Railway balise radio frequency circuit parameter optimization method and system

Abstract

The present invention provides a method and system for optimizing the parameters of a railway transponder radio frequency circuit, which belongs to the field of rail transit transponder equipment development and testing technology. It performs equivalent analysis on components in a high-frequency circuit, extracts key characteristic variables of the circuit, establishes a multi-objective optimization model of static power consumption and circuit cost, and uses a genetic optimization algorithm to optimize the parameters of the multi-objective optimization model. The present invention takes the construction of a multi-objective optimization function as a key starting point, and implements the entire process from feature vector selection to parameter optimization output; it provides a clear way to reduce the power consumption and circuit cost of existing transponder circuits, so that the implementation of parameter optimization can adopt the idea of ​​multi-objective optimization and data-driven, avoiding the limitation of time cost; it is directly applicable to the existing transponder radio frequency transmission circuit, effectively reducing the static power consumption and circuit cost of the circuit, and is also applicable to the optimization of other parameters and objectives in the circuit.

Description

Technical Field

[0001] The present invention relates to the technical field of development and testing of rail transit transponder equipment, and in particular to a method and system for optimizing radio frequency circuit parameters of a railway transponder based on multi-objective optimization. Background Art

[0002] The train operation control system is the "brain and central nervous system" of high-speed railway operation, and the balise is the key basic equipment for the train operation control system to realize train position correction. It is used to transmit absolute position information, temporary speed limit information, basic line parameters and other safety control information to the running train. According to whether the information transmitted by the balise to the train is fixed or variable, the balise can be divided into active balise and passive balise. The active balise can transparently transmit the variable message information data received from the lineside electronic unit (LEU) to the train, while the passive balise only transmits the fixed message information data stored locally to the train. Whether it is an active balise or a passive balise, no ground equipment power supply is required when transmitting message information data to the train. According to the different functions of the circuit units, the ground balise circuit can be simply divided into two parts: the downlink energy conversion circuit and the uplink signal radio frequency transmission circuit. Therefore, the optimization research on the power consumption and circuit cost of the balise circuit can be carried out around these two circuits. For the downlink signal conversion circuit, it is possible to consider improving its energy conversion efficiency, and for the uplink signal radio frequency transmission circuit, it is possible to consider reducing its energy loss that is not related to work. At present, research in this area is relatively scarce and mainly focuses on the downlink energy conversion circuit part of the transponder, and rarely involves the relevant optimization of the RF transmission circuit.

[0003] In the traditional analysis and research on improving transponder performance and circuits, the commonly used method is to predict and improve the signals it transmits, such as: using a linear gated cross-connection residual interleaving structure to eliminate the difference between signal and message parameter characteristics; or using uplink signal parameters to evaluate and predict transmission performance, and establishing corresponding models to predict and evaluate circuit parameter performance. However, these existing similar or related schemes generally have two major problems: (1) The schemes used focus on signal transmission quality and its prediction and improvement, and the scope of application is limited to the link where the transponder information is sent, without considering the impact of circuit energy transmission. (2) The schemes used are relatively rare and directly target the internal circuits of the transponder. The actual methods do not significantly improve the equipment and are difficult to implement.

[0004] With the introduction of the concept of green economy in the field of rail transit, the requirements for low power consumption and low cost of on-board equipment are increasing day by day. As an important part of the train control system, the ground transponder mainly obtains working energy through electromagnetic coupling. Since this energy conversion method is inefficient, in order to ensure the normal operation of the transponder, the on-board transponder transmission module BTM (Balise transmission module) device must always maintain a high transmission power. Therefore, the power consumption index of the transponder has become an important factor limiting the optimization of the transmission power of the on-board BTM device. For the transponder, improper parameter setting of the related components of its RF transmission circuit unit is an important reason for the high overall power consumption of the circuit. Therefore, it is very necessary to carry out research on the parameter optimization of this part of the circuit components.

[0005] To this end, how to use effective methods to construct a multi-objective optimization model of the overall circuit power consumption and circuit cost, and on this basis use a reasonable and effective way to optimize the parameters, and compared with existing equipment, it can reduce circuit power consumption, improve signal quality, and reduce circuit costs. Summary of the invention

[0006] The purpose of the present invention is to provide a method and system for optimizing the parameters of a railway transponder radio frequency circuit based on multi-objective optimization. According to the working process of the transponder radio frequency circuit, the circuit-related parameters are analyzed, characteristic factors and decision variables are extracted, and an accurate multi-objective optimization model is established. A complete transponder circuit parameter optimization solution is constructed to solve at least one technical problem existing in the above-mentioned background technology.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] On the one hand, the present invention provides a method for optimizing parameters of a railway transponder radio frequency circuit, comprising:

[0009] An equivalent analysis is performed on the components in the high-frequency circuit, key characteristic variables of the circuit are extracted, a multi-objective optimization model of static power consumption and circuit cost is established, and the parameters of the multi-objective optimization model are optimized using a genetic optimization algorithm; wherein, the equivalent analysis of the components in the high-frequency circuit is used to split the transistor and analyze its internal nodes, which are equivalent to capacitors and resistors and replaced into the original circuit; the establishment of the multi-objective optimization model of static power consumption and circuit cost is used to construct the multi-objective optimization model of static power consumption and circuit cost for the relevant decision variables, and the analysis adopts appropriate weights and coupling methods, and the decision variables are integrated by describing the association and fusion between the decision variables.

[0010] Optionally, the use of a genetic optimization algorithm to perform parameter optimization on a multi-objective optimization model is used to find the optimal solution set of the multi-objective optimization function, by considering the constraints and combining the optimization selection to further determine the optimization constraint range, and when the conditions are met, the required optimization parameters are obtained.

[0011] Optionally, the processing of performing equivalent analysis on the triode in the high-frequency circuit includes:

[0012] (1) AC equivalent model of voltage-controlled oscillator circuit

[0013] The high-frequency equivalent model of the transistor is introduced to replace the transistor in the original circuit diagram to facilitate the analysis of the circuit. First, the microstructure of the transistor is analyzed, and then the internal parameters of the transistor are brought into the voltage-controlled oscillation circuit to observe the influence of the internal parameters on the circuit structure. Then, its high-frequency equivalent model is substituted into the voltage-controlled oscillation circuit, and finally the AC equivalent model of the voltage-controlled oscillation circuit is obtained by analysis.

