Method for collecting and constructing a dataset of electrostatic induction signals of a flying charged body

Abstract

The application discloses a kind of flying charged electrostatic induction signal data set collection construction methods, including: including induction plate, insulating substrate and shielding electrode.Arranged in array on hemispherical shell, central axis points to the center of sphere, normal electrode inclination angle is 0 °.Experiment includes center point and area sweep mode, records four-way electrostatic induction signal.Using 20KHz sampling rate multichannel digital oscilloscope synchronous recording signal, carries out filtering and offset processing.Obtain time characteristic value data set by the time difference of signal peak time moment excavates flight trajectory information.The application can effectively capture the electrostatic field signal generated by flying charged body, provides data basis for analysis and application.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electrostatic induction, in particular to a method for collecting and constructing electrostatic induction signal data set of flying charged body. BACKGROUND

[0002] Unmanned platforms represented by unmanned aerial vehicles, unmanned vehicles and unmanned ships have the characteristics of high efficiency, flexibility and economy, and show the development trend of autonomy, intelligence and scalability, and have been widely used in daily life and confrontation conflicts and have achieved remarkable results. For example, unmanned aerial vehicles are used for logistics distribution, crop protection and forest fire monitoring, unmanned vehicles are used for battlefield reconnaissance and explosive handling, and unmanned ships are used for ocean monitoring. In the confrontation scene, in addition to electromagnetic interference, unmanned platforms are mainly subjected to physical attacks of high-speed flying targets such as light weapons, shrapnel jets and FPV collisions. However, the general sensors carried by unmanned platforms, such as cameras, radars and sonars, cannot detect the attacks of high-speed flying targets. Moreover, due to the limitations of energy, computing power, load and cost, unmanned platforms cannot carry complex threat perception systems such as high-speed cameras, warning radars or sound source positioning. Flying charged bodies rapidly accumulate a large amount of static electricity during flight due to friction, induction and other factors. If these static charges are not effectively discharged, they may cause discharge phenomena, strong electromagnetic interference, affect aircraft communication, and even cause fire or explosion accidents. In addition, the electromagnetic noise generated by static discharge can also cause serious interference to the radio systems such as communication and navigation of the aircraft. Therefore, in order to control the reasonable discharge of deposited static electricity, it is necessary to construct a data set of electrostatic induction signals of flying charged bodies. This data set can provide detailed data support and theoretical basis for the distribution characteristics of static electricity on the surface of the aircraft, which has important theoretical and practical significance for static protection design. By systematically collecting and analyzing the electrostatic induction signals generated by flying charged bodies, the generation and distribution rules of static electricity can be deeply understood, and more effective static protection measures can be designed to reduce the potential threat of static electricity to aviation safety and electronic device performance. SUMMARY

[0003] In order to solve the above problems, the present application provides a method for collecting and constructing electrostatic induction signal data set of flying charged body.

[0004] To achieve the above purpose, the technical scheme adopted by the present application is as follows:

[0005] The method comprises a charged target, an electrostatic field and a hemispherical electrostatic induction electrode array, a detection electrode of the hemispherical electrostatic induction electrode array comprises a cylinder structure composed of an induction plate, an insulating substrate and a shielding electrode, the inside bottom end of the cylinder structure is a circular plate-shaped induction plate for sensing a space electrostatic field and connected to a signal input end of an electrostatic sensor, the insulating substrate is used for electrically separating the induction plate from the shielding electrode, a bottom-cylindrical shielding electrode is connected to a power supply ground of the electrostatic sensor for blocking the influence of a lateral electrostatic field on the induction plate and improving the spatial directivity of the induction plate, the detection electrode obtains a current value of the charged target, and the current value is a signal amplified and recorded by a later-stage high-impedance electrostatic sensor for representing an electrostatic signal generated when the detection charged target passes through the detection electrode.

[0006] The hemispherical electrostatic induction electrode array is distributed on a hemispherical shell-shaped support structure, the central axis of each detection electrode points to the spherical center, the detection electrode whose central axis is perpendicular to the hemispherical bottom surface is defined as a normal electrode, and the inclination angle of the normal electrode is 0°; the inclination angle of other detection electrodes is defined as the included angle between the central axis of the normal electrode and the central axis of the other detection electrodes, and four detection electrodes are equidistantly arranged on the circumference of the inclination angle of 42° to detect the electrostatic field signal of the flying charged body, which comprises a center point shooting mode experiment and a region sweeping mode experiment; for the center point shooting mode experiment, the electrostatic test system is located in the center of the experimental site, the flying charged body shooting system is aimed at the position above the center of the compound eye electrode to emit, and the flying charged body is allowed to pass through the space directly above the center of the compound eye electrode to record four electrostatic induction signals; then the position of the shooting system is changed at intervals of 9 degrees, and a circle of the experimental site is traversed, and 120 groups of data are recorded at each position; for the region sweeping mode experiment, when emitting at each position, the test system is aimed at the left side, the center and the right side respectively, and 360 groups of data are recorded in each direction.

[0007] Further, the data cleaning operation of filtering and removing offset of the collected electrostatic induction signals comprises: the filtering operation uses a low-pass digital filtering function lowpass() of MATLAB to perform noise reduction processing on the recorded data, the cutoff frequency parameter is set to 10 Hz, the steepness parameter is set to 0.9999, and the stopband attenuation parameter is set to 180 dB; the offset removal operation uses a detrend() function of MATLAB, and sets the offset removal mode as constant, that is, the original data minus the direct current offset.

[0008] Furthermore, the method of mining flight trajectory information using the time difference of each signal peak moment to obtain a time characteristic value dataset includes: determining the peak values ​​of the four electrostatic induction signals, setting the minimum value of the four peak moments as the start moment, and the maximum value as the end moment. The interval between the start moment and the end moment is defined as the normalized duration of the test. The time difference between the four peak moments and the start moment is divided by the normalized duration to obtain the normalized time characteristic value, which is recorded as the normalized peak moment. The calculation method is shown in the formula;

[0009]

[0010] Further: the detection electrode obtains the current value of the charged target including:

[0011] A Cartesian coordinate system XOY is established on the horizontal ground, and the unit vectors of the three coordinate axes are and Assume that the radius of the detection electrode is r, and the center of its upper surface is defined as the origin of the space coordinate system O, and its normal vector is is the vector passing through the origin O and pointing vertically to the outside of the circular plate, where α is and The angle between The projection on the XOY plane is The amount of static charge carried by the charged target is Q(t), which will change during the flight. The dotted line is its motion trajectory, and the velocity vector is is the tangent vector of the motion trajectory, and the coordinates of the spatial position are (x p (t),y p (t),z p (t)), the unit vector of its position coordinate is and The angle between them is δ, and The angle is The angle between the projection on XOY and the X axis is γ;

[0012] The image charge method is used to calculate the field strength component of the target electric field in the normal direction of the detection electrode:

[0013]

