A method and circuit structure for dual-band impedance matching of a while-drilling transducer based on genetic algorithm and Gauss-Newton iteration fusion

Abstract

This invention discloses a dual-band impedance matching method and circuit structure for drilling transducers based on a fusion of genetic algorithm and Gauss-Newton iteration. The method first establishes a multi-mode equivalent circuit model of the drilling transducer, consisting of a static capacitor and multiple parallel branches, with the number of branches corresponding to the number of resonant points of the transducer. Then, a dual-band impedance matching network is constructed, consisting of a Π-type network and a T-type network connected in series. By dividing the network into four grid regions, a voltage grid equation is established, and the port impedance function is derived. Finally, the parameters of the dual-band impedance matching network are calculated using an adaptive genetic algorithm fused with Gauss-Newton iteration, and a learning factor is introduced for adaptive optimization, effectively solving the problem that traditional genetic algorithms are prone to getting trapped in local optima, leading to global non-convergence. This invention also discloses the corresponding dual-band impedance matching circuit structure. This method can achieve broadband impedance matching in two target frequency bands simultaneously, improving the average transmit voltage response by 12dB, increasing the effective bandwidth by 1.4 times, and increasing active power by 1.8 times.

Description

Technical Field

[0001] This invention relates to the field of logging-while-drilling acoustic transducers, and particularly to a dual-band impedance matching method and circuit structure for logging-while-drilling transducers based on a fusion of genetic algorithms and Gauss-Newton iterations. Background Technology

[0002] Transducers used while drilling are key technical equipment in deep oil and gas exploration and development, and their performance directly determines the quality of logging while drilling. Most piezoelectric transducers used while drilling are capacitive loads, typically possessing multiple resonant frequencies and exhibiting significant impedance fluctuations. Their impedance characteristics are difficult to express using specific analytical functions, resulting in unclear frequency response patterns between the excitation signal and the matching network topology. Furthermore, conventional calculation methods often fail to achieve convergence when calculating multi-component parameters. These factors constitute a critical bottleneck limiting the transducer's ability to achieve efficient power output across dual frequency bands.

[0003] Impedance matching of transducers is mainly divided into two categories: acoustic matching and electrical matching. Regarding acoustic material matching of transducers, coupling materials are typically added to make the transducer's radiating surface impedance close to the dielectric impedance, thereby obtaining better radiation characteristics. However, this method increases the external size of the transducer, making it difficult to practically apply in drilling transducers.

[0004] Existing electrical impedance matching methods can usually only be designed for a single resonant frequency of the transducer, making it difficult to cover both high-frequency and low-frequency bands at the same time. Furthermore, traditional optimization algorithms (such as simple genetic algorithms) are prone to getting stuck in local optima when optimizing multiple component parameters, resulting in global non-convergence and failing to meet the application requirements of high-efficiency dual-band power output of drilling transducers. Summary of the Invention

[0005] The purpose of this invention is to propose a dual-band impedance matching method and circuit structure for drilling transducers based on the fusion of genetic algorithm and Gauss-Newton iteration, so as to solve the technical problems in the prior art that the impedance characteristics of drilling transducers are difficult to accurately characterize, dual-band impedance matching is difficult to achieve, and traditional optimization algorithms are prone to getting trapped in local optima.

[0006] To achieve the above objectives, this invention proposes a dual-band impedance matching method for drilling transducers based on a fusion of genetic algorithms and Gauss-Newton iterations, comprising the following steps: S1: Establish a multimode equivalent circuit model of the drilling transducer. The multimode equivalent circuit model includes a static capacitor. And N parallel RLC branches, each RLC branch being a resistor. ,inductance and capacitor Composed of series, where N is a positive integer; S2: The impedance data of the transducer at different frequency points is measured by an impedance analyzer. Based on the impedance data, the least squares method is used to perform nonlinear regression fitting on the actual impedance data of the transducer to obtain the parameter values ​​of each component in the multimode equivalent circuit model. S2: Construct a dual-frequency impedance matching network, which consists of a Π-type network and a T-type network connected in series. The Π-type network includes a first series inductor. and two first parallel capacitors and The T-type network includes a second series inductor. and the third series inductor and the second parallel capacitor ; S4: In a circuit consisting of a dual-frequency impedance matching network, a step-up transformer, and a multi-mode equivalent circuit model, four grid regions are defined, voltage grid equations are established for each region, and the impedance function of the entire port is derived. ; S5: Set the target frequency band, including the first target frequency band and the second target frequency band, and set the boundary conditions; S6: Parameter calculation of dual-frequency impedance matching network is performed by fusing adaptive genetic algorithm with Gauss-Newton iteration. New points in each iteration are introduced by a learning factor. Perform adaptive optimization; S7: Based on error Perform a grading judgment, when the error When the convergence condition is not met, the learning factor Make dynamic adjustments, return to iterative calculation to continue calculating the parameters of the dual-frequency impedance matching network until convergence; S8: Determine whether the current frequency band flatness meets the preset conditions. If it does, output the parameter values ​​of each component in the dual-frequency impedance matching network.

