Method and device for measuring electrical parameters of a building wall

Abstract

The application discloses a kind of electrical parameter measurement methods of building wall, comprising the following steps: step 1, obtain reflection coefficient: step 2, electrical parameter inversion based on PSO algorithm.The measurement method disclosed by the application has high measurement accuracy and small relative error.The measurement device disclosed by the application cooperates with the measurement method of the application, and has high measurement efficiency.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of electrical parameter inversion research, and particularly relates to a method and device for measuring electrical parameters of a building wall based on a particle swarm algorithm. BACKGROUND

[0002] In 1987, Cullen et al. disclosed a free-space method for obtaining electrical parameters of a building wall in A new free-wave method for ferrite measurement at millimeter wavelengths. The method is a free-space inversion method based on the Fresnel reflection law, which uses a transmitting antenna, a receiving antenna and a vector network analyzer to measure the amplitude and phase of the material scattering parameters, and then inverses the dielectric constant. However, the accuracy of the inversion result is reduced if the sample size is too small in the measurement process of the free-space wave method, because the incident electromagnetic wave irradiated to the sample edge will produce multiple diffraction and scattering. When measuring the electromagnetic parameters of indoor building materials and constructing an indoor electromagnetic environment model, the measurement in the real indoor environment using the free-space wave method is affected by many environmental factors due to the extremely complex indoor environment. SUMMARY

[0003] The technical problem to be solved by the present application is to provide a method and device for inversing electrical parameters of a building wall based on a particle swarm algorithm combined with a super-wideband measurement system correction.

[0004] The present application adopts the following technical solutions:

[0005] A method for measuring electrical parameters of a building wall, which is improved in that it comprises the following steps:

[0006] Step 1: Obtain the reflection coefficient:

[0007] Place a metal plate on the wall surface, measure the reflected echo of the metal plate, and take the reflected echo signal strength of the metal plate as the incident wave strength in calculation;

[0008] Remove the metal plate on the wall surface, and record the reflection signal, which is the reflected echo signal strength of the building;

[0009] Lay an absorbing material on the wall surface for testing, or find a substitute site that can shield the reflected echo of the tested building and does not affect the reflection effect of other background environments, and measure the corresponding reflected echo signal strength as the actual background echo signal strength;

[0010] Separate the obtained time-domain signal through time windowing and convert it to the frequency domain, and obtain the reflection coefficient of the building to be tested through the following calculation:

[0011] Based on the total reflection characteristics of the metal plate, the reflection echo of the metal plate is taken as the actual incident wave; the measured echo without the metal plate is taken as the actual reflection echo; the background echo without the building reflection effect is taken as the actual background noise; and then the reflection coefficient R of the building is obtained by conversion according to the following formula n :

[0012]

[0013] In the above formula, f n is the measured frequency, R n is the reflection coefficient at the frequency f n ; F rn is the spectral density of the received pulse at the frequency f n ; F tn is the spectral density of the transmitting antenna pulse at the frequency f n ; B n is the spectral density of the background reflection pulse signal at the frequency f n ; R is the distance from the transmitting and receiving antenna to the measured object; G r is the gain of the receiving antenna; G t is the gain of the transmitting antenna; and C is the speed of light.

[0014] Step 2: Electrical parameter inversion based on PSO algorithm

[0015] Searches are respectively performed in the given ranges of the conductivity and the dielectric constant, and the search equation of the PSO algorithm for the dielectric constant is as follows:

[0016]

[0017] The search equation of the PSO algorithm for the conductivity is as follows:

[0018]

[0019] In the above formula, k indicates the kth generation, k+1 is the generation to be solved, and is used to complete the algorithm iteration solution, v i is the current particle velocity, w is the inertia factor of the particle, c1 and c2 are learning factors, pbest i is the individual extreme value, gbest i is the group extreme value, x i represents the current position of the particle, σ is the conductivity, ε r is the dielectric constant, and ran is a random factor for initializing the velocity of the particle by using the rand function, and the value range is 0-1.

