Automatic design method of excitation coil of sleeve type magnetic flux sensor

Abstract

The invention relates to the technical field of magnetic flux sensors, and discloses an automatic design method for an excitation coil of a sleeve type magnetic flux sensor, and the method comprises the steps: designing the excitation coil in the sleeve type magnetic flux sensor through employing an improved myxomycete algorithm; the iterative process of the improved myxomycete algorithm comprises the following steps: S1, sorting; s2, updating the weight; s3, the position is updated, an iterative formula 8 is replaced in the step S3, and after the second-dimensional data of the colistis is updated each time, namely after the layer number m of the excitation coil is updated, rounding-off and rounding-off are conducted on the layer number m of the excitation coil immediately; and after the third-dimensional data of the myxobacteria is updated each time, namely after the diameter phi of the copper wire of the exciting coil is updated, the upper and lower limit range of the current I is immediately modified to be [0, 1.5 pi phi 2 * 106]. According to the method, the design parameters of the exciting coil are rapidly generated for the steel inhaul cables with different diameters, the exciting coil designed according to the method ensures that the inhaul cables are saturated and uniformly magnetized, meanwhile, the used copper wire material is as few as possible, and then the cost of the sensor is reduced.

Description

Technical Field

[0001] The present invention relates to the field of magnetic flux sensors, and in particular to an automated design method for an excitation coil of a sleeve-type magnetic flux sensor. Background Art

[0002] A magnetic flux sensor is a high-performance measurement device based on the magnetoelastic effect. It uses the property of ferromagnetic materials that change in internal magnetic permeability when subjected to external loads to measure cable tension. Structurally, a magnetic flux sensor typically consists of an excitation coil and an induction coil. A pulsed excitation signal is applied to the excitation coil to generate a time-varying magnetic field in the component being measured. The corresponding induced voltage is then measured in the induction coil. The relative magnetic permeability of the component is then integrated and converted to the measured value. This relationship between the change in magnetic permeability and stress is then established, enabling accurate measurement of cable tension. Magnetic flux sensors are widely used in health monitoring of large infrastructure structures such as bridges and buildings. They provide an effective technical means for testing cable tension in various structures, including cable-stayed bridges, suspenders and tie rods in tied arch bridges, and cable systems in suspension bridges. The most widely used magnetic flux sensor in current engineering practice is the sleeve-type magnetic flux sensor. In sleeve-type magnetic flux sensors, the excitation and induction coils are coaxially wound. This design not only ensures uniform magnetic field distribution and improves measurement accuracy, but also makes the sensor compact and easy to install and maintain. In practical applications, sleeve-type magnetic flux sensors can accurately measure changes in cable tension in structures such as steel cables, providing reliable technical support for health monitoring of large-scale infrastructure such as bridges and buildings. Furthermore, with the continuous advancement of materials science and manufacturing processes, the performance of sleeve-type magnetic flux sensors is also continuously improving, and their application scope is also expanding.

[0003] In a sleeve-type magnetic flux sensor, the design of the excitation coil is crucial. As the source of the magnetic field, the excitation coil requires careful design to ensure uniformity and strength. Typically, the excitation coil is tightly wound to reduce leakage magnetic fields and interference. To further improve magnetic field uniformity, the excitation coil can be designed with multiple turns, each closely spaced. Specifically, the excitation coil parameters include coil length, number of coil layers, copper wire diameter, and the current applied to the excitation coil.

[0004] Currently, the design of magnetic flux sensor excitation coils mainly relies on experimental trials, resulting in a slow R&D cycle and difficulty in ensuring high performance (the measured cable can be saturated and uniformly magnetized) and low cost (less copper wire is used) of the designed excitation coils. Summary of the Invention

[0005] To overcome or alleviate one or more of the above technical problems, the present invention aims to provide an automated design method for the excitation coil of a sleeve-type magnetic flux sensor. This method utilizes an improved iterative calculation method called a slime mold algorithm to quickly and accurately generate the design parameters of the excitation coil. The present invention provides the following technical solutions:

