RU2805901C1 - Method and device for radar determination of movement parameters of cuts on hump - Google Patents

Abstract

FIELD: railway equipment.

SUBSTANCE: means for automatically controlling the speed of movement of cuts on a hump. The device includes an antenna, a transmit-receive unit TRU, a pulse modulator, a programmable synchronization and control unit PSCU, an analog-to-digital converter ADC, and a signal processing unit SPU. Moreover, the high-frequency port TRU is connected to the antenna, the output of the pulse modulator is connected to the first TRU output, and the signal input of the ADC is connected to its second output, the first PSCU output is connected to the input of the pulse modulator, its second output is connected to the SPU through the PCB programming command bus, and the third output is connected to the clock ADC input, and the ADC output port is connected to the first SPU output.

EFFECT: increase in the reliability and accuracy of determining the parameters of the cut movement is achieved under conditions of simultaneous presence of two cuts in the same braking position.

9 cl, 1 dwg

Description

The invention relates to the field of railway (railway) automation and telemechanics, in particular to methods and devices for automatically controlling the speed of movement of cuts on a hump in order to ensure trouble-free dismantling of railway trains, displaying information about the movement of cuts and their location on the hump during the dismantling of trains, as well as documenting the dissolution process, and is based on the use of radar sensors (RLD).

In accordance with the concept of interval-targeted control of the speed of cuts, the discharge part of automated and mechanized hump humps is equipped with brake positions (BP), which are located, as a rule, in front of the dividing arrow of the first TP, behind the dividing arrow of the second TP and at the beginning of the park tracks - the third TP. To regulate the speed of releases at the TP, special track devices, the so-called hump and park retarders, are used.

The main task of the first and second TP, often called hump or top, is to slow down freely rolling trailers from the hump of the hump. This is necessary to eliminate surges from passing cut trailers following specified routes along the marshalling yard path. Braking must ensure the required time intervals between car releases rolling down the hill, sufficient to move the switches along the route, and the speed of the releases at the exit from these positions, which, when the releases approach the third TP, should not exceed 6 m/s. Thus, the main task of so-called interval braking is assigned to hump-type TPs. The tasks of the park TP include targeted braking of car trailers and setting speeds sufficient for them to reach the design point on the sorting track. In this case, the collision speed of the trailers in the park should not exceed 5 km/h.

Radar speed meters, the operating principle of which is based on the use of the Doppler effect (see pp. 33-40, [1]), are used as means to control the parameters of the movement of cuts at the TP. Frequency $$F_{\text{D}}$$ Doppler signal is directly proportional to the release speed:

$$F_{\text{D}}=\frac{2V\cos \alpha }{\lambda }=\frac{2V{f}_0\cos \alpha }{c}$$ , (1)

Where

$$V$$ - speed of movement of the cutter;

$$\alpha$$ - the current angle between the cut speed vector and the direction of its irradiation;

$$\lambda$$ And $$f_0$$ - wavelength and frequency of RLD radiation, respectively;

$$c$$ - speed of propagation of radio waves.

The advantages of radar speed meters are non-contact, low inertia and continuity of the measurement process in the control zone. The range of action on the car in conditions of direct visibility of radar meters reaches 250 m.

The simplest radar speed meters are autodyne devices (see Fig. 2.1, [1]), in which the functions of a probe transmitter and a receiver of radiation reflected from the cutter are simultaneously combined by a microwave generator, called an autodyne in the domestic literature. The operating principle of these devices is based on the autodyne effect, which consists of changes with the Doppler frequency in the amplitude and frequency of oscillations of the microwave generator, as well as the current and/or bias voltage in its power supply circuit [2]. Registration of these changes in the form of autodyne signals and their processing make it possible to determine the parameters of the relative movement of the cuts. Thanks to the combination of the functions of a transmitter and a receiver, autodyne speed meters have minimal overall dimensions and weight, while they have a sufficiently high sensitivity for use as cut speed sensors on a hump [3].

An example of a speed meter based on an autodyne (generator mixer) is described in the patent [4]. The meter contains a microwave generator, which is connected to the transceiver antenna through an adjustable impedance transformer (tuner). A resistor is included in the power supply circuit of the microwave generator, from which the autodyne signal is taken, which enters the signal processing unit through an amplifier.

A more advanced autodyne sensor, which can be used as a universal device for determining the occupancy of a switch or a car speed meter, is described in article [5]. When used as an occupancy sensor, it operates in frequency modulated (FM) mode. In this case, digital processing of the signal converted by the autodyne by a signal processor provides a solution to the problem of detecting the vacancy or occupancy of a turnout. If the sensor is used as a car speed meter, its operating program changes. In this mode, there is no FM radiation, and the sensor emits unmodulated vibrations in the direction of the cut. To obtain information about the speed of cuts, time and spectral methods of processing Doppler signals are used.

The disadvantage of the known autodyne speed meters is their low energy potential compared to meters with a homodyne transceiver design. [6]. This parameter limits the maximum operating range of the meter.