[0014] (2) Analysis of factors affecting RF transmission circuit performance

[0015] By analyzing the factors affecting the loaded quality factor, loop gain and phase condition of the frequency selection network, the optimization decision variables of the RF transmission circuit are established.

[0016] Optionally, the multi-objective optimization model for establishing static power consumption and circuit cost includes:

[0017] (1) Construction of circuit static power consumption evaluation function

[0018] Static power consumption refers to the energy loss caused by the static bias of the transistor. This indicator is selected as one of the optimization indicators of the RF transmission circuit.

[0019] (2) Construction of circuit cost evaluation function

[0020] By analyzing the sensitive factors related to circuit cost, such as loaded quality factor, transistor, feedback capacitor, two resonant capacitors, bias resistor, etc., a circuit cost evaluation function is constructed.

[0021] Optionally, the use of a genetic optimization algorithm to perform parameter optimization on a multi-objective optimization model is an evolutionary algorithm based on a genetic algorithm, which finds a series of balanced solutions during the optimization process by initializing the population, evaluating fitness, and allocating congestion, covering various trade-offs under multiple objectives, thereby achieving parameter optimization of the circuit static power consumption evaluation function and the circuit cost evaluation function.

[0022] Optionally, the method of using deep learning accelerates the training speed of NSGA-II by training a deep learning network model, applies a BNN neural network for acceleration, and observes the effect.

[0023] In a second aspect, the present invention provides a railway transponder radio frequency circuit parameter optimization system, comprising:

[0024] The processing module is configured to: perform equivalent analysis on components in high-frequency circuits, extract key characteristic variables of the circuit, establish a multi-objective optimization model for static power consumption and circuit cost, and use a genetic optimization algorithm to optimize parameters of the multi-objective optimization model; wherein, the equivalent analysis on components in high-frequency circuits is used to split the transistors and analyze their internal nodes, and convert them into capacitors and resistors and replace them into the original circuit; the establishment of a multi-objective optimization model for static power consumption and circuit cost is used to construct a multi-objective optimization model for static power consumption and circuit cost for related decision variables, analyze using appropriate weights and coupling methods, and integrate the decision variables by describing the association and fusion between the decision variables.

[0025] In a third aspect, the present invention provides a non-transitory computer-readable storage medium, which is used to store computer instructions. When the computer instructions are executed by a processor, the above-mentioned method for optimizing the radio frequency circuit parameters of a railway transponder is implemented.

[0026] In a fourth aspect, the present invention provides a computer device comprising a memory and a processor, wherein the processor and the memory communicate with each other, the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the railway transponder radio frequency circuit parameter optimization method as described above.

[0027] In a fifth aspect, the present invention provides an electronic device, comprising: a processor, a memory and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device executes instructions for implementing the method for optimizing the radio frequency circuit parameters of a railway transponder as described above.

[0028] The beneficial effects of the present invention are as follows: a complete transponder circuit parameter analysis and optimization scheme is formed, with the construction of a multi-objective optimization function as the key starting point, realizing the entire process from feature vector selection to parameter optimization output; a clear approach is provided for reducing the power consumption and circuit cost of existing transponder circuits, so that the implementation of parameter optimization can adopt the idea of ​​multi-objective optimization and data-driven, and break the limitation that traditional parameter adjustment and analysis methods are highly dependent on time cost; it is directly applicable to existing transponder RF transmitting circuits, effectively reducing circuit static power consumption and circuit cost, and is also applicable to the optimization of other parameters and objectives in the circuit.

[0029] Additional advantages of the present invention will be more clearly given in the following description or learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0031] Figure 1 The present invention is a schematic diagram of the implementation principle of the method for optimizing the radio frequency circuit parameters of a railway transponder based on multi-objective optimization according to an embodiment of the present invention.

[0032] Figure 2 A microscopic structure diagram of a transistor according to an embodiment of the present invention.

[0033] Figure 3 It is a schematic diagram of a high-frequency equivalent model of a transistor according to an embodiment of the present invention.

[0034] Figure 4 This is a schematic diagram of the communication path behind the equivalent related components described in an embodiment of the present invention.

[0035] Figure 5 This is a before and after comparison diagram of the frequency selection network access circuit described in an embodiment of the present invention.

[0036] Figure 6 This is a frequency domain response curve diagram of the high frequency band of the common emitter circuit described in an embodiment of the present invention.

[0037] Figure 7 This is a diagram of impedance series-parallel conversion according to an embodiment of the present invention.

[0038] Figure 8 4 is an equivalent circuit diagram of the feedback network described in an embodiment of the present invention.

[0039] Fig. 9 4 is a relationship diagram of the phase shift of the feedback network according to an embodiment of the present invention.

[0040] Fig.10 This is a flow chart of the NSGA-II algorithm described in an embodiment of the present invention.

[0041] Fig.11 This is a curve showing the influence of DropoutRate on fitting accuracy described in an embodiment of the present invention.

[0042] Fig.12 This is a diagram of the BNN network fitting effect described in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The embodiments described below by the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0044] It should be understood by those skilled in the art that unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs.

[0045] It should also be understood that terms, such as those defined in commonly used dictionaries, should be understood to have a meaning consistent with that in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless as defined herein.

[0046] Those skilled in the art will appreciate that, unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements and / or groups thereof.

[0047] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0048] To facilitate understanding of the present invention, the present invention is further explained below with reference to specific embodiments in conjunction with the accompanying drawings, and the specific embodiments do not constitute a limitation on the embodiments of the present invention.

[0049] Those skilled in the art should understand that the drawings are merely schematic diagrams of embodiments, and the components in the drawings are not necessarily necessary for implementing the present invention.