[0007] Furthermore, the number of N parallel RLC branches is 3, corresponding to the three resonant modes of the transducer.

[0008] Furthermore, the turns ratio n of the step-up transformer is set between 1:5 and 1:10, the primary side is connected to a dual-frequency impedance matching network, and the secondary side is connected to a multi-mode equivalent circuit model.

[0009] Furthermore, the boundary conditions include convergence thresholds for the real and imaginary parts of the impedance: ; In the formula, The internal resistance of the signal source, This is the preset convergence threshold.

[0010] Furthermore, learning factors The dynamic adjustment formula is: ; in, As a learning factor, The maximum number of iterations, The maximum value of the learning factor. This represents the minimum value of the learning factor.

[0011] Furthermore, the criteria for determining flatness are: .

[0012] This invention further proposes a dual-band impedance matching circuit structure for a drilling transducer, including a signal source section, a dual-band impedance matching network, a step-up transformer, and a transducer load: The signal source section is used to provide the excitation voltage. and internal resistance The dual-frequency impedance matching network is connected to the signal source section and includes a Π-type network and a T-type network connected in series. The Π-type network includes a first series inductor. and two first parallel capacitors , A T-type network consists of two series inductors. , and a second parallel capacitor ; The turns ratio n of the step-up transformer is 1:5 to 1:10. The primary side is connected to a dual-frequency impedance matching network, and the secondary side is connected to the transducer load. The transducer load is connected to the secondary side of the step-up transformer and includes a static capacitor. And N parallel RLC branches, each RLC branch is a resistor ,inductance and capacitor Composed of series connections.

[0013] Furthermore, the number of parallel RLC branches in the transducer load is three, corresponding to the three resonant modes of the transducer.

[0014] Compared with the prior art, the advantages of the present invention are: 1. This invention establishes a multi-mode equivalent circuit model of the drilling transducer, which can fully characterize the admittance characteristics of the actual composite transducer. The model consists of a static capacitor and multiple parallel RLC branches. The number of branches corresponds to the number of resonant points of the transducer, thus accurately reflecting the multi-mode characteristics of the transducer.

[0015] 2. The present invention uses a dual-frequency impedance matching network composed of a π-type network and a T-type network connected in series, which can simultaneously cover two operating frequency bands, the high-frequency band and the low-frequency band, to achieve dual-frequency broadband impedance matching. Compared with the traditional single L-type or π-type matching network, it has a wider bandwidth coverage and better matching flatness.

[0016] 3. This invention uses a fusion algorithm of genetic algorithm and Gauss-Newton iterative algorithm to calculate the parameters of dual-frequency impedance matching network, and introduces a learning factor for adaptive optimization. This effectively solves the problem that traditional genetic algorithm is prone to forming local optimal solutions, which leads to global non-convergence, and improves the convergence and stability of multi-component parameter calculation.

[0017] 4. Experimental results show that the dual-frequency broadband impedance matching network proposed in this invention can simultaneously achieve impedance matching of the transducer in two target frequency bands. Its average transmit voltage response is improved by 12dB, the effective bandwidth is increased by 1.4 times, and the active power is significantly improved by 1.8 times. It can effectively enhance the active power output and transmit voltage response of the transducer in the target frequency band, effectively enhance the acoustic field radiation characteristics of the underwater acoustic transducer, and enhance the detection range of the logging-while-drilling instrument. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the circuit topology of a dual-frequency impedance matching network parameter calculation model according to an embodiment of the present invention.

[0019] Figure 2 This is a flowchart of the genetic algorithm and Gauss-Newton iteration fusion calculation according to an embodiment of the present invention.

[0020] Figure 3 This is a comparison chart of active power test results according to an embodiment of the present invention.

[0021] Figure 4 This is a comparison chart of the emission voltage response test according to an embodiment of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be further described below.

[0023] like Figure 1 As shown, the dual-frequency impedance matching network parameter calculation model provided in this embodiment of the invention includes, from left to right, a signal source, a dual-frequency impedance matching network, a step-up transformer, and a transducer load.

[0024] The signal source includes an excitation voltage. and internal resistance The input voltage is marked as The total input current is marked as The circuit is divided into four grid regions, numbered 1, 2, 3, and 4, which are used to establish the voltage grid equations later.