There are known radar speed meters for wagons, containing a “Doppler signal sensor (DSS)” and a measuring signal converter unit (SICU) [7, 8]. In this case, the DDS consists of a transmitting and receiving antenna, which are connected to a microwave generator and a diode mixer, respectively. The output of the diode mixer is connected to the BIPS input through a Doppler frequency signal amplifier. The Doppler signal in the DDS is obtained as a result of frequency conversion in a diode mixer of the radiation received from the moving cutter and part of the reference radiation of the microwave generator passing through the slot junction. DDS antennas are installed inside the track, and other devices are installed between the tracks.

Known devices have the following disadvantages.

The antennas are connected to the microwave generator and diode mixer using an extended waveguide path. The inevitable vibrations of this path during the passage of the cut, in turn, cause amplitude-phase modulation of the radiation and cause the formation of interference at the output of the diode mixer, which disrupts the normal processing of the Doppler signal and limits the range of the device (see pp. 393-396, [ 9]).

Installing a DDS inside the track is also not a good solution, since its maintenance in winter is difficult. The device often fails, especially in spring and in rainy weather, since the DDS is flooded with water [10]. In addition, the presence of two antennas in the meter significantly complicates the design of the device, and also increases its overall dimensions and cost, which is its additional disadvantage.

There are known radar speed meters for cars, the transceiver of which is made according to a circuit with direct (homodyne) conversion of the Doppler signal (see pp. 33-37, Fig. 2.5, [1]; [11]). These devices use a single antenna that works for both transmission and reception. Continuous radiation from the output of the microwave generator, passing along the path of the first and second ports of the circulator, enters the antenna and is radiated in the direction of the moving cut. The radiation reflected from the cut is shifted in frequency by the amount of the Doppler shift (1), proportional to the speed of movement of the cut. This radiation, through the antenna and along the path of the second - third ports of the circulator, enters one of the arms of the mixer, to the other arm of which the reference oscillations of the microwave generator are supplied. As a result of converting the oscillation frequency, a difference (Doppler) frequency signal is isolated at the output of the mixer, which then, after amplification and pre-filtering, enters the signal processing unit.

However, the above-mentioned car speed meters have a significant drawback, which is as follows.

A characteristic feature of hump speed meters is that they belong to short-range radar systems (SLRS), in which the distances between the meter and the car are commensurate with the spatial separation of concentrated reflective elements (“shiny” points) of the car structure [12]. Under such conditions, reflected radiation acquires a multiple character. This significantly complicates the structure of the reflected signal, in the formation of which not only amplitude, but also phase relationships between reflections from various elements of the car become crucial. Processing the received signal and extracting information from it about the parameters of the cut's movement are additionally complicated by the need to take into account the dynamics of changes in the radar picture. So, at the maximum distance of the cut from the speed meter, the end wall of the car is mainly irradiated. Reflective elements on its surface are concentrated in a relatively narrow sector. As the cut approaches the meter, the area of the irradiated area changes, and the irradiation area itself moves to the side of the car (see Fig. 2.4, [1]). Large, “brighter” “shiny” points fall into the irradiation zone, for example, vertical corner, intermediate, pivot and door posts, which form corner reflectors. Significant elements of the reflection pattern are the automatic coupler, pallet elements, as well as smaller reflective elements - locks, brake rods, etc. All of the listed reflective elements located in the radiation field of the radar meter create a set of elementary (partial) Doppler signals received from individual parts of the surface carriage. With existing measurement ranges and small irradiation angles, even small changes in the position of the cutter at the braking position lead to significant changes in the phase of partial signals. As a result of the interference of partial signals at the output of the receiving device (mixer), phase jumps of the resulting signal are observed, as well as its short-term and long-term loss (fading). The noted features of signal formation during their processing are not taken into account in the speed meters mentioned above. For this reason, these devices did not provide the required accuracy in measuring the parameters of the movement of cuts, and were taken out of service on hump yards.

To combat signal loss, the patent [13] proposes a method based on searching for the value of the phase difference of partial signals at which there is no signal fading. To do this, the radiation frequency of the microwave generator in the process of signal formation is modulated according to a sawtooth law. Then the signal received from the cutter is mixed with emitted oscillations, the resulting mixture is converted in a mixer, the result of the conversion is rectified, and within the modulation interval this point in time is determined $$t_1$$ from the beginning of the interval $$t_0$$ , in which the amplitude of the resulting oscillation is maximum. In subsequent intervals $$t_{01}-t_{02}$$ , $$t_{02}-t_{03}$$ etc. new samples are obtained until the Doppler frequency of the signal is measured.

The disadvantage of the proposed method is a significant expansion of the radiation spectrum of the microwave generator due to frequency modulation. Thus, with a path difference of rays of partial reflections of 0.25 m, the frequency deviation should be of the order of 300 MHz.