[0014] Where ε0 is the dielectric constant of vacuum, ε r is the dielectric constant of the space medium; from formula (1), it can be concluded that when δ=Nπ, (N=0,1,2...), E n The maximum value, that is, when and When the detection electrodes are located on the same straight line and the normal direction faces the charged target, En The value is maximum;

[0015] Let α, β ∈ (0, π / 2) and And In the same quadrant, then:

[0016]

[0017] The field strength component of the normal direction of the detection electrode can be obtained by bringing formula (2) into (1):

[0018]

[0019] Let Then

[0020]

[0021] Assuming that the charged target is located at position Then formula (4) can be simplified as:

[0022]

[0023] When α, β ∈ (0, π / 2), both are pi / 4, the normal electric field strength is maximum, and then, according to Gauss theorem, the surface induced charge density σ s of the detection electrode is:

[0024]

[0025] Assuming that the charged target carries a charge amount Q0 which remains unchanged, the posture of the detection electrode is parallel to the horizontal plane, i.e. α = 0, then formula (6) is simplified as:

[0026]

[0027] When x p (t) = 0, y p (t) = 0, the value of σ s is maximum;

[0028] Let the distance from the projection of the charged target on the XOY plane to the coordinate origin be d p (t), i.e.

[0029]

[0030] When d p (t) = 0, z p (t) = Δ, the value of σ s is the normalized quantity, and Δ is the unit length; when z p (t) ≥ 6Δ, σ s is 0 ≤ d p (t) ≤ 6Δ.The value in the circular region of (t)≤Δ is approximately constant; according to Gauss theorem, the total amount of charge Q induced on the detection electrode s is:

[0031]

[0032] Therefore, when the flying height of the charged target is greater than 6 times the radius r of the detection electrode, the charge density induced on the electrode surface can be considered constant; therefore, equation (10) can be simplified as:

[0033]

[0034] According to the law of conservation of charge, the current flowing out of the detection electrode is the change in the amount of charge induced by the detection electrode, i.e.,

[0035]

[0036] This current value is the signal amplified and recorded by the high-impedance electrostatic sensor in the rear stage, which is used to represent the electrostatic signal generated when the charged target passes through the detection electrode.

[0037] Further: the current value of the charged target obtained by the detection electrode also includes obtaining the signal waveform:

[0038] Assuming that the posture of the detection electrode is parallel to the horizontal plane, the surface charge of the detection electrode is simplified as:

[0039]

[0040] Let the components of the motion speed V of the charged target along the coordinate axes be V x , V y , and V z , then

[0041]

[0042] Let the charge amount of the charged target be constant Q0, the flying speed be constant V0, the speed direction be parallel to the x-axis, and the trajectory be parallel to the x-axis, and pass through the YOZ plane at t=0, i.e., x p (0)=0, equation (15) can be simplified as:

[0043]

[0044] The distance between the charged target and the detection electrode is D p (t), i.e.

[0045]

[0046] Then:

[0047]

[0048] Bringing in (16), get

[0049]

[0050] Therefore, t = 0, formula (19) takes the maximum value, recorded as

[0051]

[0052] Then, formula (19) is written as:

[0053]

[0054] From formula (21), the waveform shape of the induced charge is determined by , which is the waveform time scale coefficient τ:

[0055]

[0056] Then, formula (21) is written as:

[0057]

[0058] Therefore, the waveform of the induced charge of the detection electrode is determined by τ, which is jointly affected by the flight speed V(t) of the charged target and the shortest distance D p (t) of the target to the electrode, and the influence of the two parameters is inversely proportional, and then formula (23) is brought into the induced current equation (13), that is

[0059]

[0060] Compared with the prior art, the technical progress obtained by the present application is:

[0061] The present application can comprehensively collect electrostatic induction signals of flying charged bodies at different positions and directions through the experimental design of center point shooting mode and area scanning mode, ensure the comprehensiveness and diversity of the data set, use a high sampling rate of 20KHz and a multi-channel digital oscilloscope to record synchronously, ensure the high resolution and synchronization of the signal data, through data cleaning operations such as filtering and offset removal on the collected signals, and using the time difference of the signal peak time to mine flight trajectory information, accurate and clean time characteristic value data set can be obtained, which provides reliable data support for subsequent electrostatic field analysis and flight trajectory prediction. DETAILED DESCRIPTION

[0062] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application.

[0063] In the drawings:

[0064] Figure 1 Schematic diagram of the electrostatic sensing electrode array system of the present application;

[0065] Figure 2 Schematic diagram of the influence of the posture of the detection electrode on the normal electric field of the present application;

[0066] Figure 3 Schematic diagram of the normalized surface charge density of the present application;

[0067] Figure 4 Schematic diagram of the waveform and frequency spectrum of the induced charge of the detection electrode of the present application;

[0068] Figure 5 Schematic diagram of the waveform and frequency spectrum of the induced current of the detection electrode of the present application;

[0069] Figure 6 Schematic diagram of the principle and design of the shark electroreceptor organ simulation of the present application;

[0070] Figure 7 Schematic diagram of the test experiment principle of the present application;

[0071] Figure 8 Design diagram of the two-dimensional posture control mechanism of the present application;

[0072] Figure 9 Schematic diagram of the original test data of the electrostatic sensing signal of the present application;

[0073] Figure 10 Schematic diagram of the normalized time characteristic value of the electrostatic sensing signal of the present application;

[0074] Figure 11 Schematic diagram of the prediction results and errors of Model 1 on each data set of the present application;

[0075] Figure 12 Schematic diagram of the prediction results and errors of Model 2 on each data set of the present application;

[0076] Figure 13 Prediction error distribution diagram of all models of the present application. DETAILED DESCRIPTION

[0077] The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0078] Electrostatic charging of flying projectiles (i.e. flying charged bodies) has been reported since the 1980s. Due to the electrification effects of the primer explosion shock, flame and compound powder adhesion during the launching process of flying projectiles, the friction between flying projectiles and gun barrels, and the friction between flying projectiles and air molecules during the flying process, flying projectiles at high speed will carry a certain amount of static charge. The amount of charge is related to factors such as ammunition type, weapon state, and environmental conditions, and is distributed in the range of 3-5 orders of magnitude. The flying projectiles of light weapons carry static charges in the order of pC to nC, and the flying projectiles of rockets and aircraft carry more charges due to their large surface area. The static charge on the surface of the flying projectile will generate a static electric field around it. The flying process will disturb the static electric field around the space of its running track, which will be detected by various electric field sensors. Some studies have used cylindrical and window-shaped induction electrodes combined with charge amplifiers to test the electrostatic induction signals of flying projectiles. Some studies have used steel mesh to make cylindrical electrostatic sensors to test the charge of 9mm Beretta pistol flying projectiles. Some studies have compared the detection performance of flying projectiles by non-contact potential sensors, vector electric field sensors and charge induction sensors through target range experiments, and have proposed a quasi-static electric field sensor array based on varactor diodes in subsequent work. Three sensors are placed 3 meters apart, and the test signals are transmitted to the central data processing unit using a wired network. The direction of the flying projectile is indicated by using positioning algorithms such as triangulation and wavelet analysis. Some studies have used deep learning assisted self-powered multifunctional electret sensors to realize flying charged body recognition. The above work has researched the charge of flying projectiles, the collection of electrostatic signals, feature analysis and potential applications, but it is not suitable for unmanned platforms with limited space, load and computing power. The main problems are:

[0079] (1) Cylindrical induction electrodes require flying projectiles to pass through to test the signal, which cannot detect flying projectiles from any direction. Large-size flat electrodes or wide-spaced electrode arrays are not suitable for unmanned platforms with limited space and load, and will affect the concealment and maneuverability of the unmanned platform.