[0050] Example 1

[0051] In the existing technology, firstly, the ability to optimize parameters is limited, and it mainly focuses on performance prediction and analysis, and rarely considers how to improve existing circuits; secondly, the method of optimizing parameters mainly relies on modeling, but the data conditions used in modeling are not sufficient, and the coupling relationship between circuits is complex, which makes it difficult to accurately model and analyze, resulting in unstable optimization results and large deviations.

[0052] This embodiment 1 provides a method for optimizing the parameters of the RF circuit of a railway transponder based on multi-objective optimization. The scheme adopted is relatively rare for the transponder system, and the actual on-site equipment data is difficult to obtain or the conditions are not available. It is directly oriented to the transponder circuit and has a direct fit with the research object. It aims to solve the problem of lack of effective and dedicated methods for reducing power consumption and cost at the transponder circuit level in current applications. First, the circuit working mechanism is analyzed and converted, and based on this, a circuit parameter target optimization model is constructed. The model parameters are optimized using a multi-objective optimization method, and the diversity and functional availability of the transponder are comprehensively considered to test the functions and performance of the railway transponder, and improve the problem of high power consumption and circuit cost of the transponder.

[0053] In this embodiment 1, a railway transponder radio frequency circuit parameter optimization system is first provided, including: a processing module, which is configured to: perform equivalent analysis on components in a high-frequency circuit, extract key characteristic variables of the circuit, establish a multi-objective optimization model for static power consumption and circuit cost, and use a genetic optimization algorithm to optimize the parameters of the multi-objective optimization model; wherein, the equivalent analysis of components in a high-frequency circuit is used to split the transistor and analyze its internal nodes, which are equivalent to capacitors and resistors and replaced into the original circuit; the establishment of a multi-objective optimization model for static power consumption and circuit cost is used to construct a multi-objective optimization model for static power consumption and circuit cost for related decision variables, respectively, and analyze the decision variables using appropriate weights and coupling methods, and integrate the decision variables by describing the association and fusion between the decision variables.

[0054] In this embodiment 1, based on the above system, a method for optimizing the parameters of the radio frequency circuit of a railway transponder based on multi-objective optimization is implemented, which is used to analyze the radio frequency transmission circuit of the transponder, construct a high-frequency equivalent circuit model, provide decision variables for generating a multi-objective optimization function of static power consumption and circuit cost, and optimize the parameters of the multi-objective optimization function by using a hybrid model of genetic algorithm and neural network, thereby reducing circuit power consumption, improving signal quality, and reducing circuit cost. The method includes a processing method for equivalent analysis of components such as triodes in high-frequency circuits, a method for extracting key characteristic variables of circuits, establishing a multi-objective optimization model of static power consumption and circuit cost, a method for optimizing the parameters of the multi-objective optimization model using a genetic optimization algorithm, and a method for accelerating the optimization of the genetic algorithm using a neural network. The on-site operation and maintenance unit of the train control equipment uses the method provided by the present invention to first analyze the transponder circuit and seek to optimize the characteristic variables, then construct a multi-objective optimization function, call the genetic algorithm to optimize its parameters, and finally update and adjust the transponder used on site according to the optimization results, so as to provide a decision basis for its operation and maintenance and performance improvement.

[0055] The processing method for equivalent analysis of components such as transistors in high-frequency circuits is used to split the transistors and analyze their internal nodes, convert them into capacitors, resistors, etc. and replace them into the original circuit, which is conducive to refining circuit research and selecting key characteristic variables such as decision variables.

[0056] The method for establishing a multi-objective optimization model for static power consumption and circuit cost is used to construct a multi-objective optimization model for static power consumption and circuit cost using relevant decision variables, analyze and adopt appropriate weights and coupling methods, and integrate decision variables by describing the association and fusion between decision variables, thereby providing a compliant interface for the construction of a multi-objective optimization function model for static power consumption and circuit cost.

[0057] The method of optimizing parameters of a multi-objective optimization model using a genetic optimization algorithm is used to find the optimal solution set of a multi-objective optimization function. By considering the constraints and combining the optimization selection, the optimization constraint range is further determined. When the conditions are met, the required optimization parameters are obtained, thereby improving the parameters of existing circuit components, reducing power consumption, improving circuit quality, and reducing costs.

[0058] The method of using deep learning to accelerate the genetic algorithm enhances the practicality of the above method, reduces the time cost, and accurately and quickly seeks the corresponding output through the training of the deep learning neural network, which is easy to operate and implement.

[0059] The proposed processing method for equivalent analysis of components such as triodes in high-frequency circuits adopts a method of replacing the high-frequency equivalent circuit of triodes, which is mainly described as follows:

[0060] In the voltage-controlled oscillation circuit, since it is impossible to directly observe the influence of the internal parameters of the transistor on the circuit, it is necessary to introduce the high-frequency equivalent model of the transistor in order to further analyze the circuit. Figure 2 As shown in the figure, r bb’ is the base resistance, usually tens to hundreds of ohms, r c and r e They are the collector junction and emitter junction resistances, which can be ignored because their resistances are usually very small. JC and C JE They are emitter junction capacitance and collector junction capacitance, which are usually tens to tens of pico-farads. When the transistor is in the low and medium frequency bands, the junction capacitance has little effect on the circuit. However, in the high frequency band, the junction capacitance has a significant impact on the circuit performance, so it needs to be paid special attention. Therefore, the high-frequency equivalent model of the transistor is established as follows: Figure 3 shown.

[0061] Since the capacitor C JC The capacitor C is placed across b' and c, which is not conducive to the analysis of the circuit. In order to further eliminate the coupling relationship between the front and back of the circuit, JC Equivalent to C a and C b The two capacitors are connected in parallel at the input and output ends of the transistor, and the resistor r bb’ Equivalent to r b1 and r b2 The two resistors are connected in parallel to the feedback network and the input end of the transistor, and the original frequency selection network is equivalent to capacitor C, inductor L and equivalent resistor R. Q In parallel form. Since r b’c The resistor is usually large, so it is treated as an open circuit. The final simplified circuit is as follows Figure 4 As shown:

[0062] For the capacitor C JC For example, its source is connected between the transistor node b' and node c, and the voltage relationship between the two nodes is:

[0063]

[0064] Among them, g m is the transconductance of the transistor; P is the feedback coefficient of the feedback circuit; I bq is the transistor base static bias current, V EE It is the power supply for the transistor bias circuit, and R2 is the static bias resistor of the transistor.