[0006] The present invention provides an automated design method for an excitation coil of a sleeve-type magnetic flux sensor. The method uses an improved slime mold algorithm to design the excitation coil in the sleeve-type magnetic flux sensor. Each slime mold individual X contains four-dimensional data: the length L of the excitation coil, the number of layers m of the excitation coil, the diameter Φ of the excitation coil copper wire, and the current I required to be applied to the excitation coil. First, the four-dimensional data of each slime mold X is randomly initialized. The iterative process of the improved slime mold algorithm includes the following steps:

[0007] S1 sorting, calculate the fitness value F(X) of each slime mold individual X, and sort the slime mold individuals from best to worst according to their fitness values. The smaller the fitness value, the better the slime mold individual. Where F is the fitness function;

[0008] S2 updates the weights. Different slime mold individuals use different update formulas when updating weights. The weights of the better 1 / 2 of the slime mold group are updated according to the following formula 1, and the weights of the other 1 / 2 of the slime mold group are updated according to the following formula 2:

[0009]

[0010] Where rand is a random number between [0,1]; X i is the slime mold individual that the algorithm is currently updating; X best is the optimal individual with the best fitness value in the slime mold population; X worst is the individual with the worst fitness value in the slime mold population; ε is a small positive number; w d new,i Slime Mold X i The d-th dimension weight of

[0011] S3 updates the position, and the slime mold individual uses the weight to update its position. The position update formula is as follows:

[0012]

[0013] p i =tanh(|F(X best )-F(X i )|) (5)

[0014]

[0015] Among them, iter maxis the maximum number of iterations set, iter is the number of iterations currently performed; x d max The maximum value of the d-th dimension data set by the algorithm, x d min is the minimum value of the d-th dimension data set by the algorithm; r1 and r2 are two randomly selected slime mold individuals, w d r1 is the d-th dimension weight of slime mold individual r1, x d r1 is the d-dimensional data of slime mold individual r1, x d r2 is the d-th dimension data of the slime mold individual r2; rand1 and rand2 are random numbers in [0,1], Z is a constant in [0,1]; randA is a random number in [-a,a], and randB is a random number in [-b,b]; x d best is the slime mold individual X with the best fitness value in the slime mold population best The d-th dimension data, x d i is the currently updated slime mold individual X i The d-th dimension data before updating, x d new,i Update the current slime mold individual X i Updated d-th dimension data;

[0016] In step S3, each time the second dimension data of the slime mold individual is updated, that is, each time the number of excitation coil layers m is updated, the number of excitation coil layers is immediately rounded to an integer;

[0017] In step S3, each time the third dimension data of the slime mold individual is updated, that is, after the diameter Φ of the excitation coil copper wire is updated, the upper and lower limits of the fourth dimension data current I of the slime mold are immediately modified to [0,1.5πΦ 2 *10 6 ].

[0018] Preferably, the improved slime mold algorithm further includes the following process before step S1:

[0019] The entire slime mold population is assumed to contain a specific number of slime molds, with the specific number ranging from [10, 1000], and the number of algorithm iterations ranging from [100, 10000]. Upper and lower limits are set for each dimension of the slime mold population data. The excitation coil length L ranges from [0.15, 0.5], in meters. The number of excitation coil layers m ranges from [2, 25]. The excitation coil copper wire diameter Φ ranges from [0.0002, 0.002], in meters. The initial upper and lower limits of the excitation coil current I range from [0, 3], in amperes.

[0020] At the beginning of the improved slime mold algorithm, data is randomly initialized for each slime mold individual X in the slime mold population within the upper and lower limits of each dimension of the slime mold population.