[0020] The mean square error value f erAs a target function, and in order to screen the optimal position of each generation particle, the target function is expressed as:

[0021]

[0022] In the above formula, R m (θ i ,f) is the modulus value of the measured reflection coefficient corresponding to different incident angles and frequencies, R c (θ i ,f) is the modulus value of the theoretical reflection coefficient corresponding to different incident angles and frequencies, θ i represents the change of the incident angle, f represents the change of the frequency, and N represents the total number of tests;

[0023] First, a random function rand is used to randomly generate each particle to form a particle swarm set, and after traversing the particle swarm set, the optimal solution of the set is obtained by bringing it into the target function of formula (5). If the constraint condition is met, it is the solution to be found. If not, continue to update the speed and position of each particle in the next round through formula (3) and formula (4), and calculate the target function again according to formula (5). The historical optimal position of each particle is updated, and the optimal solution of the entire particle swarm in a series of update iteration processes is obtained. Continue to judge whether the final parameter condition is met. If not, continue to update the parameters.

[0024] The improvement of the building wall electrical parameter measuring device is that: a computer is used to generate an ultra-wideband signal pulse source, a transmitting antenna Tx for transmitting signals, a receiving antenna Rx for receiving signals, a filter for filtering noise, an amplifier for amplifying signals, and a sampling machine for sampling and processing signals. The computer is electrically connected with the pulse source and the sampling machine, the pulse source is electrically connected with the transmitting antenna Tx, and the sampling machine is electrically connected with the receiving antenna Rx through the amplifier and the filter in sequence.

[0025] The beneficial effects of the present application are:

[0026] The measuring method disclosed by the present application has high measuring precision and small relative error. The measuring device disclosed by the present application cooperates with the measuring method to have high measuring efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 is a flowchart of the measuring method disclosed by the present application;

[0028] Figure 2 is a composition block diagram of the measuring device disclosed by the present application;

[0029] Figure 3 is an execution flowchart of the particle swarm algorithm;

[0030] Figure 4is a connection diagram of experimental measuring equipment;

[0031] Figure 5 is a marble test sample picture;

[0032] Figure 6 is a marble test real scene picture;

[0033] Figure 7 is a time domain signal display picture of a real marble plate;

[0034] Figure 8 is a pulse frequency domain waveform and a reflection coefficient frequency domain waveform picture of a measured signal when the incident angle is 3.86°;

[0035] Figure 9 is a pulse frequency domain waveform and a reflection coefficient frequency domain waveform picture of a measured signal when the incident angle is 1.89°;

[0036] Figure 10 is a pulse frequency domain waveform and a reflection coefficient frequency domain waveform picture of a measured signal when the incident angle is 16.73°;

[0037] Figure 11 is a marble test result display picture;

[0038] Figure 12 is a frequency domain waveform picture after time window separation when the incident angle is 3.86°;

[0039] Figure 13 is a frequency domain waveform picture after time window separation when the incident angle is 1.98°;

[0040] Figure 14 is a frequency domain waveform picture after time window separation when the incident angle is 16.73°;

[0041] Figure 15 is an inversion result picture when the relative dielectric constant is 7.0 and the conductivity is 0.01 S / m under 500 MHz;

[0042] Figure 16 is an inversion result picture when the relative dielectric constant is 7.0 and the conductivity is 0.01 S / m under 1 GHz;

[0043] Figure 17 is an inversion result picture when the relative dielectric constant is 7.0 and the conductivity is 0.01 S / m under 5 GHz. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical scheme and advantages of the present application more clear, the present application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application, and are not used to limit the present application.

[0045] The application discloses a method for measuring electrical parameters of a building wall, and relates to the field of building wall measurement. Figure 1 As shown in the figure, firstly, a time window separation algorithm is used to acquire the reflection coefficient of the building wall by using an ultra-wideband time domain measurement system, and then a particle swarm optimization algorithm is used to acquire the final electrical parameter attribute based on a ray tracing algorithm according to the reflection coefficient.