There are known radar speed meters for cut movement, the transceiver of which is also made according to a homodyne circuit with direct conversion of the Doppler signal and one antenna operating both for transmission and reception (see pp. 37-40, Fig. 2.7, [1]; [ 14-22]; pp. 108-116, [23]). These meters are distinguished by the presence in the signal processing unit of technical solutions that reduce the impact on the measurement accuracy of deep signal fading due to the multiple nature of the reflected radiation from the car.

Devices described on pages 37-40, fig. 2.7 of the book [1] and in articles [14-18], contain a measuring signal converter designed to convert the frequency of the Doppler signal into a constant voltage. The converter is made on the basis of a sawtooth pulse generator, which remembers and maintains for a given time (usually two seconds) a constant frequency of the output signal and, accordingly, the value of the output voltage of the converter, starting from the moment of its loss.

Technical solutions for restoring missing pulses in the Doppler signal, proposed in [19, 20], are based on the use of a tracking filter based on a phase-locked loop (PLL) system. In the event of a signal loss, the PLL feedback circuit opens and the voltage controlled oscillator (VCO) maintains the previous signal frequency at the filter output for a set time.

The device, made in accordance with the patent [21], contains in the signal processing unit two loops for monitoring changes in both the frequency and phase of the Doppler signal, made on the basis of a digital phase detector and a digital frequency detector. The device also contains a selection unit for the duration of the input signal loss, a reference generator, and a logical communication and control unit. The proposed device allows, when restoring missing impulses, to reduce the dynamic error in measuring the parameters of the movement of cuts at the TP.

There is also a known method for measuring the speed of cuts and a radar device that implements it, according to the patent [22]. The transceiver of this device operates with continuous radiation of radio waves; it is made according to a homodyne circuit with direct conversion of the Doppler signal and one antenna. The processing unit of the device uses modern methods of digital signal processing: discrete Fourier transform (DFT), Goertzel algorithm, weight processing of input data, summation of input data with time superposition and “jumping” DFT. This invention is implemented in a radar speed meter model RIS-V3M [23].

From the presented analysis of the level of technology it follows that during the development of RLDs, made mainly in the period before 2000, preference was given to simpler and cheaper principles of their construction. Homodyne and autodyne continuous radiation devices without modulation were used as transceivers, and Doppler signal processing is based on measuring its frequency.

Currently, thanks to the development of the element base of microwave microelectronics, the development of standard modules for microwave transceiver devices and methods of digital signal processing, these limitations have become insignificant. In this regard, for the implementation of a new generation of RLDs, such methods of generating sounding and processing reflected radio signals are in demand, which more fully, using the properties of signals and their capabilities, ensure an increase in the technical and operational characteristics of newly created measuring instruments.

As a prototype, we adopted the RIS-V3M radar speed meter, which today is the most advanced and widely used in the Russian Railways marshalling station network. Its technical description is presented in the textbook: Shelukhin V.I. Automation and mechanization of hump humps: Textbook for technical schools and railway colleges. transport. - M.: Route, 2005. - 240 pp., (see pp. 108-116, Fig. 3.26, [23]).

The RLD contains (see Fig. 3.26, [23]) a microwave transceiver block, an amplifier-filter and a signal processing unit, and the microwave transceiver block consists of an antenna, a microwave generator, a circulator and a mixer, with the output of the microwave generator connected to the first the circulator port, to the second port of which the antenna is connected, the mixer input is connected to the third port of the circulator, and a signal processing unit (SPU) is connected to the output of the latter through an amplifier-filter and a digital-to-analog converter (DAC). The BOS contains a codec consisting of the aforementioned ADC and a digital-to-analog converter (DAC) connected in series, a read-only memory (ROM), a processor, an interface transceiver in the RS-485 protocol standard for communication via a two-wire line with the computer of the hump automatic centralization system (HAC) , a voltage-frequency converter of a square wave pulse sequence signal, the frequency of which corresponds to the frequency of the received Doppler signal. In addition, the BOS contains means for monitoring the performance of the RLD.

According to the description [23], the operating principle of the prototype is based on a method for radar determination of the parameters of the movement of cuts on a hump, which consists of the following.

Using the antenna of a radar sensor (RSD), installed at the beginning or end of the controlled section of the transformer substations, they emit probing microwave radiation at the height of the car automatic coupling, formed in the form of a pencil radiation pattern, coinciding in direction with the diagonal of the controlled section of the transformer substations, and receive reflected microwave radiation from “shiny” points located within the antenna radiation pattern on the surface of the moving cut, mix it with part of the probing radiation, converting the resulting mixture of reflected and probing radiation into a set of partial Doppler frequency signals from “brilliant” points on the surface of the moving cut, then the converted set of partial signals are sent to the signal processing unit, where the speed of movement of the cut is determined based on the frequency of the signal, taking into account the angle between the direction of radiation and the axis of the railway track of the controlled section of the TP.

However, the prototype and known analogues have common significant disadvantages, which are as follows.