[0080] (2) The electrostatic test experiment system generally improves the sensitivity and stability of the experiment system by increasing the induction electrode, shielding the interference signal, manually resetting, and controlling the experimental environment. The high sensitivity and high stability of the sensor circuit are not considered, and the power consumption of the sensor and the adaptability to vibration and impact are not considered.

[0081] (3) The information mining algorithm of electrostatic induction signal needs to process the entire time series in real time. The algorithm complexity of data acquisition, transmission and processing is high, and it needs to occupy high data bandwidth and computing resources, which is not suitable for edge computing platforms with limited computing power.

[0082] To solve the above problems, the present application is inspired by the Lorenzini organ (Ampullae of Lorenzini array) of sharks and other organisms, and proposes an electrostatic induction electrode array with attitude and spatial distribution differentiation characteristics. The target information is mined by using the difference of the electrode array induction signal. Firstly, the influence law of electrode parameters on the induction signal is theoretically analyzed. A high-sensitivity, high-precision and low-noise electrostatic induction signal amplification circuit is designed, which can flexibly adjust the frequency response characteristics, and a prototype system is realized based on a commercial amplifier. A flying charged body laboratory simulation system is built, and the flying charged body induction signal experimental data set is obtained. The symbolic regression machine learning algorithm is used to mine the time characteristics of the electrostatic signal, and the mapping function of the signal characteristics and the incoming direction of the flying projectile is obtained, so as to realize a low-power, low-cost, low-computing power requirement and highly integrated charged body flight trajectory perception method, and the omnidirectional prediction error is less than 8 degrees.

[0083] In the non-contact electrostatic test system composed of a charged target, an electrostatic field and a detection electrode, the electrostatic field is the medium connecting the target and the electrode. The induced charge quantity of the detection electrode can be calculated by using Gauss theorem and charge conservation law. In order to be general, the electrostatic detection schematic diagram of the charged target is shown in Figure 1

[0084] As shown in Figure 1 , the Ground plane is an infinite horizontal ground, on which a Cartesian space coordinate system is established, and the unit vectors of the three coordinate axes are , The golden yellow circular plate is a detection electrode with a finite size, conductivity and grounding in space, with a radius of r, and the upper surface center of the circular plate is defined as the origin O of the space coordinate system. The normal vector of the circular plate is , which is a vector perpendicular to the outside of the circular plate through the origin O. The two angles between the normal vector and the space coordinate axes can determine the spatial attitude of the detection electrode. Alpha is the angle between and , and beta is the angle between the projection of XOY plane and . The blue ball is a flying charged target, which carries an electrostatic charge with a charge quantity of Q(t), which changes during flight. The dash-dot line is the motion trajectory, and the velocity vector is . The tangent vector of the motion trajectory is p , and the coordinates of the space position are (x p (t), y p (t), z r (t)). The unit vector of the position coordinates is , and the angle between and is δ, and ​The angle between the projection on XOY and the X axis is γ.

[0085] The image charge method is used to calculate the field strength component of the target electric field in the normal direction of the detection electrode:

[0086]

[0087] Where ε0 is the dielectric constant of vacuum, ε r is the dielectric constant of the space medium. From formula (1), it can be seen that when δ=Nπ, (N=0,1,2...), E n The maximum value is when and When the detection electrodes are located on the same straight line and the normal direction faces the charged target, E n The spatial posture of the detection electrode determines the spatial position of the charged target to which it is most sensitive, which shows that electrostatic induction has directional selectivity.

[0088] Furthermore, according to Figure 1 The geometric relationship shown and the symmetry of the detection electrode can be set as α, β∈(0,π / 2) and and Located in the same quadrant, then:

[0089]

[0090] Substituting equation (2) into equation (1), we can obtain the field intensity component in the normal direction of the detection electrode:

[0091]

[0092] set up but

[0093]

[0094] In order to intuitively show the influence of the detection electrode posture parameters (α, β) on the sensing signal, it is assumed that the charged target is located at position The above formula can be simplified as:

[0095]

[0096] When α, β∈(0,π2), the change law of the normalized value of the normal electric field is as follows: Figure 2 As shown, from Figure 2 It can be seen that when α and β are both pi / 4, the normal electric field strength is the largest, that is, and When they coincide, the normal field strength is the largest, and the maximum value differs from the minimum value by more than twice.

[0097] Furthermore, according to Gauss's theorem, the induced charge density σ on the surface of the detection electrode iss is:

[0098]

[0099] On this basis, the influence of the charged target space position on the probe electrode surface induced charge density σ s is analyzed. In order to simplify the calculation, it is assumed that the charged target carries a constant charge amount Q0, and the probe electrode posture is parallel to the horizontal plane, i.e. α = 0, then equation (6) is simplified as:

[0100]

[0101] Obviously, when x p (t) = 0, y p (t) = 0, σ s takes the maximum value, that is, when the charged body reaches directly above the sensing electrode, the charge amount induced on the electrode is the highest.

[0102] Let the distance from the projection of the charged target on the XOY plane to the coordinate origin be d p (t), that is,

[0103]

[0104] Then,

[0105]

[0106] Take σ p (t) = 0, z p (t) = Δ, the value of σ s is a normalized quantity, and Δ is the unit length. Then, the change rule of the probe electrode surface induced charge density σ s with the relative distance d between the charged body and the probe electrode is shown in Figure 3 .

[0107] It can be seen from Figure 3 that when z p (t) ≥ 6Δ, σ s in the circular region of 0 ≤ d p (t) ≤ Δ is approximately constant. According to Gauss's theorem, the total amount of charge Q s induced on the probe electrode is:

[0108]

[0109] Therefore, when the flight height of the charged target is greater than 6 times the radius r of the probe electrode, the charge density induced on the electrode surface can be considered as a constant, and therefore, equation (10) can be simplified as:

[0110]

[0111] According to the law of conservation of charge, the current flowing out of the detection electrode is the change in the amount of charge induced by the detection electrode, that is:

[0112]

[0113] This current value is the signal amplified and recorded by the high-impedance electrostatic sensor behind, which is used to represent the electrostatic signal generated when the charged target passes through the detection electrode.