[0046] Step 1: Acquiring the reflection coefficient

[0047] Figure 2 A schematic diagram of a measurement device using a time domain ultra-wideband measurement system to measure the target reflection coefficient is given. The experimental system used in the application is an ultra-wideband time domain measurement system, which is composed of software and hardware. The system software is mainly used to display real-time signals, set measurement parameters, correct and process signals through programming, and display, record and output measurement results. The hardware part is as shown in the figure. Figure 2 The computer controls the whole system, the pulse source is used to generate an ultra-wideband signal, the transmitting antenna Tx is used to transmit the signal, the receiving antenna Rx is used to receive the signal, the filter is used to filter out noise, the amplifier is used to amplify the signal, and the sampler is used to sample and process the signal. The computer is electrically connected with the pulse source and the sampler, the pulse source is electrically connected with the transmitting antenna Tx, and the sampler is electrically connected with the receiving antenna Rx through the amplifier and the filter in sequence. Finally, the measured time domain signal is composed of three parts, one is the coupling signal echo between the transmitting and receiving antennas, the second is the reflection signal echo of the target to be measured, and the third is the reflection signal echo of the background (such as walls, floors, ceilings, other objects, people, etc.).

[0048] The working principle of the measurement device is that the transmitting antenna transmits the time domain pulse excitation provided by the pulse source, the target generates a scattering field under the irradiation of the incident plane wave, the receiving antenna receives the target scattering field and samples through the equivalent sampler synchronized with the signal source, the real-time scattering signal of the target is obtained, and the electromagnetic characteristics of the target can be obtained after data processing and analysis on the computer. For the pulse provided by the time domain pulse source, the higher the amplitude and the narrower the width of the pulse, the wider the frequency spectrum covered after Fourier transform. At the same time, the frequency spectrum of the incident wave is related to the incident antenna. In order to perform ultra-wideband measurement, the transmitting antenna and the receiving antenna are both ultra-wideband antennas.

[0049] The basic principle of measurement is: based on the full reflection characteristics of the metal plate, the reflection echo of the metal plate is regarded as the actual incident wave; the measurement echo without the metal plate is regarded as the actual reflection echo; the background echo without the reflection effect of the building is regarded as the actual background noise N i ; and then the reflection coefficient R n of the building is obtained by conversion according to the following formula:

[0050]

[0051] In the above formula, f n is the measured frequency, R n is the reflection coefficient at frequency f n ; F rn is the spectral density of the received pulse at frequency f n ; F tn is the spectral density of the transmitted pulse of the transmitting antenna at frequency f n ; B n is the spectral density of the background reflected pulse signal at frequency f n ; R is the distance from the transceiver antenna to the measured object; G r is the gain of the receiving antenna; G t is the gain of the transmitting antenna; C is the speed of light;

[0052] When R n = 1, this formula degenerates into the reflection formula of an infinite conductor plate. This is also the reason why the echo signal of a metal plate is used as the equivalent incident signal.

[0053] In summary, to obtain the reflection coefficient of the target, first measure the time domain signal of the target reflected echo at a given frequency, and after obtaining the reflected echo time domain signal, convert the time domain signal to the frequency domain. Similarly, measure the time domain signals of the background noise and the metal plate, and then subtract the background noise from the time domain signals of the incident wave and the reflected wave and take the ratio, which is the reflection coefficient of the building. The specific steps for extracting the reflection coefficient of the building wall are as follows:

[0054] Place a metal plate on the wall surface and measure the reflected echo of the metal plate. The reflected echo signal strength of the metal plate is taken as the incident wave strength E TX ;

[0055] Remove the metal plate from the wall surface and record the reflected signal, which is the reflected echo signal strength E RX of the building;

[0056] Test the wall surface by laying wave-absorbing materials, or find an alternative site that can shield the test building's reflected echo and not affect the reflection effect of other background environments. Measure the corresponding reflected echo signal strength as the actual background echo signal strength E treasureRE ;

[0057] Convert the obtained time domain signal to the frequency domain after separation by time windowing, and through correlation calculation, obtain the reflection coefficient R i of the building to be tested.

[0058] Step 2, electrical parameter inversion based on PSO algorithm:

[0059] Based on the actual electrical parameter inversion requirements, the measured results are inversed by using two-dimensional particle swarm algorithm, so as to obtain the electrical parameter results of the medium to be measured. Figure 3 The specific process of the particle swarm algorithm is given, the initial position and speed of the particles are random, and with the increase of the number of times, all the particles will approach the optimal solution.