[0021] Preferably, the fitness function F in step S1 is used to evaluate the quality of each slime mold individual; the fitness function F is expressed as follows:

[0022] F=0.001F1+20F2 (18)

[0023] F1 is the magnetic field strength adaptability, 0.001 is the set weight, and its specific expression is as follows:

[0024] F1=0,if H c1 ≤≤H(0)≤H C2 (10)

[0025] F1=|H C1 -H(0)|,ifH c1 >H(0) (11)

[0026] F1=|H c2 -H(0)|,ifH(0)>H c2 12)

[0027] Where H(0) is the magnetic field strength at the center of the excitation coil, [H c1 ,H c2 ] is the magnetic field intensity range required by the design to ensure the saturation magnetization of the tested cable; for steel cables, H c1 Set to 16000A / m, H c2 Set to 17000A / m;

[0028] F2 is the magnetic field uniformity adaptability, 20 is the set weight, and its specific expression is as follows:

[0029] F2=0,ifu≤u m (13)

[0030] F2=uu m ,ifu>u m (14)

[0031] in is the uniformity of magnetic field intensity, l is the length of the induction coil, H(0.5l) is the magnetic field intensity at the edge of the induction coil, u m =0.02 is the upper limit of the set magnetic field uniformity.

[0032] Preferably, the fitness function F in step S1 is expressed as

[0033] F=0.001F1+20F2+70F3 (19)

[0034] Where F1 is the magnetic field strength adaptability, 0.001 is the set weight, and its specific expression is as follows:

[0035] F1=0,if H C1 ≤H(0)≤H C2 10)

[0036] F1=|H C1 -H(0)|,if H c1 >H(0) (11)

[0037] F1=|H c2 -H(0)|, if H(0)>H c2 12)

[0038] Where H(0) is the magnetic field strength at the center of the excitation coil, [H c1 ,H c2 ] is the magnetic field intensity range required by the design to ensure the saturation magnetization of the tested cable; for steel cables, H c1 Set to 16000A / m, H c2 Set to 17000A / m;

[0039] F2 is the magnetic field uniformity adaptability, 20 is the set weight, and its specific expression is as follows:

[0040] F2=0,if u≤u m (13)

[0041] F2=uu m , if u>u m (14)

[0042] in is the uniformity of magnetic field intensity, l is the length of the induction coil, H(0.5l) is the magnetic field intensity at the edge of the induction coil, u m =0.02 is the upper limit of the set magnetic field uniformity;

[0043] F3 is the adaptability of the amount of copper wire used in the excitation coil, with a weight of 70. Its expression is as follows:

[0044]

[0045] Where R1 and R2 are the inner and outer radii of the excitation coil respectively, and L is the length of the excitation coil.

[0046] Preferably, the fitness function F in step S1 is expressed as

[0047] F=0.001F1+20F2+70F3+0.001F4 (9)

[0048] Where F1 is the magnetic field strength adaptability, 0.001 is the set weight, and its specific expression is as follows:

[0049] F1=0,ifH C1 ≤H(0)≤H C2 (10)

[0050] F1=|H C1 -H(0)|,ifH C1 >H(0) (11)

[0051] F1=|H C2 -H(0)|,ifH(0)>H C2 (12)

[0052] Where H(0) is the magnetic field strength at the center of the excitation coil, [H c1 ,H c2 ] is the magnetic field intensity range required by the design to ensure the saturation magnetization of the tested cable; for steel cables, H c1 Set to 16000A / m, H c2 Set to 17000A / m;

[0053] F2 is the magnetic field uniformity adaptability, 20 is the set weight, and its specific expression is as follows:

[0054] F2=0,if u≤u m (13)

[0055] F2=uu m ,if u>um (14)

[0056] in is the uniformity of magnetic field intensity, l is the length of the induction coil, H(0.5l) is the magnetic field intensity at the edge of the induction coil, u m =0.02 is the upper limit of the set magnetic field uniformity;

[0057] F3 is the adaptability of the amount of copper wire used in the excitation coil, with a weight of 70, and the expression is as follows:

[0058]

[0059] Where R1 and R2 are the inner and outer radii of the excitation coil, respectively, and L is the length of the excitation coil;

[0060] F4 is the fitness of the number of coil layers, with a weight of 0.001, and the expression is as follows:

[0061] F4=m (16)

[0062] Where m is the number of winding layers of the excitation coil.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] The present invention provides an automated design method for an excitation coil of a sleeve-type magnetic flux sensor. The method rapidly and automatically generates excitation coil design parameters for steel cables of different diameters. The excitation coil designed according to the method ensures that the cable is saturated and uniformly magnetized while using as little copper wire material as possible, thereby reducing sensor costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a schematic structural diagram of a sleeve-type magnetic flux sensor provided by an embodiment of the present invention.