Experience in operating meters at hump yards shows that the duration of signal loss at the output of the transceiver unit due to interference fading reaches 40% of the duration of the entire reflected signal in the process control area. The duration of fading can significantly exceed both the measurement time and the time the meter is in memory mode, and the greatest fading is observed precisely in that section of the TP where speed control is necessary. Moreover, in the cut-off braking mode, the magnitude of the negative acceleration reaches 3 m/s ² or more. For example, in the process of controlling retarders, sometimes non-standard situations arise, up to a complete stop of the release at the TP. Under these conditions, measuring the release speed with the required accuracy is very difficult or becomes completely impossible.

The simplest signal restoration schemes, for example, those using a generator with “memory”, when the signal disappears, form a time-limited (up to 2 s) series of pulses of a constant frequency. The duration of this series, as a rule, is less than the duration of the actual signal loss, especially at low speeds. In addition, the dynamics of the release movement are not taken into account. During intense braking, the speed, i.e. the Doppler frequency changes, and the signal restoration circuit generates a sequence of pulses of constant frequency. The difficulty of accurately reconstructing the signal is also aggravated by the wide range of speeds and the dynamics of its change. At low cut speeds, longer recovery times are required than at high speeds.

From the presented analysis of the shortcomings of known RLDs, it follows that the way to improve the method of restoring signal loss (which actually does not exist) is ineffective, it has exhausted itself. Any method, no matter how perfect it is, cannot guarantee accurate reconstruction of the signal reflecting the real dynamics of the movement of the cut at the TP. Therefore, to solve this problem, other technical solutions are in demand, aimed, for example, at improving the quality of the signal by eliminating or reducing the causes of its loss.

Another significant drawback of the known meters of cut movement parameters is that they do not provide reliable measurement results under conditions of simultaneous monitoring of two cuts at one TP. The fact is that the technology for dismantling cars at hump yards, especially high-power ones, allows for such an interval of passing cuts that there can be two cuts at the transfer point at the same time (see pp. 61-71, [10]). In this case, if one meter is installed on the TP, signals reflected from two cuts will simultaneously enter its input. The desire to eliminate this phenomenon by placing two meters on the TP, working for each moderator, on the one hand, significantly complicates the signal processing algorithm in moderator control systems (see pp. 121-124, [23]). On the other hand, such placement does not eliminate the zone within the braking position in which both meters will receive the signal reflected from one release. Moreover, when installing two meters on the TP, the irradiation angle of the moving cut increases. In this case, the meter's coverage area may include trailers moving along adjacent tracks. In addition, an interference signal in the form of beats may appear at the output of mixers when directly hit by radiation from meters with similar frequencies of microwave generators. Therefore, in order to solve this problem, in our opinion, what is in demand is not the search for the optimal placement of meters on the TP [10; 23], and new technical solutions for RLDs that limit the working areas of the meters and protect them from interference.

Thus, the technical problem to be solved by the claimed invention is the need to eliminate or significantly reduce the disappearance time of the reflected signal due to the multiple nature of its formation, which is required to increase the reliability and accuracy of determining the parameters of the movement of cuts. In addition, the functionality of obtaining reliable measurement results in conditions of the simultaneous presence of two cuts on one TP is in demand. The lack of range resolution of known meters and insufficient resistance to the effects of signals received from moving vehicles on adjacent tracks are their significant limitations, especially in modern conditions of increased train dissolution rates and the complexity of radar situation scenarios at a hump. In this regard, the task of removing the noted restrictions is urgent.

To solve this problem, a method has been proposed for radar determination of the parameters of the movement of cuts on a hump, which consists in the fact that by means of a radar sensor antenna (RSD) installed at the beginning or end of the controlled section of the TP, probing radio pulses of microwave radiation in the form of a pencil are emitted at the height of the car automatic coupler directional pattern coinciding in direction with the diagonal of the controlled section of the TP, during the emission of probing radio pulses, they receive radio pulses reflected from the “shiny” points located within the antenna radiation pattern on the surface of the moving cut, mix them with part of the energy of the probing radio pulses of radiation, convert the overlapping in time parts of these radio pulses into video pulses, sample them in time and store them at multiple points in time $$t_{kz}$$ samples of the instantaneous values of these video pulses $$u_{\text{etc}.kz}$$ , Where $$k=\mathrm{0,}\mathrm{1,}\mathrm{2,}\mathrm{...}$$ - serial number of the video pulse; $$z=\mathrm{0,}\mathrm{1,}\mathrm{2,...}Z$$ - serial number of the sample inside the video pulse, then from the current samples $$u_{\text{etc}.kz}$$ , beginning with $$z=1$$ , subtract the previous sample values $$u_{\text{etc}.k(z-1)}$$ :