[0060] The standard formula of the particle swarm algorithm is:

[0061]

[0062] In the above formula, v i is the current particle speed; w is the inertia factor of the particle, which adjusts the search ability of the solution space, a larger inertia factor will promote global search, a lower inertia factor will promote local search, and its value is between 0 and 1, and in general application, the adaptive value method is used, that is, w=0.9 at the beginning, so that the global optimization ability of the algorithm is stronger, and with the deepening of iteration, the parameter w is decreased, so that the algorithm has strong local optimization ability, and when the iteration is finished, w=0.1. c1 and c2 are learning factors for adjusting the search step, which are generally set to 0.89445. pbest i and gbest i are the individual extreme value and group extreme value respectively. x i represents the current position of the particle. v i represents the current moving step of the particle. Generally, v i is limited to prevent the step from being too large and missing the optimal solution; generally, the position of x i is also limited, because it only makes sense to solve the operation within a certain range. And a suitable solution interval will speed up the result and save the optimization algorithm time.

[0063] When the electrical parameters (relative dielectric constant and conductivity) of the building are inversed, the particle swarm equation is two, one is the equation about the conductivity σ, and the other is the equation about the relative dielectric constant ε r of the medium. The search is carried out in the given range of conductivity and dielectric constant, and the equations of the two searches are:

[0064]

[0065] The above formula (3) is the search equation of the PSO algorithm for the relative dielectric constant.

[0066]

[0067] The above formula (4) is the search equation of the PSO algorithm for the conductivity.

[0068] Initially, the position and velocity of each particle in a given particle swarm are set. A random number is generated within a given range to represent the initial position, and an initial velocity is also randomly generated within a given range. Then, using the initial values ​​and the PSO algorithm equations, a finite number of iterations are performed, continuously judging and updating the particle's position and velocity. Finally, after the iterations are complete, the optimal particle position and the optimal objective function value are returned. This invention treats the studied building as a homogeneous dielectric material; generally, the relative permittivity and conductivity of a building range from 1 to ε. r ≤15, 0≤σ≤10 -1 Therefore, in actual iterative optimization calculations, this range is used to find the optimal value.

[0069] To minimize the error between the obtained inversion results and the actual results, this invention uses the root mean square error f between the measured reflection coefficient modulus of the building and the theoretically calculated reflection coefficient modulus. er The objective function, used to select the optimal position for each generation of particles, is expressed as:

[0070]

[0071] In the above formula, R m (θ i f) represents the magnitude of the measured reflection coefficient for different incident angles and frequencies, R c (θ i f) represents the magnitude of the theoretical reflection coefficient for different incident angles and frequencies, θ i The angle of incidence represents the change, and f represents the frequency of the change, which is crucial when converting time-domain measurements to frequency-domain results. N represents the total number of tests.

[0072] First, a random function rand is used to randomly generate each particle to form a particle swarm set. After traversing the particle swarm set, the optimal solution of the set is obtained by substituting it into the objective function of formula (5). It is then determined whether the constraint conditions are met. If they are met, it is the solution to be found. If not, the velocity and position of each particle in the next round are updated by formula (3) and formula (4). The objective function is calculated again according to formula (5), and the historical optimal position of each particle is updated to obtain the optimal solution of the entire particle swarm in a series of update iterations. It is then determined whether the conditions of the final parameters are met. If not, the parameters are updated.

[0073] like Figure 4As shown, the time domain ultra-wideband measurement system includes a signal source for generating an ultra-wideband signal; an antenna for transmitting and receiving signals, the measurement range of the antenna is 0-40GHz; a data sampling system integrated in the computer software for data acquisition; a connecting line for connecting various devices and transmitting data; a turntable for adjusting the placement position of the object to be measured to prevent measurement errors caused by improper placement of the object to be measured; a material to be measured (acrylic plastic plate, wood board, marble block, etc.); a metal plate for measuring the initial incident field intensity, and the size of the metal plate should be as large as possible so that the incident field result is more accurate. Matters needing attention: the equipment is preheated for 1 hour in advance to ensure the accuracy of the measurement data.