[0114] In order to further intuitively analyze the signal waveform, without loss of generality, it is assumed that the posture of the detection electrode is parallel to the horizontal plane, and the surface charge of the detection electrode is simplified as:

[0115]

[0116] Let the components of the movement speed V of the charged target along the coordinate axes be V x , V y , and V z , respectively, so that

[0117]

[0118] Further, let the charge amount of the charged target be constant Q0, the flight speed be constant V0, the speed direction be parallel to the x-axis, and the flight trajectory be parallel to the x-axis, and the charged target passes through the YOZ plane at t=0, that is, x p (0)=0. Equation (15) can be simplified as:

[0119]

[0120] The distance between the charged target and the detection electrode is D p (t), that is,

[0121]

[0122] Then:

[0123]

[0124] Substituting equation (16) into equation (19) gives

[0125]

[0126] Therefore, at t=0, equation (19) takes the maximum value, denoted as

[0127]

[0128] Then, equation (19) can be simplified as:

[0129]

[0130] From equation (21), it can be seen that the waveform shape of the induced charge is determined by , which is denoted as the waveform time scale coefficient τ:

[0131]

[0132] Therefore, equation (21) can be simplified as:

[0133]

[0134] Therefore, the waveform of the induced charge on the probe electrode is determined by τ, which is jointly affected by the flying speed V(t) of the charged target and the shortest distance D p (t) between the target and the electrode, and the effects of the two parameters are inversely proportional. When the waveform coefficient takes different values, the charge waveform and its spectrum are as shown in Figure 4 .

[0135] Further, equation (23) is brought into the induced current equation (13), and then

[0136]

[0137] Therefore, Figure 4 The current waveforms and their spectrum graphs corresponding to the curves in equation (24) are shown in Figure 5 . From Figure 4 , 5 , it can be seen that the signal energy is mainly concentrated in the low frequency, and the low frequency bandwidth of the sensor needs to be reduced to about 0.1 Hz to effectively amplify the electrostatic induction signal without distortion. When the low frequency characteristics of the sensor are insufficient, the sensor has the characteristics of a high-pass filter, and the distortion of the induced signal is the same as that of a differentiator.

[0138] Specifically, each electrostatic induction electrode is composed of a sensing plate, an insulating substrate, and a shielding electrode in a cylindrical sandwich structure, as shown in Figure 6 (B). The bottom end of the cylinder is a circular plate-shaped copper induction plate used to induce the space electrostatic field and connected to the signal input end of the electrostatic sensor. The insulating substrate made of polyester imine is used to electrically separate the induction plate from the shielding electrode. The copper shielding electrode with a bottom cylinder is connected to the power supply ground of the electrostatic sensor and is used to block the influence of the lateral electrostatic field on the induction plate, thereby improving the spatial directivity of the electrostatic induction plate. The higher the cylinder height, the better the spatial directivity, but the weaker the induced signal strength; the lower the cylinder height, the worse the spatial directivity, but the stronger the induced signal strength. The present application sets the thickness of the induction plate to 1 mm, the diameter to 20 mm, the thickness of the insulating substrate to 0.5 mm, the radius of the shielding electrode to 21 mm, and the cylinder height to 20 mm.

[0139] The electrostatic sensing electrode array is distributed on the hemispherical shell support structure, and the central axis of each electrode points to the spherical center. The electrode whose central axis is perpendicular to the hemispherical bottom surface is defined as the normal electrode, and its inclination angle is recorded as 0°. The angle between the central axis of other electrodes and the central axis of the normal electrode is defined as the inclination angle of the electrode, that is, the direction in which the spatial sensitivity of the electrode is the highest. The present application equally sets 6 electrodes on the circumferences with inclination angles of 0, 21, 42 and 63 degrees, and 5 electrodes on the circumferences with inclination angles of 36 and 72 degrees.

[0140] The electrostatic sensing signal amplification circuit is described in the literature M. Man, M. Wei, Y. Zhang, G. Ma, and Y. Chen, "Biomimetic Measurement Method for Surface Electric Potential Imaging Inspired by Visual Lateral Inhibition," IEEE Transactions on Industrial Electronics, pp. 1-11, 2023, doi: 10.1109 / TIE.2023.3314910.

[0141] The electrostatic sensing signal amplification circuit is described in the literature M. Man, M. Wei, Y. Zhang, G. Ma, and Y. Chen, "Biomimetic Measurement Method for Surface Electric Potential Imaging Inspired by Visual Lateral Inhibition," IEEE Transactions on Industrial Electronics, pp. 1-11, 2023, doi: 10.1109 / TIE.2023.3314910.

[0142] M. Man and M. Wei, "Remote monitoring method for human body electrostatic potential based on symbolic regression machine learning," Measurement Science and Technology, vol. 34, no. 6, p. 065116, 2023 / 03 / 24 2023, doi: 10.1088 / 1361-6501 / acc3b6. The DC voltage gain of the sensor in the range of 0.1 Hz to 1 kHz is 34 dB, which can effectively amplify the weak electrostatic induction signal. The 1 / f corner frequency of flicker noise is about 30 Hz, and the sensor Gaussian white noise above this frequency point is It should be noted that in order to improve the spatial resolution of the space electrostatic field test, it is necessary to reduce the physical size of the induction electrode, thereby reducing the induction area. This also has a negative impact, that is, the equivalent coupling capacitance between the charged body and the induction electrode is greatly reduced to the picofarad or even femtofarad level. This requires the amplifier circuit to have extremely small input capacitance and extremely high input resistance to match the equivalent coupling capacitance to improve the voltage division ratio and sensitivity. The use of the most advanced positive feedback circuit design technology (bootstrap, neutralization) has to some extent alleviated this problem. However, the mutual restraint problem of spatial resolution and signal sensitivity of such sensors has not been completely solved.

[0143] Each electrode of the electrostatic induction electrode array is connected to an independent amplification circuit, which amplifies the weak electrostatic induction signal and transmits it to the rear-end connected multi-channel oscilloscope for digital acquisition and recording. Thus, the multi-channel synchronous recording of the surrounding space electrostatic field is realized.

[0144] 3.3 Symbolic regression algorithm

[0145] When the flying projectile passes over the unmanned platform at a certain distance (≤10 m), the disturbance of the space electrostatic field caused by the flying projectile carrying static charge will be detected and recorded by the electrostatic field sensor, and saved as a set of time series signals with differences. Analyzing this data can obtain the speed, trajectory, direction, etc. of the flying projectile. Since the energy, computing power, load and cost of the unmanned platform are limited, the present application selects symbolic regression algorithm for data mining. Symbolic regression is a method that searches the mathematical expression space using evolutionary algorithms, with the goal of minimizing the error between measured data and predicted data, and automatically finding the mathematical expression behind the measured data. The trained symbolic model is convenient to deploy without the need for special software and hardware environments (such as neural networks, vector machines, etc.). The present application uses a multi-gene symbolic regression method, which combines the population search ability of multi-gene genetic algorithm (Multi-Gene Genetic programming, MGGP) and the parameter estimation ability of linear least squares method, to find the optimal linear combination of mathematical expressions corresponding to all gene individuals in the population, so that the error between the predicted output response of the mathematical model and the target output response is minimized. The algorithm principle and implementation details are explained in detail in the literature M. Man and M. Wei, "Remote monitoring method for human body electrostatic potential based on symbolic regression machine learning," Measurement Science and Technology, vol. 34, no. 6, p. 065116, 2023 / 03 / 24 2023, doi: 10.1088 / 1361-6501 / acc3b6.