$$x_{kz}=(u_{\text{etc}\text{.}kz}-u_{\text{etc}\text{.}k(z-1)})$$ ,

and according to the obtained difference values $$x_{km}$$ , using many samples along $$k$$ video pulses are obtained for each sample $$z$$ spectrum of the Doppler signal, for example, by performing a fast Fourier transform (FFT) operation using the “Radix2” algorithm using the Hanning window function, for which the amplitude is calculated $$A_z$$ and frequency $${\Omega }_z$$ signal, then for the received amplitude samples $$A_z$$ calculate signal-to-noise ratio values $$q_z=A_z/{\sigma }_{\text{w}}$$ , Where $${\sigma }_{\text{w}}$$ - root mean square value of the noise level at the receiver output, then from a set of values $$q_z$$ signal-to-noise ratios select those sequence number values $$z$$ , amplitudes $$A_z$$ and frequencies $${\Omega }_z$$ , at which the signal-to-noise level exceeds the threshold value: $$q_z\ge q_{\text{since then}}$$ , Where $$q_{\text{since then}}$$ - threshold signal-to-noise ratio, then for selected amplitudes $$A_z$$ and frequencies $${\Omega }_z$$ in ascending order of number $$z$$ determine the number of readings $$z_n$$ for signals received from $$n$$ -th “shiny” point, after which the current distance is calculated $$R_n$$ to each of $$n$$ "shiny" points on the surface of the cut according to the formula:

$$R_n=\frac{z_n{t}_{\text{And}}c}{2Z}$$ ,

Where $$z_n$$ - number of samples $$z$$ before $$n$$ -th “shiny” point; $$t_{\text{And}}$$ - duration of the radio pulse; $$c$$ - speed of propagation of microwave radiation; $$Z$$ - number of samples per time $$t_{\text{And}}$$ , then calculate the current positions of the “shiny” points on the cut relative to the reference line - perpendicular from the place where the RLD is installed on the rail track, using the following formula:

$$l_{\text{father}\text{.}n}=\sqrt{R_n^2-{[l_{\text{RLD}}-(b/2)]}^2}$$ ,

Where $$l_{\text{father}\text{.}n}$$ - current distance $$n$$ -th “shiny” point from the reference line (perpendicular from the place where the RLD is installed on the rail track); $$R_n$$ - current distance from the RLD installation location to $$n$$ -th “shiny” point; $$l_{\text{RLD}}$$ - the distance from the location of the RLD installation between tracks to the axis of the railway track, stored in the RAM memory of the DSP; $$b$$ - car width (see Fig. 3.28, [23]), then from the set of current distance values $$l_{\text{father}\text{.}n}$$ “shiny” points are chosen those values for which the two-sided inequality holds:

$$l_{\min }\le l_{\text{father}\text{.}n}\le l_{\max }$$

Where $$l_{\text{father}\text{.}n}$$ - current distance $$n$$ -th “shiny” point from the reference line (perpendicular from the location of the RLD installation on the rail track); $$l_{\text{min}}$$ , $$l_{\text{max}}$$ - the near and far boundaries of the measuring section of the TP, which are determined as the distance from the reference line (perpendicular from the installation site of the RLD on the rail track), then for the “shiny” points that satisfy this inequality, from the data array for frequencies $${\Omega }_n$$ and distances $$R_n$$ compile and store data arrays:

$${\overrightarrow{\Omega }}_m=[{\Omega }_1{\Omega }_2{\Omega }_3\mathrm{...}{\Omega }_M]$$ , $${\overrightarrow{R}}_m=[R_1{R}_2{R}_3\mathrm{...}{R}_M]$$ ,

Where $${\Omega }_m$$ - frequency of the Doppler signal spectrum obtained from $$m$$ -th “shiny” point (here $$m=\mathrm{1,}\mathrm{2,}\mathrm{...}M$$ ); $$R_m$$ - current distance from the radar to $$m$$ -th “brilliant” point, after that, based on the obtained values of Doppler frequencies $${\Omega }_m$$ and distances $$R_m$$ calculate ground speed $$V_{\text{father}}$$ movement of the release, according to the following formula:

$$V_{\text{father}}=\frac{c}{4\pi f_{0}M}{\displaystyle \sum _{m=1}^M\frac{{\Omega }_m{R}_m}{\sqrt{R_m^2-{[l_{\text{RLD}}-(b/2)]}^2}}}$$ ,

Where $$c$$ - speed of propagation of radio waves; $$f_0$$ - cyclic frequency of radiation; $$M$$ - number of “shiny” points on the cut surface for which Doppler frequency values were obtained $${\Omega }_m$$ and current distance $$R_m$$ ; $${\Omega }_m$$ - Doppler frequency value $$m$$ -th “shiny” point; $$R_m$$ - current distance from the RLD installation location to $$m$$ -th “shiny” point; $$l_{\text{RLD}}$$ - distance from the location of the RLD installation between tracks to the axis of the railway track; $$b$$ - width of the car, then the obtained ground speed values $$V_{\text{father}}$$ movement of the release sequentially when the number of the video pulse changes $$k$$ smoothed with reference to the moments of formation $$k$$ th video pulse in the form $${\widehat{V}}_{\text{father}}(t_k)$$ , then using the smoothed values of the velocity data array $${\widehat{V}}_{\text{father}}(t_k)$$ when changing the video pulse number $$k$$ , perform numerical integration according to the expression:

$$L_{\text{father}}={\displaystyle \underset{t_{\text{input}}}{\overset{t_{\text{out}}}{\int }}{\widehat{V}}_{\text{father}}(t_k)d{t}_k}-L_{\text{ku}}$$ ,