[0074] Basic parameter settings for measurement:

[0075] (1) Related parameters of the transmitting and receiving antenna: the transmitting and receiving antenna uses a 2GHz-26GHz ultra-wideband double-ridge horn antenna.

[0076] (2) Height of the transmitting and receiving antenna: keep the same height as the center of the test sample to ensure the correct receiving strength.

[0077] (3) Distance between the transmitting and receiving antenna and the measurement sample: the distance between the apertures of the transmitting and receiving antenna is constantly changing according to the test requirements, and the ultimate purpose of changing the distance is to change the incident angle of the incident electromagnetic wave.

[0078] (4) The frequency range of the time domain system is 0-40GHz, but under high frequency conditions, the propagation distance of the electric wave is short; under low frequency conditions, the diffraction phenomenon is more obvious, therefore, when selecting the final result, the frequency interval of 2GHz-20GHz is used, which can ensure the accuracy of the measurement result.

[0079] Take the measurement process of marble as an example to illustrate the method for measuring the reflection coefficient of the medium, as shown in Figure 5 The test sample is a square marble block with a length of 20cm, a width of 20cm, and a height of 4.55cm.

[0080] The test scene is shown in Figure 6 The marble is the test target, which is placed at the same height as the transmitting and receiving antenna, and a laser pen is used to try to make the light emitted from the transmitting and receiving antenna exactly in the center of the marble brick, so as to ensure that the energy emitted by the transmitting antenna is as much as possible to be received by the receiving antenna after passing through the target. The same measurement position should be maintained for other test samples.

[0081] The following briefly describes the actual measurement steps of the marble sample:

[0082] (1) Connect the equipment and turn it on for more than one hour of preheating;

[0083] (2) Put the marble on the turntable, adjust the position of the turntable, find the maximum value of the time domain result on the display, stop the rotation of the turntable, and mark the position of the front of the marble at this time;

[0084] (3) After determining the position of the marble, sample the marble reflection echo data and record them;

[0085] (4) Place the metal plate in front of the marble, ensure that the front surface position of the metal plate is the same as the marked position of the metal plate, then sample the echo of the metal plate and record the data;

[0086] (5) Remove the marble and metal plate on the turntable, sample and record the test background data;

[0087] (6) Repeat the above test steps one to two times to ensure the accuracy of the measurement results at the same incident angle;

[0088] (7) Change the position of the transmitting and receiving antenna, and repeat the above steps again to realize the collection of multiple sets of measurement data;

[0089] (8) Turn off the power and disassemble the equipment;

[0090] (9) Process the data to obtain the results.

[0091] The measured time domain waveform results are shown in Figure 7 , (a) shows the time domain distribution of the measured background signal, it can be seen that the distribution of the background signal is irregular and chaotic, (b) shows the time domain distribution of the received echo signal after the test marble is added, it can be clearly seen that there are two wave peaks, which proves that the echo signals of the front and back surfaces of the marble plate are received; (c) shows the time domain distribution of the echo signal intensity when the metal plate is placed behind the marble and closely attached to the back surface of the marble, it can be seen that there are three wave peak signals, which indicates that the echo signal of the metal plate is also detected; (d) shows the time domain intensity distribution of the received echo signal when the metal plate is placed in front of the marble and closely attached to the front surface of the marble; (e) shows all the signals in the same time distribution, it can be seen that the intensity of the background signal is relatively weak, the reflection echo signals of the marble and the metal plate are relatively strong, and the intensity change of the echo signal in the time domain is very obvious, which provides guarantee for the subsequent data processing.

[0092] When the height of the transmitting and receiving antenna is 1.1 m, the aperture distance of the transmitting and receiving antenna is 0.18 m, the distance between the antenna and the test sample is 1.334 m, and the incident angle is 3.86°, the pulse frequency domain waveform and the reflection coefficient frequency domain waveform change are shown in Figure 8 .

[0093] The height of the transceiving antenna is 1.1 m, the distance between the transceiving antenna and the aperture is 0.18 m, the distance between the antenna and the test sample is 2.725 m, and the change of the pulse frequency domain waveform and the reflection coefficient frequency domain waveform when the incident angle is 1.89° is as shown in Figure 9 .