[0065] According to Miller's theorem, C a and C b The value of is:

[0066]

[0067] For the original resistance r bb’ For example, its source is connected between the transistor node b and node b', and the voltage relationship between the two nodes is:

[0068]

[0069] in, is the resistance r be The conductivity of is the capacitance C JE The conductivity of is the capacitance C a The conductivity of the resistor r is known from Miller's theorem. b1 and resistor r b2 The values ​​are:

[0070]

[0071] The method for establishing a multi-objective optimization model for static power consumption and circuit cost is used to establish a multi-objective optimization model for the parameters to be optimized and to integrate the decision variables. The specific method includes three parts: analysis of factors affecting the performance of the RF transmission circuit, establishment of the decision variables for optimizing the RF transmission circuit, and construction of the objective function for optimizing the RF transmission circuit.

[0072] S3.1 Analysis of factors affecting RF transmission circuit performance

[0073] The influencing factors of the RF transmitting circuit will be studied from three aspects: frequency selection network, loop gain and phase conditions.

[0074] (1) Analysis of factors affecting the quality factor of the frequency selection network

[0075] Comparison of the frequency selection network before and after it is connected to the circuit Figure 5 shown. Figure 5 (b) Lieutenant General C b Connected to the frequency-selective network, Figure 5 (c) Resistors R2 and R b1 Equivalent to R L , the resonant frequency of the circuit is given by becomes It can be concluded that the parasitic parameters of the transistor will affect the resonant frequency of the circuit. In order to offset this effect, this paper adds capacitors C1 and C4 to the frequency selection network, and the total resistance connected to the frequency selection network is changed from R Q becomes R' L , as shown in formula (5), where “ / / ” means parallel connection:

[0076]

[0077] Among them, R L The resistors R2 and r b1 The equivalent resistance. R' L is the total load equivalent resistance of the circuit. From the quality factor formula of the frequency selection network, we know that the connection R L The resistance value will decrease and the quality factor will decrease accordingly.

[0078] (2) Analysis of factors affecting loop gain

[0079] For an oscillator circuit, the circuit needs to meet the condition that the loop gain is greater than 1 for it to start oscillating. From the node voltage relationship obtained by simplifying the circuit using Miller's theorem twice in the previous article, combined with formula (1) and formula (3), we can know that:

[0080]

[0081] The loop gain |T| is:

[0082] |T|=|K 2 P 2 K1| (7)

[0083] (3) Analysis of factors affecting phase conditions

[0084] The oscillation circuit must meet the requirement of additional phase shift of 2nπ in the loop. The additional phase shift is mainly produced by the single-tube common-emitter amplifier circuit. The frequency domain response curve of this circuit in the high frequency band is as follows: Figure 6 As shown in the figure. H represents the high frequency cutoff frequency corresponding to the transistor frequency domain. Before calculating it, you need to Figure 5 The circuit elements between the two stages of the triode b'e are connected in parallel to the equivalent of series impedance, such as Figure 6 After the conversion, the circuit forms a low-pass filter, the cut-off frequency of which is the high-frequency cut-off frequency f of the transistor. H , the calculation formula is as follows:

[0085]

[0086] Where: r m ——Equivalent series resistance;

[0087] Q a - the quality factor of the circuit in this part;

[0088] Since the entire oscillation circuit operates in the high frequency band of the transistor, its operating frequency will be 0.1f H After that, combined with Figure 7 It can be seen that the circuit will introduce a phase shift of more than -180 degrees. The oscillation circuit contains two identical amplification links, so an additional phase shift of more than -360 degrees will be introduced. The phase condition for oscillation cannot be met, so the feedback link must introduce a positive phase shift.

[0089] Feedback The circuit diagram is as follows Figure 8 As shown in the figure, it can be seen that the phase shift of the feedback network is determined by multiple components. This paper uses the method of controlling variables to study the influence of relevant parameters on the phase shift characteristics of the feedback network, such as Fig. 9 As shown. Combining the phase shift characteristic curves of the amplification network and the feedback network, although a single transistor produces an additional phase shift of more than 180 degrees, the feedback network can produce a positive phase shift to compensate for the part of the former that exceeds 180 degrees. As long as the appropriate parameters are determined, the additional phase shift of a single transistor can be 180 degrees. There are two transistor amplification links in the circuit, so the phase shift condition of 2nπ can be met under certain conditions.

[0090] S3.1. Establishment of decision variables for RF transmission circuit optimization

[0091] In the analysis of the factors affecting circuit performance in the previous article, many secondary parameters were generated by introducing Miller's theorem, but these secondary parameters are all obtained based on the primary parameters in Table 1. The 10 primary parameters in Table 1 are the key factors affecting circuit performance. This article finally defines the decision variable of the circuit as a 10-dimensional vector x. The specific meaning of each dimension is shown in Table 1 below:

[0092] Table 1 Decision variables table

[0093]

[0094] S3.2. Construction of RF transmission circuit optimization objective function

[0095] a) Construction of circuit static power consumption evaluation function

[0096] Static power consumption refers to the energy loss caused by the static bias of the transistor. This article selects this indicator as one of the optimization indicators of the RF transmission circuit. There are two main parameters related to the static power consumption of the circuit, which are the base bias resistor R bias And the DC gain β of the transistor, the two together determine the static bias current of the circuit. bias The smaller β is, the larger the bias current of the circuit is, and the power supply voltage is fixed, so the power consumption is higher. The relationship between the above variables is shown in formula (9):

[0097]

[0098] Where: Ic is the quiescent operating current of the transistor; V cc is the power supply voltage of the circuit, in V; V bias For R bias The voltage drop is in V; P is equal to the static power consumption of the transistor; R bias is the static bias resistor of the transistor; β is the DC gain of the transistor. Finally, the circuit static power consumption evaluation function P(x) is expressed as shown in formula (10):

[0099] P(x)=11.5*x3 / x8 (10)