[0066] Figure 2 This is a graph showing the results of running the improved position-updating slime mold algorithm provided by an embodiment of the present invention.

[0067] Figure 3 Graph showing the moderate mean values of the slime mold population according to an embodiment of the present invention. DETAILED DESCRIPTION

[0068] Existing literature, https: / / www.sciencedirect.com / science / article / abs / pii / S0167739X19320941 and https: / / zhuanlan.zhihu.com / p / 712901735, provides a description of the slime mold algorithm, an optimization algorithm based on the vegetative growth process of slime molds. After the mitotically formed amoeba matures, the slime mold enters a vegetative growth phase, forming a network-like structure and varying its surface area based on the nutrients it needs, such as food, water, and oxygen. In the slime mold algorithm, the slime mold determines the weight of each individual's position based on the objective conditions of its current location (the quality of its fitness function). Each individual then determines its new position based on this weight.

[0069] The present invention uses an improved slime mold algorithm to design the excitation coil in a sleeve-type magnetic flux sensor. The design parameters include the length L of the excitation coil, the number of layers m of the excitation coil, the diameter d of the copper wire of the excitation coil, and the current I required to be applied to the excitation coil. First, the existing slime mold algorithm is introduced.

[0070] In the existing slime mold algorithm, each iteration process after initializing the slime mold population is divided into three steps: sorting, updating weights, and updating positions.

[0071] The first step is sorting, that is, sorting the slime mold individuals from best to worst according to their fitness functions. The smaller the fitness value, the better the slime mold individual.

[0072] There is no universal, specific expression for the fitness function. Different problems require different fitness functions, and different people may design different fitness functions for the same problem. In this embodiment, the fitness function is designed as Equation (9). The smaller the fitness value, the better the slime mold individual.

[0073] Step 2: Update the weights

[0074] When updating the weights, different slime mold individuals have different update formulas. The weights of the better 1 / 2 of the slime mold group are updated according to the following formula (1), and the weights of the other 1 / 2 of the slime mold group are updated according to formula (2).

[0075]

[0076] Where rand is a random number between [0, 1]; F is the fitness function; Xi is the slime mold individual of the current iteration; X best is the individual with the best fitness value in the slime mold population; X worst is the individual with the worst fitness in the slime mold population; ε is a small positive number, generally ranging from 1e-4 to 1e-9, to prevent the denominator from reaching zero. It can be seen that the weights of the better individuals are larger, within the range [1, 1.3], with the better individuals having weights closer to 1; the worse individuals have weights within the range [0.7, 1], with the worse individuals having weights closer to 0.7.

[0077] Step 3: Update location

[0078] After updating the weights, the slime mold individuals will use the weights to update their positions. The position update formula is as follows:

[0079]

[0080] p i =tanh(|F(X best )-F(X i )|) (5)

[0081]

[0082] iterma x is the maximum number of iterations set, iter is the number of iterations currently performed; x d is the d-th dimension data of slime mold X, x d max is the maximum value of the d-dimensional data of the slime mold population, x d minis the minimum value of the d-th dimension data of the set slime mold population; r1 and r2 are two randomly selected slime mold individuals, w d r 1 is the d-th dimension weight of slime mold individual r1, x d r1 is the d-dimensional data of slime mold individual r1, x d r2 is the d-th dimension of the slime mold individual r2. rand1 and rand2 are random numbers in the range [0, 1]. Z is a constant in the range [0, 1], typically set to 0.03. randA is a random number in the range [-a, a], and randB is a random number in the range [-b, b].

[0083] Formulas (6), (7), and (8) are specific position update formulas. It can be seen that formula (6) is an operation to obtain a global random position with a low probability, similar to the mutation of a genetic algorithm. Formula (7) searches near the current optimal position. Formula (8) is an operation to allow the individual to converge to 0. This formula needs to be improved in order to allow the individual to converge to 0 so as to obtain good results on the test function whose optimal solution is 0. When the optimal solution is not 0, the effect is not good.