[0094] The height of the transceiving antenna is 1.1 m, the distance between the transceiving antenna and the aperture is 1.564 m, the distance between the antenna and the test sample is 2.833 m, and the change of the pulse frequency domain waveform and the reflection coefficient frequency domain waveform when the incident angle is 16.73° is as shown in Figure 10 .

[0095] The above figures respectively show the change of the time domain signal intensity and the frequency domain reflection coefficient measured at different incident angles. Take Figure 8 for example, it can be seen that the upper line in the (a) figure display represents the distribution of the signal intensity of the echo signal of the metal plate minus the background intensity in the time domain. The lower line in (a) represents the actual target echo intensity in the time domain after the echo signal intensity of the test target minus the background echo signal intensity. (b) shows the change of the reflection coefficient in the frequency domain. By comparing Figure 8 , Figure 9 , Figure 10 , it can be found that Figure 8 , because the distance between the antenna and the test target is relatively close, the peak value of the pulse spectrum in the frequency domain is higher than that of Figure 9 and Figure 10 . By comparing Figure 9 and Figure 10 , it can be seen that when the incident angle increases, the peak value of the spectrum intensity will decrease. In addition, it can be seen that although the incident angle and the distance between the antenna and the target of Figure 8 , Figure 9 , Figure 10 are different, the change of the pulse spectrum of the incident wave and the reflected wave is similar, and the change trend of the reflection coefficient in the frequency domain is also the same, which provides a precision guarantee for multi-angle measurement.

[0096] As the incident frequency changes, the reflection coefficient of the marble slab also changes continuously, which is the same as the frequency domain variation law of the reflection coefficient of layered media. This indirectly confirms the accuracy of the experimental measurement results and lays the foundation for subsequent data processing and electrical parameter inversion. However, in the traditional free-space method for measuring the electrical parameters of the test medium, the thickness of the test medium is known, so the previously measured reflection coefficient results can be substituted into the objective function to directly invert the electrical parameters of the building. But in actual situations, the thickness of the building is unknown, so this method cannot be used. In this case, the echo separation method must be used to solve for the electrical parameters. The so-called separation means isolating the overall echo of the building in the time domain according to the different return times, and only analyzing the reflection of the first reflection from the front surface of the building. At this time, the reflected echo is equivalent to the case of electromagnetic waves obliquely incident into half space and reflected, and the electrical parameters of the reflecting surface can be inverted. It is worth mentioning that this method is only applicable to homogeneous media. The marble slab tested in this application can be regarded as a homogeneous medium because of its regular shape and uniform distribution.

[0097] The following comparison of the time-domain test results of marble and plastic board illustrates the principle of the time-window separation method for obtaining the reflection coefficient used in this application.

[0098] by Figure 11 Taking a marble sample as an example, this section illustrates the steps and principles of the time-window separation method. First, the time-domain echo signal of the marble target is measured. Then, a metal plate is placed behind the target plate, and the echo signals of the marble and the metal plate are measured. This measurement is performed to verify the correctness of the time-window separation method; therefore, this step is unnecessary when actually measuring building walls. Figure 11 As shown, the solid lines represent the time-domain echo measured from the marble, while the dotted lines represent the echo signal measured after placing a metal plate directly behind the marble.

[0099] exist Figure 11The three obvious peak echo signals can be seen in the rectangular frame. Echo 1 is the echo signal produced by the interface when the electromagnetic wave is incident on the marble from the air. The degree of coincidence between the two sets of measurement data represented by the solid line and the dotted line is very high, indicating that the marble position did not change during the two measurements, which is one of the key factors to ensure the accuracy of the comparison results. In echo 2, the dotted line is the echo signal from the interface between the marble and the metal plate, and the solid line is the reflected echo signal from the interface between the marble and the metal plate. In echo 3, the solid line represents the echo signal of the electromagnetic wave reflected from the metal plate to the marble. After the reflection of the contact surface between the front surface of the marble and the air, the reflected signal penetrates the contact surface between the back surface of the marble and the metal plate, and then returns to the receiving antenna (i.e. the second reflection echo signal produced by the interface between the marble and the metal plate).