Where $$L_{\text{father}}$$ - cut length; $${\widehat{V}}_{\text{father}}(t_k)$$ - smoothed value of the ground speed of the controlled trailer at the current time $$t_k$$ , tied to $$k$$ -th video pulse; $$t_{\text{input}}$$ , $$t_{\text{out}}$$ - respectively, the moments of entry and exit of the controlled cut to the controlled section of the TP; $$L_{\text{ku}}=l_{\text{max}}-l_{\text{min}}$$ - length of the controlled section of the transformer substation; $$l_{\text{min}}$$ , $$l_{\text{max}}$$ - near and far boundaries of the controlled section of the TP, further, using smoothed values of the speed data array $${\widehat{V}}_{\text{father}}(t_k)$$ , when changing the video pulse number $$k$$ by numerical differentiation, the uncoupling acceleration is calculated according to the expression:

$$a_{\text{father}}(t_k)=\frac{d}{d{t}_k}{\widehat{V}}_{\text{father}}(t_k)=\frac{{\widehat{V}}_{\text{father}}(t_k+\Delta t_{\text{change}})-{\widehat{V}}_{\text{father}}(t_k-\Delta t_{\text{change}})}{2\Delta t_{\text{change}}}$$ ,

Where $$a_{\text{father}}(t_k)$$ - acceleration of the movement of the cutter in the controlled section of the TP at the current time $$t_k$$ , tied to $$k$$ -th video pulse; $${\widehat{V}}_{\text{father}}(t_k)$$ - smoothed value of the ground speed of the controlled trailer at the current time $$t_k$$ , tied to $$k$$ -th video pulse; $$\Delta t_{\text{change}}$$ - time interval for measuring acceleration. At the same time, the duration $$t_{\text{And}}$$ probing radio pulses of microwave radiation are selected from the condition:

$$t_{\text{And}}\ge \frac{2}{c}\sqrt{{[l_{\text{RLD}}-(b/2)]}^2+l_{\max }^2}$$ ,

Where $$c$$ - speed of propagation of radio waves; $$l_{\text{RLD}}$$ - distance from the location of the RLD installation between tracks to the axis of the railway track; $$b$$ - carriage width $$l_{\text{max}}$$ - distance to the far border of the controlled section of the TP.

To implement this method, a device is proposed for determining the parameters of the movement of cuts on a hump, containing an antenna, a transmitting and receiving unit PPB, a pulse modulator, a programmable synchronization and control unit PBSU, an analog-to-digital converter ADC, as well as a BOS signal processing unit, and the high-frequency port PPB is connected to antenna, the output of the pulse modulator is connected to the first output of the PPB, and the signal input of the ADC is connected to its second output, the first output of the PBSU is connected to the input of the pulse modulator, its second output is connected to the BOS through the ShKP programming command bus, and the third output is connected to the clock input of the ADC , and the output port of the ADC is connected to the first output of the BOS, and:

The PPB is made according to a homodyne circuit and contains a microwave generator, a mixer and a circulator, with the microwave generator connected by its high-frequency port to the first port of the circulator, and the second to the antenna, while the input of the mixer is connected to the third port of the circulator, and the modulation input of the microwave generator and the mixer output are connected to the first and second terminals of the PPB, respectively;

The PSB is made according to an autodyne circuit based on a microwave generator connected to the antenna, containing in its oscillatory system or connected to it through a transmission line an amplitude detector, and the modulation input of the microwave generator and the output of the amplitude detector are connected to the first and second terminals of the PPB, respectively;

The PBSU contains a reference clock generator connected to the inputs of the programmable PUCH multiplier and the PDCH frequency divider, while the PDCH output is the first output of the PBSU, the second output is the bus of programming commands of the ShKP, the outputs of the PDCH and PDCH are connected to the inputs of the SI pulse selector, the output of which is the third output PBSU.

The BOS contains a central signal processor DSP, which consists of read-only ROM and RAM storage devices, a computing core, the first PP-1 and second PP-2 data transceivers (mutual connections within the DSP are shown conditionally due to their virtuality), as well as a digital-to-analog converter DAC, linear voltage-frequency converter PN-F signal of the square wave type, signal inversion cascade and paraphase amplifier.

As follows from a comparison of known and proposed methods and devices, the technical result of solving this problem is achieved through the use of a new method for generating probing radiation and processing the reflected radio signal, introducing additional blocks and nodes (IM and PBSU), as well as new connections between blocks and nodes. The proposed technical solutions made it possible to separate the mixture of Doppler signals at the output of the PPB, obtained from “shiny” points on the surface of a moving cut, into separate independent partial components that are not subject to interference. Accordingly, this eliminates the main drawback of the prototype and known analogues - fading (loss) of the signal. In addition, the proposed RLD has a new property - the ability to separately measure the speed of passing cuts on different TP retarders and increased resistance to interference from vehicles moving on adjacent tracks.

