics of randomness, ergodicity and regularity of chaos phenomena. The system equation is as follows:

[0076] u1(n+1)=u1(n)(1-u1(n))u1(n)

[0077] u1(n+1)∈[0,1]

[0078] Where: n is the number of iterations; μ is the control parameter;

[0079] When 3.5699456<μ≤4, the Logistic map is in a chaotic state. When μ is closer to 4, it presents a pseudo-random distribution. When μ is in other ranges, the function will eventually converge to a certain value. Here, μ=4 is selected to make the system in a chaotic state. The transformation formula of the chaotic variable Cu1 is as follows:

[0080] Cu1(n+1) i =4Cu1(n) i (1-Cu1(n) i )

[0081] Where: i = 1, 2, ..., n;

[0082] Where: Cu1(n) is the size of the i-th chaotic variable after the n-th chaotic transformation. The chaotic variable Cu1(n+1) after the n+1-th transformation is obtained by the above formula: i Then, according to the original data u1 before chaotic mapping, the maximum development coefficient a max , minimum development coefficient a min With variable Cx i ∈[0,1] repeatedly performs chaotic mapping, the formula is as follows:

[0083]

[0084]

[0085] Where: is the i-th original data before chaotic mapping, and is the value after chaotic mapping;

[0086] S32: Introduce a dynamic search radius strategy and redesign the dynamic search radius generation method to achieve adaptive dynamic adjustment of the algorithm's search radius, improve the algorithm's convergence accuracy and reduce the convergence time. The formula is as follows;

[0087]

[0088] In the above formula, rmax and r min Represent the maximum and minimum search radius respectively, iter represents the current number of iterations, and Miter represents the maximum number of iterations;

[0089] It can be understood that the above algorithm solves the shortcomings of the fixed radius, making the algorithm have a larger search radius in the early stage, enhancing the global search ability of the algorithm, avoiding local optimality, and the search radius decreases with the increase of the number of iterations, thereby improving the overall convergence accuracy of the algorithm and reducing the convergence time;

[0090] S33: Improve the odor concentration determination formula of the fruit fly algorithm to avoid the influence of the odor concentration determination value being always positive on the algorithm, expand the search range of the algorithm, and enhance the adaptability of the algorithm. The odor concentration determination optimization formula is as follows:

[0091]

[0092] Among them, the exponential function ensures the negative correlation between the candidate solution and the position of the fruit fly. At the same time, the sgn function is used. When the fruit fly individual is in the second or fourth quadrant of the two-dimensional coordinate, Si is a negative value;

[0093] The above formula can realize a comprehensive search of the negative value space, expand the application scenarios of the algorithm, and improve the algorithm's ability to solve high-dimensional and complex problems.

[0094] In a more specific embodiment of the present invention, step S4 is specifically as follows:

[0095] The dynamic generation coefficient α(i) is used to replace the fixed value in the background value of the grey model GM(1,1), and the optimized fruit fly algorithm is used to find the optimal value. By dynamically adjusting the generation coefficient α(i) of each interval, the background value error is minimized. The new background value construction formula is:

[0096]

[0097] In the above formula, α(i) is the dynamic generation coefficient, 0≤α(i)≤1, i=1, 2,…, n-1, k=

[0098] 2, 3,…, n.

[0099] It should be noted that the actual background value should be The integral on the interval [k-1, k] is used, while the background value of the traditional modeling method is replaced by the trapezoidal area. When processing some drastically changing data, the traditional background value construction method will bring large errors, resulting in a decrease in the accuracy of the model. Therefore, a dynamic generation coefficient α(i) is used instead of a fixed value.

[0100] It is understandable that the optimization of the background value of the grey model requires dynamically adjusting the generation coefficients of each interval to minimize the background value error, and there are many parameters for solving the dynamic coefficients, which are difficult to solve by general methods; the global optimization ability of the fruit fly algorithm can obtain the global optimal solution of the parameters, so the optimized fruit fly algorithm is used to solve the dynamic generation coefficients of the grey model; after optimizing the background value of the grey model, the optimized fruit fly algorithm can reduce the impact of drastic changes in data on the model accuracy, thereby improving the prediction accuracy and prediction efficiency of the grey model.

[0101] In a more specific embodiment of the present invention, step S5 is specifically as follows:

[0102] S51: For the original sequence U (0) Data in u1 (0) (k) Perform exponential transformation to obtain the new original sequence, and the formula is as follows

[0103]

[0104] Where: i = k = 1, 2, ..., n; c is a constant base number that can be set arbitrarily according to actual conditions;

[0105] S52: Substitute the new original sequence into the optimized grey model GM(1,1) to obtain a new response sequence:

[0106]

[0107] S53: Perform cumulative reduction on the new response sequence. The formula is as follows:

[0108]

[0109] S54: According to The restored value of the original sequence is:

[0110]

[0111] The restored value Substitute the original sequence as a new data set into U (0) The ground resistance value is predicted and the new U (0) The sequence is predicted using the improved GM(1,1) model, and the sequence after n+1 is derived from this sequence to predict the ground resistance value. The prediction steps for the grounding grid electromagnetic field, contact potential, and step potential are the same as those for the ground resistance value.