[0084] The above is an introduction to the iterative process of the existing slime mold algorithm.

[0085] The present invention is described in detail below with reference to the embodiments and accompanying drawings. However, it should be understood that the embodiments and accompanying drawings are merely exemplary descriptions of the present invention and do not constitute any limitation on the scope of protection of the present invention. All reasonable variations and combinations within the scope of the inventive concept of the present invention fall within the scope of protection of the present invention.

[0086] Example 1

[0087] This embodiment uses the slime mold algorithm to design a magnetic flux sensor. First, the structure of the magnetic flux sensor is analyzed. Figure 1 The following is a schematic diagram of a sleeve-type magnetic flux sensor. The length of the excitation coil is L, the number of excitation coil layers is m, and the diameter of the excitation coil copper wire is Φ. The length of the induction coil is l, the number of induction coil layers is m1, the diameter of the induction coil copper wire is d1, the radius of the steel cable is R0, the inner radius of the excitation coil is R1 = R0 + m1d1, and the outer radius of the excitation coil is R2 = R1 + mΦ. According to the Boit-Savart theorem and the principle of superposition of magnetic fields, when a current I is passed through the excitation coil shown in the figure above, the direction of the axial magnetic field is consistent with the axis of the solenoid, and the magnetic field strength at any point x on the axis, with the center of the induction coil as the origin, is as follows:

[0088]

[0089] The above formula is the prior art, wherein n=1 / Φ is the number of turns of the excitation coil per unit length.

[0090] In this example, the entire slime mold colony consists of several individuals, for example, 100, and the algorithm iteration count is set to 4000. Each individual slime mold, X, is represented by four-dimensional data (L, m, Φ, I), where each dimension represents the excitation coil length, the number of excitation coil layers, the excitation coil wire diameter, and the current required to be applied to the excitation coil. Upper and lower limits are set for each dimension: the excitation coil length, L, is in the range [0.15, 0.5], in meters; the number of excitation coil layers, m, is in the range [2, 25]; the excitation coil wire diameter, Φ, is in the range [0.0002, 0.002], in meters; and the initial upper and lower limits for the excitation coil current, I, are in the range [0, 3], in amperes.

[0091] One improvement of the original slime mold algorithm in this embodiment is that in the third step of updating the position, the upper and lower limits of the excitation coil current I are dynamically changed and set to [0, 1.5πΦ 2 *10 6 ], in amperes. It will change continuously according to the copper wire diameter Φ during the algorithm iteration, otherwise the excitation coil will be at risk of overcurrent and burnout; when any slime mold individual updates the third dimension data (copper wire diameter) and before updating the fourth dimension data (current), the upper and lower limits of the current are immediately set to [0, 1.5πΦ 2 *10 6 ].

[0092] In order for the improved slime mold algorithm to work, it is necessary to set the necessary fitness function F in the first sorting process, which is used to evaluate the quality of each slime mold individual. For the design task of the sleeve-type magnetic flux sensor excitation coil in this embodiment, the fitness function F is expressed as follows:

[0093] F=0.001F1+20F2+70F3+0.001F4 (9)

[0094] F1 is the magnetic field strength adaptability, 0.001 is the set weight, and its specific expression is as follows:

[0095] F1=0,ifH C1 ≤H(0)≤H C2 (10)

[0096] F1=|H C1 -H(0)|,if H C1 >H(0) (11)

[0097] F1=|H C2 -H(0)|, if H(0)>H C2 (12)

[0098] Where H(0) is the magnetic field strength at the center of the excitation coil, [H c1 ,H c2 ] is the magnetic field strength range required by the design to ensure the saturation magnetization of the tested cable. For steel cables, H c1 Can be set to 16000A / m, H c2 Can be set to 17000A / m.