[0100] Based on the measured results, the reflection signal of the electromagnetic wave incident on the surface of the building from the air can be separated according to the time and intensity variation. From Figure 11 As can be seen from the figure, the energy before the rising edge of the second echo signal and after the rising edge of the first echo signal is relatively concentrated, so the intensity of the first incident reflection echo is determined by the rising edges of the two echoes. By separating the front surface reflection echo signal through the time window between the two rising edges, the reflection coefficient of the front surface of the medium plate can be obtained.

[0101] Although this processing method has errors, it is still of reference value. In terms of building measurement, because the thickness of the building itself is relatively large, there is a relatively large time interval between the first echo and the second echo, so it is not necessary to place a metal plate behind the building to obtain the front surface reflection echo intensity. It is only necessary to find the waveform distribution between the relatively flat place after the falling edge of the first reflection echo and the beginning of the rising edge of the second echo, which is the front surface reflection echo of the building. Therefore, this method is more suitable for relatively thick medium plates, and is not suitable for relatively thin measurements, such as Figure 11 As shown in the time domain echo variation diagram of a plastic plate with a thickness of about 1 cm, it can be seen that the echoes of the front and back surfaces are difficult to distinguish in time. The main reason is the accuracy of the measurement equipment. In theory, as long as the medium to be measured has a thickness, the echo signals of the front and back surfaces can be distinguished. However, this method is completely applicable to the study of buildings, because the thickness of the building is very thick, and the obtained echo signal is either the echo signal of the front and back walls, or because the building is too thick, there is only a front surface echo signal.

[0102] It should be noted that, although the time difference of the echo reflected by the front and back surfaces of the wall can be seen in the time domain, the length of the time difference may be different from the actual propagation time, so it is not applicable to obtain more accurate building thickness, but the energy represented by the intensity itself can be used.

[0103] Figure 12 、 13 , 14 are the pulse spectrum waveforms and the reflection coefficient frequency domain waveforms of the echo reflected by the front surface after time window separation, respectively, obtained by vertically polarized electromagnetic waves with incident angles of 3.86°, 1.96°, and 16.73°, respectively, incident on marble. Compared with the results without time window separation, the changes are still very obvious, especially the changes in the reflection coefficient frequency domain waveform.

[0104] In actual test conditions, because of the diffraction of low-frequency signals and the strong attenuation of high-frequency signals, the results of the reflected echoes of the low-frequency part and the high-frequency part will have errors. When the frequency is less than 2 GHz and greater than 20 GHz, the waveform of the reflection coefficient changes chaotically, indicating that the diffraction of low-frequency signals and the strong attenuation of high-frequency signals have affected the results. The reflection coefficient changes relatively gently with frequency in the frequency range of 5 GHz to 15 GHz.

[0105] Next, several typical frequency bands of electromagnetic waves, such as 500 MHz, 1 GHz, and 5 GHz, are selected for testing to study the changes in the inversion results of the optimal particles in each generation of the particle swarm with the number of iterations. In order to avoid errors in the results caused by the random selection of initialization points in the particle swarm algorithm, three tests are performed for each inversion result. Because the polarization of electromagnetic waves does not affect the final inversion result, all simulations in this application use vertically polarized electromagnetic waves. The number of iterations is set to 200, the particle swarm size is 1000, the inertia factor is w = 0.4, and the acceleration factors are c1 = 0.89445 and c2 = 0.89445. The theoretically calculated reflection coefficients under different incident angles are used as input parameters to verify the correctness of the inversion of the electrical parameters.

[0106] Figure 15 、 16 , 17 are different inversion results when the incident frequency of electromagnetic waves is 500 MHz, 1 GHz, and 5 GHz, respectively, and the relative permittivity of the medium to be inverted is 7.0 and the conductivity is 0.01 S / m. It can be seen that the inversion results of the conductivity and the relative permittivity are correct.

[0107] The time domain results of the measurement data with incident angles of 3.86°, 1.96°, and 16.73° are combined with the reflection coefficient frequency domain results obtained after time window separation, and the electrical parameter inversion results obtained after fitting the frequency domain results with the particle swarm algorithm.