[0112] In a more preferred embodiment of the present invention, step S1 further comprises: acquiring the electromagnetic field u2(i), the contact potential u3(i), and the step potential u4(i) at the buried position of the grounding grid according to a fixed time sequence;

[0113] A prediction model for grounding grid corrosion diagnosis is established based on the grey model GM(1,1);

[0114] The Logistic chaotic map is introduced to improve the convergence accuracy of the traditional fruit fly algorithm and expand the search range to obtain the optimized fruit fly algorithm;

[0115] The optimized fruit fly algorithm is used to optimize the background value of the grey model GM (1, 1) to obtain the optimized grey model;

[0116] The electromagnetic field u2(i), contact potential u3(i), and step potential u4(i) at the buried location of the grounding grid are updated according to the optimized grey model. The n+1 sequence is deduced to predict the subsequent changes in the grounding resistance value and obtain the grounding grid corrosion rate.

[0117] It can be understood that by obtaining the electromagnetic field u2(i), contact potential u3(i), and step potential u4(i) at the buried position of the grounding grid according to a fixed time series, the n+1 sequence of the electromagnetic field u2(i), contact potential u3(i), and step potential u4(i) data at the buried position of the grounding grid can be obtained respectively, thereby more accurately inferring the corrosion state of the grounding grid.

[0118] In a more specific embodiment of the present invention, each data in the sequence after n+1 of the grounding resistance u1(i) of the grounding grid, the electromagnetic field u2(i) at the buried position of the grounding grid, the contact potential u3(i), and the step potential u4(i) is arithmetic averaged to infer the corrosion state of the grounding grid.

[0119] By measuring the grounding resistance, electromagnetic field, contact potential, and step potential of the grounding grid, the current corrosion status of the grounding grid can be determined through analysis. Substituting the measured data into the improved grey model, the grounding resistance, electromagnetic field, contact potential, and step potential can be predicted, thereby achieving the goal of predicting grounding grid corrosion.

[0120] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0121] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0122] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0123] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0124] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

Claims

1. A grounding grid corrosion prediction method based on fruit fly algorithm and grey model, characterized in that: The following steps are involved: S1: Obtain the grounding resistance of the grounding grid according to a fixed time series; S2: Using a grey model to process the grounding resistance, and establishing a prediction model for grounding grid corrosion diagnosis; S3: Introducing the Logistic Chaos Map to generate the initial position of the fruit fly population, redesigning the dynamic search radius and expanding the search range to obtain the optimized fruit fly algorithm; S4: optimizing the background value of the prediction model for grounding grid corrosion diagnosis using an optimized fruit fly algorithm to obtain an optimized prediction model; S5: updating the grounding resistance according to the optimized prediction model, and predicting the subsequent change of the grounding resistance to obtain the corrosion rate of the grounding grid; In step S2, the grounding resistance is processed using a grey model to establish a prediction model for grounding grid corrosion diagnosis, specifically: S21: Introduce the grey model in grey theory and construct the original sequence; S22: performing cumulative calculation according to the original sequence to obtain a cumulative sequence; S23: Calculating the adjacent mean according to the accumulated sequence to obtain an adjacent mean sequence; S24: establishing a whitened differential equation according to the original sequence, and solving the whitened differential equation to obtain a prediction model of the accumulated sequence; S25: Obtaining a time response sequence of a grey model according to the prediction model of the cumulative sequence; S26: Perform cumulative reduction on the time response sequence of the grey model to obtain a prediction model for grounding grid corrosion diagnosis; In step S3, the Logistic chaotic map is introduced to generate the initial position of the fruit fly population, the dynamic search radius is redesigned, and the search range is expanded to obtain the optimized fruit fly algorithm, specifically: S31: Introduce the Logistic chaotic map to initialize the positions of all fruit flies and generate the initial position of the fruit fly population; S32: Introduce dynamic search radius strategy and redesign the dynamic search radius generation method; S33: Pass The function improves the determination formula of the odor concentration of the fruit fly algorithm, and expands the search range through the improved determination formula of the odor concentration of the fruit fly algorithm to obtain the optimized fruit fly algorithm; In step S4, the optimized fruit fly algorithm is used to optimize the background value of the prediction model for grounding grid corrosion diagnosis to obtain an optimized prediction model, specifically: The fixed value in the background value of the grey model is replaced by the dynamically generated coefficient to obtain the optimized grey model; In step S5, the grounding resistance is updated according to the optimized prediction model, and subsequent changes in the grounding resistance are predicted to obtain the corrosion rate of the grounding grid, specifically: S51: Perform exponential transformation on the data in the original sequence to obtain a new original sequence; S52: Substitute the new original sequence into the optimized grey model to obtain a new response sequence; S53: perform cumulative reduction on the new response sequence; S54: Obtaining a restored value of the original sequence according to the restored new response sequence and the new original sequence.

2. The method for predicting grounding grid corrosion based on the fruit fly algorithm and the grey model according to claim 1, characterized in that: Step S1 also includes: acquiring the electromagnetic field, contact potential, and step potential of the buried position of the grounding grid according to a fixed time sequence.

Images