[0100] b) Construction of circuit cost evaluation function

[0101] There are three main sensitive factors related to circuit cost. The first is the circuit's no-load quality factor Q0. The larger the Q0, the better the circuit performance. However, if you want to increase Q0, you must use capacitors and antenna coils with high quality factors in the frequency selection circuit (equivalent inductance in the frequency selection circuit), which undoubtedly increases the cost of the circuit. The second factor that affects the circuit cost is the transistor. For transistors, C JC and r bb , the product of which is used to represent its high-frequency performance. The smaller the product of the two, the better the high-frequency performance and the higher the cost. The DC gain and characteristic frequency of the transistor are opposite to its high-frequency performance indicators. The larger the values ​​of the first two, the higher the price of the transistor. In addition, there are feedback capacitor C1 and two resonant fine-tuning capacitors C ajt1 , C ajt2 And the bias resistor R bias , their costs are proportional to their values, forming the third factor affecting the circuit cost. Since the dimensions of the multiple independent variables of the cost evaluation function are not uniform, and there is no clear physical relationship like the static power evaluation function. Therefore, this paper chooses to normalize the multiple independent variables and then use the weighted summation method to construct the cost evaluation function. The final cost evaluation function C(x) of the circuit is shown in formula (11):

[0102]

[0103] The method of using genetic optimization algorithm to optimize the parameters of multi-objective optimization model is used to find the optimal solution set of multi-objective optimization function, and further determine the optimization constraint range by considering the constraint conditions and combining the optimization selection. When the conditions are met, the required optimization parameters are obtained to improve the parameters of existing circuit components, reduce power consumption, improve circuit quality, and reduce costs. It is specifically divided into the setting of RF transmission circuit optimization constraints and the use of NSGA-II to optimize the parameters of multi-objective functions:

[0104] S4.1. Setting of constraints for RF transmission circuit optimization

[0105] (1) RF transmission circuit signal quality evaluation indicators

[0106] According to the transponder test specification, the uplink FSK signal of the transponder needs to meet certain requirements in terms of center frequency, frequency offset, average symbol rate, amplitude jitter, bandwidth and other indicators. The signal analyzed in this article is the current signal generated in the transmitting antenna after the transponder is activated, that is, the source signal emitted by the RF transmitting circuit of the transponder. In order to ensure that the transponder can work normally, in addition to meeting the above-mentioned related parameter indicators, the transponder also needs to meet the minimum operating current requirement of not less than 60mA. The final RF transmitting circuit signal quality evaluation indicators are shown in Table 2:

[0107] Table 2 RF signal transmission quality evaluation table

[0108]

[0109] (2) Obtaining the signal quality evaluation index of the RF transmission circuit

[0110] First, the current signal in the antenna of the RF transmitting circuit needs to be processed. Since it is a 2FSK signal, commonly used demodulation methods include frequency discrimination, differential detection, coherent demodulation, zero-crossing detection, etc. Since the focus of this paper is not on signal demodulation, the easy-to-implement zero-crossing detection method is selected to demodulate the current signal while ensuring the demodulation accuracy. When the sampling rate reaches 50MHz and the spline interpolation method is used to increase the sampling rate by 10 times, the maximum error of the frequency estimation will be much less than 0.065KHz, as shown in formula (12):

[0111]

[0112] It can be seen that as long as the signal sampling rate is set to 50MHz, the normal demodulation requirements can be met by using the zero-crossing detection method combined with spline interpolation.

[0113] a) Center frequency and offset frequency analysis

[0114] After the zero-crossing point of the signal is detected by the zero-crossing detection method, a frequency estimation value can be calculated using the time difference between two adjacent zero-phase zero-crossing points, so that multiple frequency estimation values ​​can be obtained within the entire sampling period. According to the transponder test specification, a 16-bit sliding window is required to evaluate the center frequency and frequency offset of the current signal, and segment-by-segment analysis is performed in each window. The specific test steps are as follows:

[0115] 1. Use the first 16 bits of the acquired signal as the beginning of the 16-bit sliding window.

[0116] 2. Calculate the average frequency for each bit in the window

[0117] 3. Find the minimum value f of the average frequency of bit0 in the window respectively Lmin and the maximum value f Lmax Similarly, find the minimum value f of the average rate of bit1 Hmin and f Hmax .

[0118] 4. In the worst case, the four values ​​obtained from step 3 should satisfy the following relationship:

[0119]

[0120] 5. The window slides back 1 bit and repeats steps 2 to 5 until all bits of data have been tested.

[0121] From the test step 4) above, it can be seen that the maximum value f of the center frequency in the current window can be calculated in each 16-bit window. Cmax and the minimum value f Cmin And the maximum value of the offset frequency f Dmax and the minimum value f Dmin .

[0122] Therefore, after all windows are analyzed, this paper selects the maximum center frequency and its offset frequency and the minimum center frequency and its offset frequency in all windows as the parameters for the final evaluation of the center frequency and its offset frequency.

[0123] b) Amplitude jitter analysis

[0124] According to the test specification, the measurement method of amplitude jitter needs to define the calculation center window W i and long time window W m The ratio of the average amplitude within. m The window requirement is 50us~100us, W i The window is 1.77us, and W i Window is located at W m The calculation formulas for the two are:

[0125]

[0126]

[0127] Where N is the number of sampling points in the window period, and x[k] is the sampling value. The third output variable is determined by this

[0128] c) Bandwidth analysis

[0129] According to the test specification, the ratio of the sum of the signal energy in the 1MHz window before and after the two upper and lower sidebands to the energy in the 1MHz window before and after the center frequency should be ≤-10dB, thus determining the bandwidth B w for:

[0130]

[0131] d) Average symbol rate analysis

[0132] According to the test specification requirements, the calculation formula for the average symbol rate v is:

[0133]

[0134] Where T (b0) and T (b1500) They are the conversion times of the message 0 bit and 1500 bit respectively, in seconds.

[0135] e) Working current analysis

[0136] The specification does not provide a specific method for analyzing the working current. This article chooses to use a method based on fast Fourier decomposition to find the maximum current amplitude in the upper and lower sidebands of 40KHz bandwidth, and then selects the smaller of the two values ​​as the final working current I C The value of .