[0025] The dual-frequency impedance matching network consists of a Π-type network and a T-type network connected in series. The Π-type network includes a first series inductor. and two first parallel capacitors , The node voltages are marked in the diagram. and branch current , The T-shaped network follows immediately after the Π-shaped network and includes two series inductors. and and a second parallel capacitor The node voltages are marked in the diagram. and branch current , , The Π-type network and the T-type network are directly connected in series to form a composite matching structure, which aims to cover both high-frequency and low-frequency bands simultaneously.

[0026] The step-up transformer is located between the matching network and the load, with a turns ratio n set between 1:5 and 1:10. The primary side of the transformer is connected to the matching network, and the secondary side is connected to the transducer load, serving to transform impedance and amplify voltage.

[0027] The transducer load is a multimode equivalent circuit model used to accurately characterize the admittance characteristics of a real composite transducer. This model includes a static capacitor. This includes a branch connected in parallel with subsequent branches, and three parallel RLC dynamic branches, each corresponding to one of the three resonant modes of the transducer. Specifically, branch 1 consists of... , , Composed of series connections; Branch 2 consists of , , Composed of series connections; Branch 3 consists of , , Composed of series connections. The load-side impedance is marked as... .

[0028] like Figure 2 As shown, the genetic algorithm and Gauss-Newton iteration fusion calculation process provided in this embodiment of the invention includes the following steps: First, target frequency band initialization is performed, setting the first and second target frequency bands, and dynamic boundary conditions are set, including convergence thresholds for the real and imaginary parts of impedance: ; In the formula, The internal resistance of the signal source, This is the preset convergence threshold.

[0029] Then, the core iterative calculation phase begins. This is based on the current iteration step U or the error. Perform a grading judgment, when the error When different threshold conditions are met (less than 0.055, less than 0.075, or less than 0.15, respectively), the learning factor is adaptively adjusted.

[0030] The core calculation process includes: deriving the port impedance function using the mesh method. Calculate the input impedance under the current network parameters: ; Where Qi is: ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; Next, the parameters of the dual-frequency impedance matching network are calculated by fusing genetic algorithms with Gauss-Newton iterations. First, the target frequency band is determined. Then, weighting coefficients, boundary conditions, population size, maximum number of iterations, mutation rate, crossover rate, and the proportion of local optimization in each generation are established. New points in each iteration are adaptively optimized by introducing a learning factor, which is dynamically adjusted as shown in the formula: ; in, As a learning factor, The maximum number of iterations, The maximum value of the learning factor. This represents the minimum value of the learning factor.

[0031] After the iteration converges, the impedance flatness within the frequency band is further determined, and the determination condition is as follows: ; If the flatness condition is not met, adjustments are made, and the process returns to the core calculation box for re-optimization. If all conditions are met, the final result is output. , , , , , Component parameter values. Specific implementation examples: First, the drilling transducer was selected, and the turns ratio of the step-up transformer was set to 1:6. The transducer's admittance data was measured using an impedance analyzer. Based on the least squares method, a nonlinear regression fitting was performed on the admittance data to obtain the transducer's multimodal equivalent circuit parameters, as shown in the table below.

[0033] Substituting the above parameters into the port impedance function In the calculation, the parameter values ​​of each component of the dual-frequency impedance matching network are obtained by combining genetic algorithm and Gauss-Newton iterative calculation.

[0034] Based on the calculation results, an actual dual-frequency impedance matching circuit was prepared and experimentally verified.

[0035] like Figure 3 The figure shows a comparison of active power tests. The horizontal axis represents frequency (kHz), ranging from 0 to 20kHz; the vertical axis represents active power (W). The solid line in the figure represents the power curve before impedance matching, and the dashed line represents the power curve after impedance matching. As can be seen from the figure, in the two target frequency bands (approximately 4.5kHz and 14.5kHz), the active power after impedance matching was significantly improved: the low-frequency power increased from approximately 50W to approximately 140W; and the high-frequency power increased from approximately 150W to over 250W. The overall test results indicate that the active power was increased by 1.8 times.

[0036] like Figure 4The figure shows a comparison of Transmitting Voltage Response (TVR) tests. The horizontal axis represents frequency (kHz), ranging from 0 to 30 kHz; the vertical axis represents the transmit voltage response (dB). The solid line in the figure represents the response curve before impedance matching, and the dashed line represents the response curve after impedance matching. As can be seen from the figure, the overall response level is significantly improved after impedance matching, with the average transmit voltage response increasing by 12 dB. Within a wide frequency band from 10 kHz to 20 kHz, the response curve is flatter and higher. The effective bandwidth increases by 1.4 times.