[0099] F2 is the magnetic field uniformity adaptability, 20 is the set weight, and its specific expression is as follows:

[0100] F2=0,if u≤u m (13)

[0101] F2=uu m ,ifu>u m (14)

[0102] in is the uniformity of magnetic field intensity, l is the length of the induction coil, H(0.51) is the magnetic field intensity at the edge of the induction coil, H(0) is the magnetic field intensity at the center of the excitation coil, and u m =0.02 is the upper limit of the set magnetic field uniformity.

[0103] F3 is the adaptability of the amount of copper wire used in the excitation coil, with a weight of 70. Its purpose is to use as little copper wire as possible to save costs while ensuring that the magnetic field strength and uniformity meet the requirements. Its expression is as follows:

[0104]

[0105] Where R1 and R2 are the inner and outer radii of the excitation coil respectively, and L is the length of the excitation coil.

[0106] F4 is the adaptability of the number of coil layers, with a weight of 0.001. Its purpose is to minimize the number of coil layers to facilitate coil winding, provided that the magnitude and uniformity of the magnetic field strength are met and the copper usage is sufficient. Its expression is as follows:

[0107] F4=m (16)

[0108] Where m is the number of winding layers of the excitation coil.

[0109] In the iterative process of the improved slime mold algorithm, another improvement to the existing slime mold algorithm is that in the third step of updating the position, one of the iterative formulas (8) is replaced by Improved to:

[0110]

[0111] To solve the problem that the optimal parameters of coil design are not 0.

[0112] The final improvement of this embodiment to the original slime mold algorithm is that after each iteration of the second dimension data (number of excitation coil layers) of any slime mold, the number of coil layers is immediately rounded to an integer to ensure that the number of excitation coil layers is an integer.

[0113] This embodiment is specifically for a steel cable with a diameter of 90mm, an induction coil length of 0.016m, 2 layers, and a copper wire diameter of 0.00032m. Using the method provided by the invention, the length, number of layers, copper wire diameter, and applied excitation current of the magnetic flux sensor are quickly designed. The required magnetic field strength range is [16000A / m, 17000A / m], and the required magnetic field uniformity is better than 0.02 to ensure that the measured steel cable can be saturated and uniformly magnetized. Use python code to implement this algorithm, and the results after the algorithm is run are as follows Figure 2 As shown, the length of the excitation coil is 0.15m, the number of layers of the excitation coil is 3, the diameter of the excitation coil is 0.00134m, and the current required by the excitation coil is 8.47A.

[0114] In addition, if Figure 3 As shown in Figure 3, the moderate average value of the slime mold population gradually decreases during the algorithm iteration process, which also shows the effectiveness of the design of this method.

[0115] In summary, three improvements are made to the position update step of the original slime mold algorithm so that it can be used for excitation coil design.

[0116] First, the improvement for the position update step is to change one of the iterative formulas from Improved to To solve the problem that the optimal parameters of coil design are not 0.

[0117] Second, during the algorithm iteration process, each time the second-dimensional data of the slime mold is updated, that is, after the number of excitation coil layers m is updated, the number of excitation coil layers is immediately rounded to ensure that the number of excitation coil layers is an integer.

[0118] Third, during the algorithm iteration, each time the third-dimensional data of the slime mold is updated, that is, after the diameter Φ of the excitation coil copper wire is updated, the upper and lower limits of the current I are immediately modified to [0, 1.5πΦ 2 *10 6 ].

[0119] In addition, the fitness function of the improved slime mold algorithm is designed to be the fitness function F, and the expression is as follows:

[0120] F=0.001F1+20F2+70F3+0.001F4 (9)

[0121] The parameters in this embodiment can also be changed, for example, the number of iterations is [100, 10000], the number of slime molds is [10, 1000]; the fitness function F only contains F1-F2, or only contains F1-F3, or only contains F1-F4. These combinations can meet the design requirements of the magnetic flux sensor, and their effects are gradually improved; the first combination value considers that the performance of the magnetic flux sensor meets the requirements, the second combination can further reduce the amount of copper wire used, and the third combination can further reduce the number of copper wire layers (reducing the difficulty of industrial production winding). That is, the fitness function F can be expressed in the following forms:

[0122] F=0.001F1+20F2 (18)

[0123] Or the following:

[0124] F=0.001F1+20F2+70F3 (19)

[0125] Or:

[0126] F=0.001F1+20F2+70F3+0.001F4 (9)

[0127] The above embodiments are merely preferred embodiments of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions that fall within the scope of protection of the present invention are within the scope of protection of the present invention. It should be noted that improvements and modifications that can be made by a person skilled in the art without departing from the principles of the present invention are also considered to be within the scope of protection of the present invention.