[0108] Table 1 Inversion results of electrical parameters

[0109]

[0110] The above table is the inversion results of electrical parameters, and the inversion results of six inertia factors are averaged to obtain the final electrical parameters under different incident angles. It can be seen that the final results are the relative dielectric constant of 5.37-5.54 and the conductivity of 0.00429-0.00515 S / m. In order to verify the correctness of the results, the reflection coefficient frequency domain value obtained can be compared with the measured results.

[0111] Table 2 Relative dielectric constant of common medium

[0112]

[0113] From the above table, it can be seen that the relative dielectric constant of marble is 6.2, and the relative dielectric constant of marble obtained in this experiment under the incident angles of 3.86°, 1.96° and 16.73° is 5.545, 5.395 and 5.370 respectively, and the average value is 5.44. Comparing the actual relative dielectric constant of marble 6.2 with the inversion result 5.44, the relative error is 12.26%, and the relative error of medium electrical property is ≤20%.

Claims

1. A method of measuring electrical parameters of a building wall, characterized in that, Comprising the following steps: Step 1, obtaining the reflection coefficient: Place a metal plate on the wall, measure the reflected echo of the metal plate, and take the reflected echo signal strength of the metal plate as the incident wave strength for calculation; Remove the metal plate on the wall and record the reflected signal, which is the reflected echo signal strength of the building; Lay a wave-absorbing material on the wall for testing, or find a substitute site that can shield the reflected echo of the test building and does not affect the reflection effect of other background environment, and measure the corresponding reflected echo signal strength as the actual background echo signal strength; Convert the obtained time-domain signal to the frequency domain after separation by time window, and obtain the reflection coefficient of the building under test by the following calculation: Based on the total reflection characteristics of the metal plate, the reflected echo of the metal plate is taken as the actual incident wave, the measured echo after removing the metal plate is taken as the actual reflected echo, and the background echo after removing the reflection effect of the building is taken as the actual background noise; The reflectance R of the building is then converted by the following equation n : In the above equations, f n is the measured frequency, R n is the reflection coefficient at frequency f n ; F rn is the spectral density of the received pulse at frequency f n ; F tn is the spectral density of the transmitted pulse of the transmitting antenna at frequency f n ; B n is the spectral density of the background reflected pulse signal at frequency f n ; R is the distance of the transceiver antenna to the measured object; G r is the gain of the receiving antenna; G t is the gain of the transmitting antenna; C is the speed of light. Step 2, electrical parameter inversion based on PSO algorithm: Search in the given range of conductivity and dielectric constant respectively, the search equation of PSO algorithm for dielectric constant is: The search equation of PSO algorithm for conductivity is: In the above formula, k refers to the kth generation, k+1 is the algebra to be solved, used to complete the algorithm iteration solution, v i is the current particle velocity, w is the inertia factor of the particle, c1, c2 are learning factors, pbest i is the individual extreme value, gbest i is the group extreme value, x i represents the current position of the particle, σ is the conductivity, ε r is the dielectric constant, and ran is a random factor attached to the particle initialization speed using the rand function, with a value range of 0-1. The mean square error value f of the modulus of the reflection coefficient of the actual building and the modulus of the reflection coefficient calculated theoretically er As the objective function, and to filter the optimal position of each generation of particles, the objective function is expressed as: In the above formula, R m (θ i ,f) is the measured reflection coefficient corresponding to different incident angles and frequencies, R c (θ i ,f) is the theoretical reflection coefficient corresponding to different incident angles and frequencies, θ i represents the varying incident angle, f represents the varying frequency, and N represents the total number of tests. First, a random function rand is used to randomly generate each particle to form a particle swarm set, after traversing the particle swarm set, the optimal solution of the set is obtained by bringing it into the objective function of formula (5), it is judged whether the constraint condition is met, if yes, it is the solution to be found, if not, the speed and position of each particle in the next round are updated through formula (3) and formula (4), the objective function is calculated again according to formula (5), the historical optimal position of each particle is updated, the optimal solution of the entire particle swarm in a series of update iteration processes is obtained, and it is judged whether the final parameter condition is met, if not, the parameter is continuously updated.

Images