As a result of the search for alternative solutions in the field of using RLDs in railway transport, among various sources of information, the fact of using these technical solutions was not found (see, for example, literature: [24-26]). Sources of information revealing the essence of the proposed invention were also not found in the literature on radar (see, for example, literature: [27-29]). Based on the above, it can be argued that the proposed technical solution has significant differences from the prototype and meets the “Novelty” criterion.

This solution is associated with obtaining new properties of the device, which do not clearly follow from the prior art, and meets the “Inventive Step” criterion.

The invention is aimed at improving the characteristics of the RLD for control systems for retarders on the descent part of the hump, which is necessary to increase its processing capacity and reduce potential losses from derailment of cars in the event of failure of floor equipment. Therefore, such a device is in demand on the railway network and can be produced industrially, since its manufacture requires conventional radio-electronic components. Thus, the claimed invention meets the criterion of “Industrial applicability”.

The essence of the invention is illustrated by drawings.

In fig. Figure 1 shows a block diagram of the RLD, which contains antenna A, transmitter-receiver unit PPB, pulse modulator IM, programmable synchronization and control unit PBSU, analog-to-digital converter ADC, signal processing unit BOS, programming command bus ShKP, frequency output bus ShChV, digital data bus SDC, digital-to-analog converter DAC, voltage-frequency converter PN-CH, inverter INV, first U1 and second U2 amplifiers, RS-485 bus interface. The BOS signal processing unit is made on the basis of a digital signal processor DSP, while the DSP contains a read-only memory device ROM, a computing core and a random access memory device RAM, as well as the first and second transceivers of the serial port buses PP-1 and PP-2. F1 and F2 are paraphase signal outputs in the form of a meander, the frequency of which is directly proportional to the cutting speed.

In fig. Figure 2 shows homodyne ( a ) and autodyne ( b ) versions of the PPB, where A is the antenna; SHF-G - microwave generator; C - circulator; SM - mixer; AD - amplitude detector.

In fig. Figure 3 shows one of the variants of the PBSU, which contains a built-in VKP programming controller, a reference OTG clock generator, a programmable PUC multiplier and a PFC frequency divider, and an SI pulse selector.

In fig. Figure 4 shows time diagrams of the processes of generating and processing signals in the RLD to determine the parameters of the movement of cuts on the hump: ( a ) - output voltage $$u_{\text{them}}(t)$$ pulse modulator; ( b ) - probe voltage $$u_{\text{probe}}(t)$$ radio signal; ( v ) - voltage $$u_{\text{negative}\text{.1}}(t)$$ radio signal received (reflected) from the first “shiny” point of the moving cut; ( g ) - voltage $$u_{\text{negative}\text{.2}}(t)$$ radio signal received (reflected) from the second “shiny” point of the moving cut; ( d ) - voltage of the converted $$u_{\text{etc}}(t)$$ signal at the output of a SM mixer or an autodyne microwave generator; ( e ) - clock voltage $$u_{\text{you}}(t)$$ on the third output of the PBSU; ( w ) - voltage $$u_{\text{b-z}}(t)$$ signals after the sampling-storage (storage) operation; ( z ) - voltage $$u_{\text{calc}}(t)$$ signals after the operation of subtracting previous samples from the current values.

The RLD (see Fig. 1) contains an antenna A connected to the PPB transceiver unit, a programmable synchronization and control unit PBSU, connected with its first output to the input of the pulse modulator IM, the output of which is connected to the first output of the PPB, and the second output of the PPB is connected to the signal input of the analog-to-digital converter ADC. The ADC output is connected to the first (signal) pin of the BOS signal processing unit. In this case, the second output of the PBSU is connected via the ShKP programming command bus to the second output of the BOS, and the third output of the PBSU is connected to the clock input of the ADC. The third output of the BOS is connected to the input of the digital-to-analog converter DAC, the output of which is directly connected to the voltage-frequency converter PN-CH, and the output of the latter is connected to the input of the first amplifier U1 and, through the INV inverter, to the input of the second amplifier U2. The fourth output of the BOS is connected to the computer of the hump automatic centralization system of the GAC through the digital data bus SCD, an interface in the RS-485 protocol standard and a two-wire line.

Antenna A (see Fig. 1) can have various designs, for example, in the form of a slot or strip phased array, horn, dielectric, lens antenna or horn-lens antenna (see page 110, Fig. 3.26 in [23]; pp. 21, 114, 142, 151, 200 in [30]).

PPB also has alternative technical solutions. It can be made according to a homodyne (see Fig. 2, a ) or autodyne (see Fig. 2, b ) scheme. The first option is presented in the description of the RIS-V3M manual [23], adopted as a prototype, while both embodiments of the microwave module are described in patent US3750171, 07/31/1973 (see Fig. 1 and 2).