[0146] M. Man, Y. Zhang, G. Ma, Z. Zhang, and M. Wei, "Indoor Localization Method of Personnel Movement Based on Non-Contact Electrostatic Potential Measurements," Sensors, vol. 22, no. 13, p. 4698, 2022.

[0147] The difference between the symbolic regression algorithm used in this experiment is the special fitness function. The fitness function selects the circular absolute error (Circle absolute error) between the predicted value (Ye) and the target value (Ya), and the calculation method is as follows:

[0148] CAE = min (|Y e -Ya |,360-|Y e -Y a |)

[0149] This is because the direction of the flying projectile is a point on the circumference, the predicted direction and the real direction both belong to the point on the circumference, and the value of the point on the circumference is periodic with 360. For example, the real shooting direction is 10°, when the predicted direction of attack is 30°, the prediction error is 20°. When the predicted direction of attack is 350°, the prediction error is also 20° instead of 340°, because the interval between 10° and 350° on the circumference is 20°. Therefore, CAE is used as the fitness function in this experiment.

[0150] A test scene of flying charged body electrostatic induction signal is built in the laboratory, as shown in Figure 7 (a). In the laboratory circular field with a radius of 4 m, the electromagnetic ejection module and the two-dimensional attitude control mechanism constitute the flying charged body launching system, which launches a metal cylinder carrying static electricity. The mechanical dog carries a static electric field sensor to synchronously collect the spatial static electric field disturbance signals caused by the flying charged body.

[0151] 1. Flying charged body experimental device

[0152] An electromagnetic ejection module and a two-dimensional attitude control mechanism are used to constitute a flying charged body launching system. The electromagnetic ejection module is a 5-stage copper coil accelerator (wire diameter 0.8 mm, 195 turns, outer diameter 20 mm, inner diameter 8 mm, thickness 26 mm, inductance 2.4 mh), and the driving capacitors of each stage of coil are 2000 uF, 1440 uF, 1440 uF, 1000 uF, and 1000 uF, respectively. The capacitor charging voltage range is (50 V-400 V), which is quantitatively controlled by a contact switch, and the guide rail is a PVC circular pipe (radius 3.2 mm, length 500 mm). The launched flying projectile is a stainless steel cylindrical body (radius 3 mm, height 30 mm), and the maximum initial speed of the flying projectile is about 48 m / s. When the capacitor is triggered, the static electricity stored in the capacitor flows through the coil quickly, generating a transient magnetic field to excite the flying projectile to move in the guide rail. The friction between the flying projectile and the guide rail generates static electricity, and after flying out, it becomes a charged flying body. Repeated experiments have proved that this electrification process is relatively stable, and about 95% of the launch experiments will make the flying projectile charged.

[0153] The trajectory of the flying projectile is determined by the guide rail pointing and the initial speed of the flying projectile. In order to accurately control the trajectory of the flying projectile and improve the repeatability of the experiment, a two-dimensional attitude control mechanism is designed, as shown in Figure 8As shown in the figure, the control mechanism includes a base, a rotating platform, a rotating drive, a limiter, and a manual control device. The rotating platform consists of two axes: the azimuth axis and the pitch axis. Both axes are supported by two pairs of precision angular bearings to ensure axial and radial stiffness and rotational accuracy. The azimuth axis has a rotation angle range of (-40° to +40°), and the pitch axis has a rotation angle range of (0° to +15°). Precise control is achieved by sending control signals to the rotating drive via an electronic handwheel, with an adjustment resolution of 1°.

[0154] 2. Two experimental modes

[0155] Using the four sensing electrodes with an inclination angle of 42° in the space electrostatic field sensor ( Figure 7 In Figure 1, the four colors (red, blue, green, and orange) are marked as Sensor A, B, C, and D, respectively, to detect the electrostatic field signal of the flying charged object and mine its flight angle information. In order to verify the effectiveness and versatility of the present invention, the present invention designed two types of experiments, namely the center point shooting mode experiment and the area scanning mode experiment. Figure 7 As shown in (b) and (c). For the center point shooting mode experiment, the electrostatic test system is located in the center of the experimental site. The flying projectile launch system is aimed at 1m above the center of the compound eye electrode and launched, causing the flying projectile to pass through the airspace directly above it. The electrostatic test system records four electrostatic induction signals. Then, the position of the launch system is changed at intervals of 9 degrees, traversing the experimental site once. The experimental results are recorded three times at each position, and a total of 120 sets of data are collected. The difference for the area sweeping mode experiment is that when launching at each position, it is aimed at the left, center, and right sides of the test system respectively, and launched three times in each direction, so a total of 360 sets of data are recorded.

[0156] 3. Data Processing

[0157] During the experiment, a multi-channel digital oscilloscope (PicoScope 4824) was used to synchronously record the four-way electrostatic induction signals at a sampling rate of 20 kHz. Figure 9 shown.

[0158] Figure 9 (a) In order to facilitate manual observation, the offset is added to each channel signal in the oscilloscope. As can be seen from the figure, the four-way signal shows good consistency. The peak part of each signal is amplified and the offset is removed, as shown in the figure. Figure 9 As shown in (b), the signal characteristics meet the Figure 5Theoretical analysis of the flying charged body electrostatic induction signal, and the peak value and peak time of four signals have obvious differences, which shows that the system can accurately test the electrostatic induction signal when the flying charged body sweeps, and the direction selectivity of electrostatic induction is realized by using the spatial position and attitude difference of the induction electrode, and the electrostatic test with spatial filtering characteristics is realized.

[0159] There are DC bias, 50Hz power frequency noise and high frequency noise in the test results, which is due to the good low frequency characteristics of the sensor, the test bandwidth up to 100kHz, and the influence of the electrostatic discharge phenomenon in the experiment. Therefore, the collected raw data needs to be filtered and offset for data cleaning operation. The filtering operation is to use the low pass digital filter function lowpass() of MATLAB to process the recorded data, and the cutoff frequency parameter is set to 10Hz, the steepness parameter is set to 0.9999, and the stopband attenuation parameter is set to 180dB. The offset operation uses the detrend() function of MATLAB, and sets the offset removal mode as constant, that is, the original data minus the DC offset. In addition, there are two relatively large noise points in the signal, the first noise point is before the electrostatic induction signal, which is due to the electromagnetic radiation signal generated in the moment when the flying projectile separates from the sliding track, and the second noise is after the electrostatic induction signal, which is due to the electrostatic discharge signal generated when the flying charged body hits the protective foam. These two parts of noise are not the electrostatic field induction signal to be analyzed in this test, so they are directly eliminated. Finally, taking the peak time of SensorA signal as the reference point, each channel data only retains the data of 0.1s before and after the reference point. The data cleaning result is as Figure 9 (c) shows.