Claims

1. An automated design method for an excitation coil of a sleeve-type magnetic flux sensor, characterized in that: The excitation coil in the sleeve-type magnetic flux sensor is designed using an improved slime mold algorithm. Each slime mold individual X contains four-dimensional data, namely, the length L of the excitation coil, the number of layers m of the excitation coil, the diameter Φ of the excitation coil copper wire, and the current I required to be applied to the excitation coil. First, the four-dimensional data of each slime mold X is randomly initialized. The iterative process of the improved slime mold algorithm includes the following steps: S1 sorting, calculate the fitness value F(X) of each slime mold individual X, and sort the slime mold individuals from best to worst according to their fitness values. The smaller the fitness value, the better the slime mold individual. Where F is the fitness function; S2 updates the weights. Different slime mold individuals use different update formulas when updating weights. The weights of the better 1 / 2 of the slime mold group are updated according to the following formula 1, and the weights of the other 1 / 2 of the slime mold group are updated according to the following formula 2: Where rand is a random number between [0,1]; X i is the slime mold individual that the algorithm is currently updating; X best is the optimal individual with the best fitness value in the slime mold population; X worst is the individual with the worst fitness value in the slime mold population; ε is a small positive number; w d new,i Slime Mold X i The d-th dimension weight of S3 updates the position, and the slime mold individual uses the weight to update its position. The position update formula is as follows: Among them, iter max is the maximum number of iterations set, iter is the number of iterations currently performed; x d max The maximum value of the d-th dimension data set by the algorithm, x d min is the minimum value of the d-th dimension data set by the algorithm; r1 and r2 are two randomly selected slime mold individuals, w d r1 is the d-th dimension weight of slime mold individual r1, x d r1 is the d-dimensional data of slime mold individual r1, x d r2 is the d-th dimension data of the slime mold individual r2; rand1 and rand2 are random numbers in [0,1], Z is a constant in [0,1]; randA is a random number in [-a,a], and randB is a random number in [-b,b]; x d best is the slime mold individual X with the best fitness value in the slime mold population best The d-th dimension data, x d i is the currently updated slime mold individual X i The d-th dimension data before updating, x d new,i Update the current slime mold individual X i Updated d-th dimension data; In step S3, each time the second dimension data of the slime mold individual is updated, that is, each time the number of excitation coil layers m is updated, the number of excitation coil layers is immediately rounded to an integer; In step S3, each time the third dimension data of the slime mold individual is updated, that is, after the diameter Φ of the excitation coil copper wire is updated, the upper and lower limits of the fourth dimension data current I of the slime mold are immediately modified to [0,1.5πΦ 2 *10 6 ].

2. The automated design method for the excitation coil of the sleeve-type magnetic flux sensor according to claim 1, characterized in that: The improved slime mold algorithm also The following processes are included: The entire slime mold population is assumed to contain a specific number of slime molds, with the specific number ranging from [10, 1000], and the number of algorithm iterations ranging from [100, 10000]. Upper and lower limits are set for each dimension of the slime mold population data. The excitation coil length L ranges from [0.15, 0.5], in meters. The number of excitation coil layers m ranges from [2, 25]. The excitation coil copper wire diameter Φ ranges from [0.0002, 0.002], in meters. The initial upper and lower limits of the excitation coil current I range from [0, 3], in amperes. At the beginning of the improved slime mold algorithm, data is randomly initialized for each slime mold individual X in the slime mold population within the upper and lower limits of each dimension of the slime mold population.