[0137] (3) Establishment of RF transmission circuit optimization constraints

[0138] For the three parameters of center frequency, offset frequency and average symbol rate, the signal quality can be measured only by the deviation from the standard value. Therefore, the maximum center frequency deviation, the maximum offset frequency deviation and the maximum average symbol rate deviation are used as the final signal quality evaluation parameters. The definition of the three variables is as follows:

[0139]

[0140] Finally, this paper defines the total constraint variable as y, which is a 6-dimensional variable. The specific meaning of each dimension is shown in Table 3:

[0141] Table 3 Parameters of constraint variables in each dimension

[0142]

[0143] The set constraint number condition is shown in formula (19):

[0144]

[0145] S4.2, NSGA-II algorithm

[0146] Multi-Objective Optimization Problem (MOOP) refers to a problem that requires optimizing multiple objective functions at the same time. Unlike single-objective optimization problems, the various objectives in multi-objective optimization must conflict with each other, and the optimization of one objective will inevitably lead to the degradation of other objectives. Therefore, in multi-objective optimization problems, there is no unique solution like single-objective optimization that makes all objective functions optimal. Instead, there is a set of solutions that includes all possible solutions, and different solutions cannot be simply compared in terms of advantages and disadvantages, but need to be weighed according to actual conditions. Therefore, such a solution is called a non-dominated solution or a Pareto optimal solution. The solution to multi-objective optimization problems is generally divided into the following steps: First, determine the decision variables: determine the decision variables that can be optimized, and define the search range and search accuracy for each decision variable; second, determine the objective function: it is necessary to determine the definition and form of all objective functions, that is, map the design variables to the target space; third, determine the constraints: give the definition and form of all constraints, that is, the constraints on the design variables in practical applications; then, determine the solution method: select a suitable solution method to search for a set of Pareto optimal solutions, such as particle swarm algorithm, genetic algorithm, etc.; finally, optimize the solution and analyze the solution set: write a computer program to solve the MOOP problem, analyze and screen the final Pareto frontier, and find the final solution.

[0147] After the objective function and constraints are given, how to choose the right algorithm to solve the problem becomes the key. The solution of multi-objective optimization problems usually requires the use of multi-objective optimization algorithms, which can be divided into two categories: traditional methods and heuristic methods. Traditional methods include linear programming methods based on weighted averages, methods based on goal programming, and methods based on boundary analysis. The advantage of these methods is that they are simple and easy to implement, but they usually cannot handle nonlinear, non-convex or high-dimensional problems well. Heuristic methods are a type of search-based algorithms that use evolutionary algorithms, genetic algorithms, simulated annealing, particle swarm algorithms, etc. to search for the Pareto frontier. These methods can usually handle complex problems well. This article takes the NSGA algorithm as an example to explain the working principle of this type of search algorithm.

[0148] NSGA (Non-dominated Sorting Genetic Algorithms) is a genetic algorithm based on Pareto optimal theory. It divides individuals into different levels according to their dominance and non-domination relationships, and then selects and regenerates these levels to obtain very satisfactory multi-objective optimization results. However, in practical applications, it is found that this algorithm has some shortcomings, mainly including the following three aspects: First, the computational complexity of the original algorithm is very high. For a problem with a population size of N and the number of objective functions of m, the non-dominated sorting algorithm needs to perform m*N 3 The algorithm has to compare and sort the populations of the NSGA algorithm. As the problem size increases, the computational complexity of the algorithm will increase accordingly. Second, there is a lack of elite strategy. Studies have shown that the introduction of elite strategy can improve the performance of genetic algorithms and prevent excellent individuals from being missed. Third, the shared parameters need to be manually specified. The population maintenance and diversity retention in the NSGA algorithm rely on the concept of sharing, and the selection of shared parameters needs to be manually specified, which affects the performance and effect of the algorithm.

[0149] The non-dominated sorting genetic algorithm NSGA-II is designed to overcome the defects of NSGA and improve the accuracy and diversity of optimization results. The NSGA-II algorithm flow chart is as follows Fig.10 shown.

[0150] The NSGA-II algorithm has made the following three improvements: First, the use of fast non-dominated sorting algorithm: NSGA-II uses a fast non-dominated sorting algorithm to reduce the complexity of the calculation program, thereby improving the efficiency and accuracy of the algorithm. Second, the introduction of elite strategy: NSGA-II introduces an elite strategy to select excellent individuals as "elites", thereby expanding the sampling space, improving the search efficiency of the algorithm and the accuracy of the optimization results. Through the elite strategy, NSGA-II can retain excellent individuals and prevent the population from falling into the local optimal solution. Finally, the introduction of crowding and crowding comparison operators: The traditional NSGA algorithm must manually specify shared parameters, which will introduce non-objective factors to the solution of the model, while the NSGA-II algorithm introduces a more objective concept of crowding, using crowding to quantify the degree of dispersion of the current population, ensuring that the population maintains a certain degree of dispersion during the search process to avoid falling into the local optimal solution. Through crowding, NSGA-II can make the solutions in the population evenly distributed on the Pareto frontier, so as to better explore the Pareto optimal solution.

[0151] The method of using deep learning to accelerate the genetic algorithm enhances the practicality of the above method, reduces time cost, and accurately and quickly seeks corresponding outputs through the training of deep learning neural networks, which is easy to operate and implement. Deep learning networks can use three types: BNN, LSTM and DNN. After testing, BNN has the best effect. The following is a description using BNN as an example:

[0152] S5.1. BNN neural network structure design and hyperparameter selection

[0153] The BNN network mainly consists of input, hidden layer, MCdropout layer and output layer. The final BNN network structure is shown in Table 4:

[0154] Table 4 BNN network structure

[0155]

[0156] The final hyperparameters of the BNN network are lr = 0.001, batch_size = 25, epoch = 1000, optimizer = Adam, and the patience parameter in Earlystopping is 50. In addition, since the BNN network can provide uncertainty information of the output parameters, the number of predictions is designed to be 100 times.