[0160] 4. Feature extraction

[0161] Due to the difference of the friction pair state between the surface of the flying projectile and the surface of the guide rail, the voltage attenuation of the electromagnetic ejection energy storage capacitor, the size of the flying projectile and other factors, the actual charge carried by the flying projectile in each launch has strong uncertainty. Therefore, the amplitude of the electrostatic induction signal obtained by testing under the same test conditions has strong randomness, which cannot be used for information mining. The relative time of the flying charged body flying through different induction electrodes is a fixed value, which is only affected by the flying speed and the relative position of the electrode array. Therefore, in the test, the time difference of the peak value of each signal is used for information mining of the flight trajectory.

[0162] Firstly, the peak value of four electrostatic induction signals is determined (the peak() function of matlab is used), as shown in Figure 10(a) as shown, respectively labeled as Peak A, Peak B, Peak C, Peak D, whose abscissa is the peak time, respectively recorded as T1, T2, T3, T4. Further, the minimum value of the four peak times is set as the starting time, the maximum value is set as the ending time, and the interval between the starting time and the ending time is defined as the normalized duration of this test. Then, the time difference between the four peak times and the starting time divided by the normalized duration is the normalized time characteristic value, recorded as the normalized peak time, and the calculation method is shown in the formula.

[0163]

[0164] The radar chart of the normalized time characteristic values of the four signals obtained in one test is shown in FIG. Figure 10 (b) as shown. In the first test mode, the normalized time characteristic values of the test data obtained by 40 shooting angles, and the radar chart thereof are shown in FIG. Figure 10 (c, d) as shown. It can be seen from the figure that the signal characteristics have obvious correlation with the shooting angle.

[0165] The advantage of this normalization processing is that the four time characteristic values are only affected by the shooting angle of the flying projectile, and are not affected by the flight speed and the relative distance. The signal waveform corresponds to the dynamic process of the flying projectile flying across the electrode sensing area. The waveform of the electrostatic induction signal is mainly determined by the time scale coefficient τ, that is, the flight speed V(t) and the shortest distance D p (t) together. Because after the electrode structure and posture are fixed, the relative distance determines the sensing area. When the relative distance is unchanged and only the flight speed changes, the time length and peak time of the four-way induction signal also change, but the relative interval of the peak time changes in the same proportion, inversely proportional to the flight speed, and this proportion coefficient is exactly divided when the normalized time characteristic value is calculated. Therefore, the change of the flight speed will not affect the normalized time characteristic value. Similarly, when the flight speed is unchanged and only the relative distance changes, the sensing areas of the four electrodes change in the same proportion, and the peak times of the four-way induction signal also change in the same proportion, which is directly proportional to the relative distance. When the normalized time characteristic value is calculated, this proportion coefficient can also be divided. Therefore, the change of the relative distance will not affect the normalized time characteristic value.

[0166] 5. Inversion results

[0167] The symbolic regression algorithm is used to mine the mapping function between the normalized time characteristic values of the electrostatic induction signal and the shooting angle. The shooting angle of each experiment is taken as the dependent variable of the mapping function, and the normalized time characteristic values of the four-way induction signal are taken as the independent variable. The symbolic regression algorithm is used to optimize the mapping function, and a mathematical regression model from the sensor induction signal to the bearing of the charged body is expected to be obtained.

[0168] The present application respectively in the center point shooting mode and area scanning mode two experimental mode, each carried on three round omnidirectional 360 degree traversal experiment, obtained six groups of data sets, named M1R1 (Mode1Run1), M1R2, M1R3, M2R1, M2R2 and M2R3. Further, each data set is used as a training sample for model training, and the remaining data is used as a validation sample set. Running a symbol regression algorithm, a set of optimal models is obtained. Finally, the above process is repeated ten times, each time the initial value of the random evolution algorithm is randomly evolved, and the other parameters remain unchanged, to obtain the optimal model of each data set.

[0169] The optimal model obtained by training the M1R1 data set is:

[0170] Y e = 0.453 + 0.453T2 cos(T4) - 0.211T2 + 0.506T1 cos(4.07cos(3.07+2.51T2 cos(T2 cos(0.808+2.51T2))))

[0171] Where Ye represents the model estimate of the direction of the incoming projectile, T1-T4 represent the normalized time characteristic values of the four induction signals. The prediction results of this model on the training data set M1R1 and the validation data sets M1R2 and M2R1 are shown in Figure 11 (a, b, c).

[0172] Figure 11 In (a), the horizontal coordinate is the 40 shooting angles (interval 9 degrees traversal a week) in the training data set, the vertical coordinate is the angle value (0-360 degrees), the red line is the target value of the shooting angle, the blue line is the predicted value of the shooting angle, and the yellow bar corresponds to the prediction error of the shooting angle. As can be seen from the figure, this model has very high prediction accuracy, with an average absolute error of 3.39 degrees, as shown by the green line. The error of the model at individual shooting angles (such as 45, 54 and 99 degrees) is larger, reaching more than 10 degrees, which is considered to be caused by the instability of electrostatic charging in the test. It can be seen that the symbol regression algorithm has sufficient modeling and prediction ability, meeting the demand for predicting the direction of attack.

[0173] Further, the generalization ability of this model is verified on other data sets. Figure 11(b) is the prediction result of this model on the validation data set M1R2, the prediction accuracy of this model still maintains a high level, the average absolute error is 5.55 degrees. In addition, the average absolute error of this model on M1R3 is 4.83 degrees, the result graph is similar to this graph, and will not be described again. Data sets M1R1, M1R2, M1R3 are all from the first test mode. Therefore, this model has good generalization ability for point shooting test mode, and the prediction absolute error is about 5 degrees.

[0174] Figure 11 (c) is the prediction result of this model on the validation data set M2R1 of the sweep shooting test mode, the prediction accuracy of this model decreases significantly, and the average absolute error increases to 13.01 degrees. In the sweep shooting test mode, each shooting angle is shot three times, which are left side shooting, central shooting and right side shooting. Therefore, in the sweep mode, 120 groups of data will be collected by traversing a circle every 9 degrees, including 40 groups of data for left side shooting, central shooting and right side shooting. In the figure, the dark red line is the prediction result of the model for the forty groups of data of the left side shooting, the blue line is the prediction result of the model for the central shooting data, and the orange line is the prediction result of the model for the right side shooting data. In the figure, the blue line is closer to the red line than the dark red and orange lines, and the blue error bar is shorter than the dark red and orange error bars. This shows that the prediction accuracy of the model for the central shooting data is higher, which produces the overfitting phenomenon commonly seen in machine learning training. This is not conducive to the generalization and application of the model, and when the projection of the flying projectile trajectory does not pass through the center of the sensor, the prediction error of the model will be greatly increased, greatly reducing the universality of this method.