3. The automated design method for the excitation coil of the sleeve-type magnetic flux sensor according to any one of claims 1 to 2, characterized in that: In step S1, the fitness function F is used to evaluate the quality of each slime mold individual; the expression of the fitness function F is as follows: F=0.001F1+20F2 (18) F1 is the magnetic field strength adaptability, 0.001 is the set weight, and its specific expression is as follows: F1=0,if H C1 ≤H(0)≤H C2 (10) F1=|H C1 -H(0)|,if H C1 >H(0) (11) F1=|H C2 -H(0)|,if H(0)>H C2 (12) Where H(0) is the magnetic field strength at the center of the excitation coil, [H c1 ,H c2 ] is the magnetic field intensity range required by the design to ensure the saturation magnetization of the tested cable; for steel cables, H c1 Set to 16000A / m, H c2 Set to 17000A / m; F2 is the magnetic field uniformity adaptability, 20 is the set weight, and its specific expression is as follows: F2=0,if u≤u m (13) F2=u-u m ,if u>u m (14) in is the uniformity of magnetic field intensity, l is the length of the induction coil, H(0.5l) is the magnetic field intensity at the edge of the induction coil, u m =0.02 is the upper limit of the set magnetic field uniformity.

4. The automated design method for the excitation coil of the sleeve-type magnetic flux sensor according to any one of claims 1 to 2, characterized in that: The fitness function F in step S1 is expressed as F=0.001F1 + 20F2+70F3 (19) Where F1 is the magnetic field strength adaptability, 0.001 is the set weight, and its specific expression is as follows: F1=0,if H C1 ≤H(0)≤H C2 (10) F1=|H C1 -H(0)|,if H C1 >H(0) (11) F1=|H C2 -H(0)|,if H(0)>H C2 (12) Where H(0) is the magnetic field strength at the center of the excitation coil, [H c1 ,H c2 ] is the magnetic field intensity range required by the design to ensure the saturation magnetization of the tested cable; for steel cables, H c1 Set to 16000A / m, H c2 Set to 17000A / m; F2 is the magnetic field uniformity adaptability, 20 is the set weight, and its specific expression is as follows: in is the uniformity of magnetic field intensity, l is the length of the induction coil, H(0.5l) is the magnetic field intensity at the edge of the induction coil, u m =0.02 is the upper limit of the set magnetic field uniformity; F3 is the adaptability of the amount of copper wire used in the excitation coil, with a weight of 70. Its expression is as follows: Where R1 and R2 are the inner and outer radii of the excitation coil respectively, and L is the length of the excitation coil.

5. The automated design method for the excitation coil of the sleeve-type magnetic flux sensor according to any one of claims 1 to 2, characterized in that: The fitness function F in step S1 is expressed as F=0.001F1+20F2+70F3+0.001F4 (9) Where F1 is the magnetic field strength adaptability, 0.001 is the set weight, and its specific expression is as follows: F1=0,if H C1 ≤H(0)≤H C2 (10) F1=|H C1 -H(0)|,if H C1 >H(0) (11) F1=|H C2 -H(0)|,if H(0)>H C2 (12) Where H(0) is the magnetic field strength at the center of the excitation coil, [H c1 ,H c2 ] is the magnetic field intensity range required by the design to ensure the saturation magnetization of the tested cable; for steel cables, H c1 Set to 16000A / m, H c2 Set to 17000A / m; F2 is the magnetic field uniformity adaptability, 20 is the set weight, and its specific expression is as follows: F2=0,if u≤u m (13) F2=u-u m ,if u>u m (14) in is the uniformity of magnetic field intensity, l is the length of the induction coil, H(0.5l) is the magnetic field intensity at the edge of the induction coil, u m =0.02 is the upper limit of the set magnetic field uniformity; F3 is the adaptability of the amount of copper wire used in the excitation coil, with a weight of 70, and the expression is as follows: Where R1 and R2 are the inner and outer radii of the excitation coil, respectively, and L is the length of the excitation coil; F4 is the fitness of the number of coil layers, with a weight of 0.001, and the expression is as follows: F4=m (16) Where m is the number of winding layers of the excitation coil.

Images