[0157] In addition, since the MCdropout layer is added, the DropoutRate parameter needs to be passed to this layer. The impact of different Dropout rates on the fitting accuracy is as follows Fig.11 As shown:

[0158] It can be seen from the figure that when DropoutRate = 0.05, the accuracy is higher. Keep the final selection of DropoutRate as 0.05.

[0159] S5.2. Analysis of Constraint Variable Prediction Results

[0160] The prediction effect of the BNN solution is as follows Fig.12 As shown in the figure, the red curve is the true value of the output variable of the test set, and the light green area is the 95% confidence interval output after 100 predictions on the test set. The mean of the 95% confidence interval is selected as the final prediction value, as shown in the green curve in the figure. It can be seen from the figure that there is a high fitting accuracy between the predicted value and the true value, and a high prediction accuracy is achieved. According to Tables 5 and 6, it is concluded that the use of BNN neural network can optimize parameters, reduce circuit cost and static power consumption while significantly reducing time cost.

[0161] Table 5 Comparison of indicators before and after optimization based on NSGA-II

[0162]

[0163] Table 6 Comparison of indicators before and after optimization based on NSGA-II and BNN neural network

[0164]

[0165] In summary, from the perspective of implementation, other parameter optimization methods, such as optimization by modeling, can also achieve the purpose of parameter optimization. However, the railway transponder is a safety device with complex circuits and numerous coupling relationships. It is very difficult to analyze and directly model its working mechanism, and the accuracy is difficult to guarantee. In actual applications, there are large deviations and lack of stability. In this embodiment, based on the existing conditions of the equipment, the circuit parameters are analyzed, and an optimization function is established with static power consumption and circuit cost as the goals. Multi-objective optimization and deep learning methods are used for analysis, which avoids complex models, can achieve the optimization of circuit parameters, and support the implementation of the situation-based maintenance model.

[0166] Example 2

[0167] This embodiment 2 provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the above-mentioned method for optimizing the parameters of a railway transponder radio frequency circuit is implemented. The method includes:

[0168] An equivalent analysis is performed on the components in the high-frequency circuit, key characteristic variables of the circuit are extracted, a multi-objective optimization model of static power consumption and circuit cost is established, and the parameters of the multi-objective optimization model are optimized using a genetic optimization algorithm; wherein, the equivalent analysis of the components in the high-frequency circuit is used to split the transistor and analyze its internal nodes, which are equivalent to capacitors and resistors and replaced into the original circuit; the establishment of the multi-objective optimization model of static power consumption and circuit cost is used to construct the multi-objective optimization model of static power consumption and circuit cost for the relevant decision variables, and the analysis adopts appropriate weights and coupling methods, and the decision variables are integrated by describing the association and fusion between the decision variables.

[0169] Example 3

[0170] This embodiment 3 provides a computer device, including a memory and a processor, the processor and the memory communicate with each other, the memory stores program instructions that can be executed by the processor, the processor calls the program instructions to execute the above-mentioned railway transponder radio frequency circuit parameter optimization method, the method comprising:

[0171] An equivalent analysis is performed on the components in the high-frequency circuit, key characteristic variables of the circuit are extracted, a multi-objective optimization model of static power consumption and circuit cost is established, and the parameters of the multi-objective optimization model are optimized using a genetic optimization algorithm; wherein, the equivalent analysis of the components in the high-frequency circuit is used to split the transistor and analyze its internal nodes, which are equivalent to capacitors and resistors and replaced into the original circuit; the establishment of the multi-objective optimization model of static power consumption and circuit cost is used to construct the multi-objective optimization model of static power consumption and circuit cost for the relevant decision variables, and the analysis adopts appropriate weights and coupling methods, and the decision variables are integrated by describing the association and fusion between the decision variables.

[0172] Example 4

[0173] This embodiment 4 provides an electronic device, including: a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes instructions for implementing the above-mentioned method for optimizing the parameters of the radio frequency circuit of the railway transponder, and the method includes:

[0174] An equivalent analysis is performed on the components in the high-frequency circuit, key characteristic variables of the circuit are extracted, a multi-objective optimization model of static power consumption and circuit cost is established, and the parameters of the multi-objective optimization model are optimized using a genetic optimization algorithm; wherein, the equivalent analysis of the components in the high-frequency circuit is used to split the transistor and analyze its internal nodes, which are equivalent to capacitors and resistors and replaced into the original circuit; the establishment of the multi-objective optimization model of static power consumption and circuit cost is used to construct the multi-objective optimization model of static power consumption and circuit cost for the relevant decision variables, and the analysis adopts appropriate weights and coupling methods, and the decision variables are integrated by describing the association and fusion between the decision variables.

[0175] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative work on the basis of the technical solution disclosed in the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for optimizing parameters of a railway transponder radio frequency circuit, characterized in that: include: Perform equivalent analysis on components in high-frequency circuits, extract key characteristic variables of circuits, establish a multi-objective optimization model for static power consumption and circuit cost, and use genetic optimization algorithms to optimize parameters of the multi-objective optimization model; wherein, the equivalent analysis on components in high-frequency circuits is used to split transistors and analyze their internal nodes, convert them into capacitors and resistors and replace them in the original circuit; the establishment of a multi-objective optimization model for static power consumption and circuit cost is used to construct a multi-objective optimization model for static power consumption and circuit cost for related decision variables, analyze and use appropriate weights and coupling methods, and integrate decision variables by describing the association and fusion between decision variables; Among them, an equivalent analysis is performed on the components in the high-frequency circuit to extract the key characteristic variables of the circuit, including: in the voltage-controlled oscillation circuit, a high-frequency equivalent model of the transistor is introduced, including the base body resistance, the collector junction, the emitter body resistance, the emitter junction capacitance and the collector junction capacitance; the emitter junction capacitance is equivalent to two capacitors connected in parallel at the input and output ends of the transistor respectively, and the base body resistance is equivalent to two resistors connected in parallel at the feedback network and the input end of the transistor respectively; for the emitter junction capacitance, it is originally connected between the node b' and the node c of the transistor, and according to the voltage relationship between the node b' and the node c, the capacitance of the two capacitors equivalent to the emitter junction capacitance is obtained in combination with Miller's theorem; for the base body resistance, it is originally connected between the node b and the node b' of the transistor, and according to the voltage relationship between the node b and the node b', the resistance of the two resistors equivalent to the base body resistance is obtained by using Miller's theorem; Among them, establishing a multi-objective optimization model for static power consumption and circuit cost includes: according to the factors affecting the performance of the RF transmission circuit, establishing the optimization decision variables of the RF transmission circuit, and constructing the RF transmission circuit optimization objective function and the circuit static power consumption evaluation function; wherein, the RF transmission circuit performance influencing factors include the loaded quality factor of the frequency selection network, the loop gain and the phase condition; Among them, the genetic optimization algorithm is used to optimize the parameters of the multi-objective optimization model, including: finding a series of balanced solutions in the optimization process by initializing the population, evaluating fitness, and allocating congestion, covering various trade-offs under multiple objectives, and realizing parameter optimization of the circuit static power consumption evaluation function and the circuit cost evaluation function.