[0175] The prediction error of this model on all data sets is shown in Figure 11 (d) The average prediction absolute error of the model on the data sets M1R1, M1R2, M1R3 is about 5 degrees, and there are fewer outliers, and the extreme value will not exceed 40 degrees. The average prediction absolute error of the model on the data sets M2R1, M2R2, M2R3 increases to about 15 degrees, and the outliers increase significantly, and the maximum error reaches 160 degrees. This shows that this model has good prediction ability for point shooting mode test, with an error of about 5 degrees. The prediction ability for sweep shooting test data is poor, with an error of about 15 degrees. The generalization ability of this model is weak, and the fluctuation of the trajectory deviation from the center of the sensor will greatly affect the prediction accuracy.

[0176] The optimal model obtained by training with the M2R3 data set is:

[0177] Y e = 0.571 + T1T2 + 0.353T1T4 + 0.0972sin(T2 2) -0.214T4-2.13T1 mod (0.571T2-8.03, 0.489)

[0178] The prediction results of this model on the training dataset M2R3 and the validation datasets M2R1, M1R1 are shown in Figures Figure 12 (a, b, c) respectively.

[0179] Figure 12 In (a, b), the mean absolute error of this model on the training dataset M2R3 is 7.22 degrees, and on the validation dataset M2R1 is 7.29 degrees. Figure 12 In (a), when the shooting angle is 0, the predicted angle is 357.8 degrees, and when the shooting angle is 351 degrees, the predicted angle is 0.02 degrees. Figure 12 In (b), when the shooting angle is 9, the predicted angle is 354.9 degrees. Since the angle value is periodic with 360, the actual absolute error of the model at these three shooting angles is not significant. Therefore, the absolute error distribution of the model on all data is consistent. In addition, the mean absolute error of this model on M2R2 is 7.66 degrees, and the result figure is similar to Figure 12 (b) and will not be repeated here. Therefore, this model has high prediction accuracy for all data in the sweep mode, with an absolute error of about 7 degrees.

[0180] Figure 12 (c) is the prediction result of this model on the point shooting mode dataset M1R1. The prediction accuracy of this model does not decrease but increases, and the mean absolute error decreases to 5.79 degrees. There are two reasons for this. First, M1R1 has a small amount of data and is more regular. Because the trajectory of central shooting is relatively easy to control, the shooting aiming sensor is directly above the center. The left and right shooting trajectories have high randomness in deviation, and the shooting aiming offset needs to be controlled manually, resulting in larger data noise. Second, because the sweep mode data contains point shooting mode data, model 2 has learned the rules contained in the point shooting mode data during the training process. In order to verify this conclusion, the prediction errors of this model on all datasets are summarized in Figure 12 (d). The mean prediction absolute error of the model on the M1R1, M1R2, M1R3 datasets is about 6 degrees, with fewer outliers and an extreme value of no more than 40 degrees. The mean prediction absolute error of the model on the M2R1, M2R2, M2R3 datasets is about 7 degrees, with also fewer outliers and an extreme value of no more than 30 degrees. This proves that this model has good prediction ability for point shooting mode and sweep mode tests, with a mean error of less than 8 degrees. Therefore, this model has high prediction accuracy and strong generalization ability, and can accurately predict the incoming angle when the charged flying body passes through the area above the sensor.

[0181] In addition, the optimal models trained by the data sets M1R2, M1R3, M2R1, M2R2 are respectively:

[0182]

[0183] Y e = 0.444 + 0.265T2 + 1.32T1T2 + 0.34T1T4 - 0.627T1 - 1.49T1mod(T2, 0.696) - 0.146T2T4cos(T2T3)

[0184] Y e = 0.85 + 0.847T1T2 + 0.358T1T4 - 0.182T4 - 0.311cos(sin(T2)) - 2.74T1mod(4.29 + 0.37T2, 0.37)

[0185] Y e = 0.556 + 0.305T1 + 0.085T2 - 0.151T4 + 0.33T1T4 + 0.122mod(25.2 - T2 - mod(T1, 0.305), -8.23)

[0186] The average absolute errors of these models on all data sets are shown in Table 1. Figure 13

[0187] The first three rows are the performances of the optimal models trained by the point-shooting mode data sets (M1R*) on all data sets. They perform well on the first three columns (i.e. M1R* data sets) with errors ranging from [3.2 6.4], while their performances on the last three columns (i.e. M2R* data sets) are significantly reduced with errors ranging from [9.2 15.9]. Therefore, the models trained by the point-shooting mode data suffer from overfitting and poor generalization, and are not suitable for the sweep-shooting mode data. The last three rows are the performances of the optimal models trained by the sweep-shooting mode data sets (M1R*) on all data sets. The errors of each model on all data sets remain around 7 degrees with a fluctuation range of no more than 2 degrees. Therefore, the models trained by the sweep-shooting mode data have strong generalization ability, good versatility, and an omnidirectional prediction error of less than 8 degrees.

[0188] ​Based on the principle of electrostatic induction, an array of sensing electrodes with posture and spatial distribution structure characteristics is designed by imitating compound eye electrode array; long-distance and non-contact measurement of electrostatic induction signal of flying charged body is realized by using high-sensitivity electrostatic induction signal conditioning amplifier circuit; a prototype test system is built based on electromagnetic launching device, and the electrostatic induction signal dataset of omnidirectional incoming charged body is obtained, the symbolic regression algorithm is trained, and the prediction model of incoming direction of charged body is obtained, and the omnidirectional prediction error of the optimal model is less than 8 degrees. The present application belongs to passive receiving test, and does not actively emit any form of energy (such as infrared, electromagnetic wave, ultrasonic wave, etc.), so the power consumption is very low. The system is simple, only four-way electrostatic induction signal amplifier and low-cost embedded single-chip microcomputer (Raspberry Pi, STM32 series MCU) are used, so the cost is low and the flexibility is high, and it is easy to integrate. The prediction model is a simple mathematical expression, and the calculation complexity is very low, and any embedded processor can run, and the model has strong universality. Therefore, the present application has the characteristics of low power consumption, low cost, low computing power requirement and high integration, and has high application value.

[0189] Finally, it should be noted that: the above only describes the preferred embodiments of the present application, and is not used to limit the present application, although the present application has been described in detail with reference to the foregoing embodiments, for those skilled in the art, the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the scope of protection of the claims of the present application.