[0037] Experimental test results show that the dual-frequency broadband impedance matching design method for the drilling transducer proposed in this invention can effectively achieve broadband impedance matching in two target frequency bands, and can achieve efficient output in both high-frequency and low-frequency bands at the same time, effectively enhancing the detection range of the drilling instrument and improving the oil and gas drilling encounter rate.

[0038] The above are merely preferred embodiments of the present invention and do not constitute any limitation on the present invention. Any equivalent substitutions or modifications made by those skilled in the art to the technical solutions and content disclosed in the present invention without departing from the scope of the present invention shall be deemed to have remained within the protection scope of the present invention.

Claims

1. A dual-band impedance matching method for drilling transducers based on a fusion of genetic algorithm and Gauss-Newton iteration, characterized in that, Includes the following steps: S1: Establish a multimode equivalent circuit model of the drilling transducer, wherein the multimode equivalent circuit model includes a static capacitor. And N parallel RLC branches, each RLC branch being a resistor. ,inductance and capacitor Composed of series, where N is a positive integer; S2: The impedance data of the transducer at different frequency points is measured by an impedance analyzer. Based on the impedance data, the actual impedance data of the transducer is fitted by nonlinear regression using the least squares method to obtain the parameter values ​​of each component in the multimode equivalent circuit model. S2: Construct a dual-frequency impedance matching network, which consists of a Π-type network and a T-type network connected in series. The Π-type network includes a first series inductor. and two first parallel capacitors and The T-network includes a second series inductor. and the third series inductor and the second parallel capacitor ; S4: In the circuit composed of the dual-frequency impedance matching network, the step-up transformer, and the multi-mode equivalent circuit model, four grid regions are defined, voltage grid equations are established for each region, and the impedance function of the entire port is derived. ; S5: Set the target frequency band, including the first target frequency band and the second target frequency band, and set the boundary conditions; S6: Parameter calculation of dual-frequency impedance matching network is performed by fusing adaptive genetic algorithm with Gauss-Newton iteration. New points in each iteration are introduced by a learning factor. Perform adaptive optimization; S7: Based on error Perform a grading judgment, when the error When the convergence condition is not met, the learning factor Make dynamic adjustments, return to iterative calculation to continue calculating the parameters of the dual-frequency impedance matching network until convergence; S8: Determine whether the current frequency band flatness meets the preset conditions. If it does, output the parameter values ​​of each element in the dual-frequency impedance matching network.

2. The dual-band impedance matching method for drilling transducers based on the fusion of genetic algorithm and Gauss-Newton iteration as described in claim 1, characterized in that, The number of the N parallel RLC branches is 3, which correspond to the three resonant modes of the transducer.

3. The dual-band impedance matching method for drilling transducers based on the fusion of genetic algorithm and Gauss-Newton iteration as described in claim 1, characterized in that, The turns ratio n of the step-up transformer is set between 1:5 and 1:

10. The primary side is connected to the dual-frequency impedance matching network, and the secondary side is connected to the multi-mode equivalent circuit model.

4. The dual-band impedance matching method for drilling transducers based on the fusion of genetic algorithm and Gauss-Newton iteration as described in claim 1, characterized in that, The boundary conditions include convergence thresholds for the real and imaginary parts of the impedance: ； In the formula, The internal resistance of the signal source, This is the preset convergence threshold.

5. The dual-band impedance matching method for drilling transducers based on the fusion of genetic algorithm and Gauss-Newton iteration as described in claim 1, characterized in that, The learning factor The dynamic adjustment formula is: ； in, As a learning factor, The maximum number of iterations, The maximum value of the learning factor. This represents the minimum value of the learning factor.

6. The dual-band impedance matching method for drilling transducers based on the fusion of genetic algorithm and Gauss-Newton iteration as described in claim 1, characterized in that, The flatness determination criteria are as follows: .

7. A dual-band impedance matching circuit structure for a drilling transducer, comprising a signal source, a dual-band impedance matching network, a step-up transformer, and a transducer load, characterized in that: The signal source section is used to provide excitation voltage. and internal resistance The dual-frequency impedance matching network is connected to the signal source section and includes a Π-type network and a T-type network connected in series, wherein the Π-type network includes a first series inductor. and two first parallel capacitors , The T-network includes two series inductors. , and a second parallel capacitor ; The turns ratio n of the step-up transformer is 1:5 to 1:10, the primary side is connected to the dual-frequency impedance matching network, and the secondary side is connected to the transducer load. The transducer load is connected to the secondary side of the step-up transformer and includes a static capacitor. And N parallel RLC branches, each RLC branch is a resistor ,inductance and capacitor Composed of series connections.

8. The dual-band impedance matching circuit structure for a drilling transducer according to claim 7, characterized in that, The number of parallel RLC branches in the transducer load is three, corresponding to the three resonant modes of the transducer.

Images