2. The method for optimizing the parameters of the radio frequency circuit of a railway transponder according to claim 1, characterized in that: The genetic optimization algorithm is used to optimize the parameters of the multi-objective optimization model to find the optimal solution set of the multi-objective optimization function, and the optimization constraint range is further determined by considering the constraint conditions and combining the optimization selection. When the conditions are met, the required optimization parameters are obtained; Including, setting of optimization constraints for RF transmission circuit and optimization of parameters for multi-objective functions using NSGA-II: Setting of optimization constraints for RF transmission circuit includes determining and obtaining the evaluation index of signal quality of RF transmission circuit, including: processing the current signal in the antenna of RF transmission circuit and demodulating the current signal using zero-crossing detection method; After detecting the zero-crossing point of the signal using the zero-crossing detection method, a frequency estimate is calculated using the time difference between two adjacent zero-phase zero-crossing points. In this way, multiple frequency estimates can be obtained within the entire sampling period. According to the transponder test specifications, the center frequency and frequency offset of the current signal are evaluated. The amplitude jitter, bandwidth, average symbol rate, and operating current are measured. The optimization constraints of the RF transmission circuit are established.

3. The method for optimizing the parameters of the radio frequency circuit of a railway transponder according to claim 1, characterized in that: The multi-objective optimization model for establishing static power consumption and circuit cost includes: (1) Construction of circuit static power consumption evaluation function Static power consumption refers to the energy loss caused by the static bias of the transistor. Static power consumption is selected as one of the optimization indicators of the RF transmission circuit. (2) Construction of circuit cost evaluation function By analyzing the sensitive factors related to circuit cost, such as loaded quality factor, transistor, feedback capacitor, two resonant capacitors and bias resistor, a circuit cost evaluation function is constructed.

4. The method for optimizing the parameters of the radio frequency circuit of a railway transponder according to claim 1, characterized in that: Use deep learning methods to speed up the training of NSGA-II by training deep learning network models, apply BNN neural network for acceleration, and observe the effect.

5. A railway transponder radio frequency circuit parameter optimization system, characterized in that: include: The processing module is configured to: perform equivalent analysis on components in high-frequency circuits, extract key characteristic variables of the circuit, establish a multi-objective optimization model for static power consumption and circuit cost, and use a genetic optimization algorithm to optimize parameters of the multi-objective optimization model; wherein, the equivalent analysis on components in high-frequency circuits is used to split the triodes and analyze their internal nodes, and convert them into capacitors and resistors and replace them into the original circuit; the establishment of a multi-objective optimization model for static power consumption and circuit cost is used to construct a multi-objective optimization model for static power consumption and circuit cost for related decision variables, analyze and use appropriate weights and coupling methods, and integrate the decision variables by describing the association and fusion between the decision variables; Among them, an equivalent analysis is performed on the components in the high-frequency circuit to extract the key characteristic variables of the circuit, including: in the voltage-controlled oscillation circuit, a high-frequency equivalent model of the transistor is introduced, including the base body resistance, the collector junction, the emitter body resistance, the emitter junction capacitance and the collector junction capacitance; the emitter junction capacitance is equivalent to two capacitors connected in parallel at the input and output ends of the transistor respectively, and the base body resistance is equivalent to two resistors connected in parallel at the feedback network and the input end of the transistor respectively; for the emitter junction capacitance, it is originally connected between the node b' and the node c of the transistor, and according to the voltage relationship between the node b' and the node c, the capacitance of the two capacitors equivalent to the emitter junction capacitance is obtained in combination with Miller's theorem; for the base body resistance, it is originally connected between the node b and the node b' of the transistor, and according to the voltage relationship between the node b and the node b', the resistance of the two resistors equivalent to the base body resistance is obtained by using Miller's theorem; Among them, establishing a multi-objective optimization model for static power consumption and circuit cost includes: according to the factors affecting the performance of the RF transmission circuit, establishing the optimization decision variables of the RF transmission circuit, and constructing the RF transmission circuit optimization objective function and the circuit static power consumption evaluation function; wherein, the RF transmission circuit performance influencing factors include the loaded quality factor of the frequency selection network, the loop gain and the phase condition; Among them, the genetic optimization algorithm is used to optimize the parameters of the multi-objective optimization model, including: finding a series of balanced solutions in the optimization process by initializing the population, evaluating fitness, and allocating congestion, covering various trade-offs under multiple objectives, and realizing parameter optimization of the circuit static power consumption evaluation function and the circuit cost evaluation function.

6. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by the processor, the method for optimizing the parameters of the railway transponder radio frequency circuit is implemented as described in any one of claims 1 to 4.

7. A computer device, characterized in that: It includes a memory and a processor, the processor and the memory communicate with each other, the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the railway transponder radio frequency circuit parameter optimization method as described in any one of claims 1-4.

8. An electronic device, characterized in that: include: A processor, a memory and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device executes instructions for implementing the method for optimizing the radio frequency circuit parameters of a railway transponder as described in any one of claims 1 to 4.

Images