Claims

1. A method for collecting and constructing a data set of electrostatic induction signals of a flying charged object, characterized in that: include: A charged target, an electrostatic field, and a hemispherical electrostatic induction electrode array. The detection electrode of the hemispherical electrostatic induction electrode array includes a cylindrical structure consisting of a sensing plate, an insulating substrate, and a shielding electrode. The sensing plate has a circular plate-shaped bottom end, which is used to sense the spatial electrostatic field and is connected to the signal input end of the electrostatic sensor. The insulating substrate is used to electrically separate the sensing plate from the shielding electrode. The shielding electrode with a bottom is connected to the power ground of the electrostatic sensor to block the influence of the lateral electrostatic field on the sensing plate and improve the spatial directivity of the sensing plate. The detection electrode obtains the current value of the charged target. The current value is amplified and recorded by the subsequent high-impedance electrostatic sensor as a signal, which is used to represent the electrostatic signal generated when the charged target passes through the detection electrode. The hemispherical electrostatic induction electrode array is distributed on a hemispherical shell-shaped support structure, and the central axis of each detection electrode points to the center of the sphere. The detection electrode whose central axis is perpendicular to the bottom surface of the hemisphere is defined as the normal electrode, and its inclination angle is 0°; the angle between the central axis of the other detection electrodes and the central axis of the normal electrode is defined as its inclination angle, and four detection electrodes are equidistantly arranged on the circumference of the circle with an inclination angle of 42° to detect the electrostatic field signal of the flying charged body, including: a central point shooting mode experiment and an area scanning mode experiment. For the central point shooting mode experiment, the electrostatic test system is located in the center of the experimental site, and the flying charged body launch system is aimed at the center of the compound eye electrode to launch, so that the flying charged body passes through the airspace directly above it, and records four electrostatic induction signals; then the position of the launch system is changed at intervals of 9 degrees, traversing the experimental site once, and the experimental results are recorded 3 times at each position, and a total of 120 sets of data are collected; for the area scanning mode experiment, when launching at each position, it is aimed at the left, center, and right side of the test system respectively, and launched 3 times in each direction, and a total of 360 sets of data are recorded; A multi-channel digital oscilloscope is used to synchronously record four-way electrostatic induction signals. The collected electrostatic induction signals are filtered and de-skewed, and the flight trajectory information is mined using the time difference of each signal peak moment to obtain a time eigenvalue dataset.

2. The method for collecting and constructing a flying charged body electrostatic induction signal data set according to claim 1, characterized in that: The data cleaning operation of filtering and de-offsetting the collected electrostatic induction signal includes: the filtering operation uses MATLAB's low-pass digital filter function lowpass() to perform noise reduction processing on the recorded data, with the cutoff frequency parameter set to 10 Hz, the steepness parameter set to 0.9999, and the stopband attenuation parameter set to 180 dB; the de-offset operation uses MATLAB's detrend() function, setting the de-offset mode to constant, that is, subtracting the DC offset from the original data.

3. The method for collecting and constructing a flying charged body electrostatic induction signal data set according to claim 1, characterized in that: The method of mining flight trajectory information using the time difference of each signal peak moment to obtain a time characteristic value data set includes: determining the peak values ​​of the four electrostatic induction signals, setting the minimum value of the four peak moments as the starting moment, and the maximum value as the ending moment, defining the interval between the starting moment and the ending moment as the normalized duration of the test, and dividing the time difference between the four peak moments and the starting moment by the normalized duration to obtain the normalized time characteristic value, which is recorded as the normalized peak moment. The calculation method is shown in the formula; 4. The method for collecting and constructing a flying charged body electrostatic induction signal data set according to claim 1, characterized in that: The detection electrode obtains the current value of the charged target including: A Cartesian coordinate system XOY is established on the horizontal ground, and the unit vectors of the three coordinate axes are and Assume that the radius of the detection electrode is r, and the center of its upper surface is defined as the origin of the space coordinate system O, and its normal vector is is the vector passing through the origin O and pointing vertically to the outside of the circular plate, where α is and The angle between The projection on the XOY plane is The amount of static charge carried by the charged target is Q(t), which will change during the flight. The dotted line is its motion trajectory, and the velocity vector is is the tangent vector of the motion trajectory, and the coordinates of the spatial position are (x p (t),y p (t),z p (t)), the unit vector of its position coordinate is and The angle between them is δ, and The angle is The angle between the projection on XOY and the X axis is γ; The image charge method is used to calculate the field strength component of the target electric field in the normal direction of the detection electrode: Where ε0 is the dielectric constant of vacuum, ε r is the dielectric constant of the space medium; from formula (1), it can be concluded that when δ=Nπ, (N=0,1,2...), E n The maximum value, that is, when and When the detection electrodes are located on the same straight line and the normal direction faces the charged target, E n The maximum value; Let α, β∈(0,π / 2) and and Located in the same quadrant, then: Substituting equation (2) into equation (1), we can obtain the field intensity component in the normal direction of the detection electrode: set up but Assume that the charged target is located at position Then formula (4) can be simplified as: When α, β∈(0,π / 2) are both pi / 4, the normal electric field strength is the largest. Then, according to Gauss's theorem, the induced charge density σ on the detection electrode surface is s for: Assuming that the charge Q0 carried by the charged target remains unchanged and the detection electrode posture is parallel to the horizontal plane, that is, α = 0, then Equation (6) is simplified to: When x p (t)=0,y p When (t) = 0, σ s The maximum value; Assume that the distance from the projection of the charged target on the XOY plane to the coordinate origin is d p (t), i.e. Take d p (t)=0,z p σ when (t)=Δ s The value is a normalized quantity, Δ is the unit length; when z p When (t)≥6Δ, σ s In 0≤d p The value of (t)≤Δ in the circular region is approximately constant; according to Gauss's theorem, the total amount of charge induced on the detection electrode Q s for: Therefore, when the flying height of the charged target is greater than 6 times the radius r of the detection electrode, the charge density induced on the electrode surface can be considered to be constant; therefore, Equation (10) can be simplified to: According to the law of conservation of charge, the current flowing out of the detection electrode is the change in the amount of induced charge on the detection electrode, that is: This current value is amplified and recorded by the subsequent high-impedance electrostatic sensor, and is used to characterize the electrostatic signal generated when a charged target passes through the detection electrode.

5. The method for collecting and constructing a flying charged body electrostatic induction signal data set according to claim 4, characterized in that: The detection electrode obtains the current value of the charged target and further includes obtaining the signal waveform: Assuming that the detection electrode is parallel to the horizontal plane, the surface charge of the detection electrode is simplified to: Assume that the components of the charged target velocity V along the coordinate axis are V x , V y , V z ,So Assume that the charge of the charged target is constant at Q0, the flight speed is constant at V0, the speed direction and the flight trajectory are parallel to the x-axis, and it flies through the YOZ plane at t=0, that is, x p (0) = 0, formula (15) can be simplified to: The distance between the charged target and the detection electrode is D p (t), that is but: Substituting into formula (16), we get Therefore, when t = 0, the value of formula (19) is the largest, which is recorded as Then, formula (19) can be abbreviated as: From formula (21), the waveform shape of the induced charge is given by Determine, let it be the waveform time scale coefficient τ: Then formula (21) can be abbreviated as: Therefore, the waveform of the induced charge on the detection electrode is determined by τ, which is affected by the flight speed V(t) of the charged target and the shortest distance D from the target to the electrode. p (t) and the influence of the two parameters are inversely proportional. Then, we can substitute Equation (23) into the induced current equation (13), and we get

Images