An example of implementing a PPB using a homodyne circuit, shown in Fig. 2, a , contains a microwave generator SHF-G, a mixer SM and a circulator C, while the microwave generator SHF-G is connected by its high-frequency port to the first port of the circulator C, and its second port to antenna A, and to the third port circulator C is connected to the input of the mixer SM, and the modulation input of the microwave generator SHF-G and the output of the mixer are connected to the first and second terminals of the PPB, respectively (see page 110, Fig. 3.26 in [23]). Homodyne PPB modules are produced by industry (see, for example, the description of Doppler modules: “24.125 GHz Ranging Sensor Head, Dual Channel, Short Range”, Model SSD-24303-20M-DW on the website https://www.eravant.com/) .

An example of the implementation of PPB according to the autodyne scheme, presented in Fig. 2, b , contains a microwave generator SHF-G, connected by its high-frequency port to antenna A and to the amplitude detector AD. Structurally, the detector diode of the amplitude AD detector is usually placed directly into the resonator of the autodyne microwave generator SHF-G or into the transmission line connected to the resonator, as shown in Fig. 2 patents RU2295911С1, publ. 03/27/2007, bulletin. No. 9 and in Fig. 6 a and 9 a of article [2]. The modulation input of the microwave generator SHF-G and the output of the amplitude detector AD are connected to the first and second terminals of the PPB, respectively.

The PPB may include additional elements that do not change the essence of the invention. For example, after the SM mixer and the amplitude detector AD, a low-noise amplifier and a Doppler frequency signal filter can be installed, and between the microwave generator SHF-G and the first port of the circulator Ts - a power amplifier.

The microwave generator SHF-G can be made, for example, in the form of a microwave generator module in a volume or strip design based on a transistor (see page 88, Fig. 3.7 of the book [31]), on a Gunn diode or an avalanche-flight diode in waveguide or stripline design (see pages 194, 195, Fig. 4.24 and 4.25, [32]). The oscillation frequency of the microwave generator SHF-G can be stabilized by a frequency synthesizer or using an additional high-quality resonator, which does not change the essence of the proposed invention.

The pulse modulator IM is designed to generate a pulse effect on the microwave generator SHF-G, during which, during this pulse, microwave oscillations are excited in the microwave generator SHF-G. Examples of circuit solutions for pulsed modulators in the power supply circuit for microwave generators microwave-G, made on Gunn diodes, avalanche diodes or transistors are widely known (see Fig. 15 a , Fig. 22 a of article [2]).

PBSU can be made on the basis of “hard” logic, FPGA or using specialized microcircuits. One of the implementation options for the PBSU (see Fig. 3), made on the Si5368 chip, contains a built-in VCP programming controller, a reference OTG clock generator connected to the inputs of the programmable PUCH frequency multiplier and the PDCH frequency divider, as well as an SI pulse selector. In this case, the output of the MAP is connected to the first input of the SI and is the first output of the PBSU, and the output of the PDCH is connected to the second input of the SI, while the output of the SI is the third output of the PBSU. This microcircuit is characterized by a low level of output oscillation phase jitter in the frequency range from 2 kHz to 1.4 GHz (see the Silicon Laboratories website: http://www.silabs.com).

It is preferable to use high-speed microcircuits as ADCs [33; 34]. For example, the AD9689 chip from Analog Devices is a dual 14-bit ADC with a JESD204B interface, speed 2.6 GB/s (see website: https://www.analog.com/ru/products/ad9689.html# product-overview). This ADC is capable of directly sampling analog signals with -3 dB bandwidth up to 9 GHz. Similar ADCs such as DAC38RF82 and DAC38RF89 are produced by Texas Instruments (see website https://www.ti.com/).

The BOS (see Fig. 1) simultaneously performs the functions of generating commands for programming the PBSU, processing signals, converting the frequency of the Doppler signal into a digital code and exchanging digital data with the GAC. It can be implemented on the basis of a signal processor chip, for example, the TMS320F2808 type from Texas Instrument [35]. The DSP includes blocks that perform the following functions: a universal asynchronous transceiver bus of the first port PP-1, which exchanges information with the computer of the GAC system via the RS-485 interface; bus transceiver of the second serial port PP-2; read-only memory (ROM) storing the signal processing program, controls and constants necessary for signal processing; a high-speed computing core that performs all the functions of digital signal processing (spectral analysis, digital signal filtering and data generation to generate the frequency output of the RLD and display information); random access memory (RAM), which performs the functions of storing current values and signal processing results.

The digital-to-analog converter (DAC) of the BOS signal processing unit is designed to convert a sequence of digital data on the cut speed coming from the third port of the DSP into voltage. The principles of DAC design are widely known, and a wide range of DAC chips are produced by industry (see, for example, [36]).

Voltage-frequency converter PN-Ch, designed to convert the output voltage of the DAC into an analog pulse signal of the square wave type, the frequency of which is directly proportional to the uncoupling speed. PN-Ch can be implemented on the AD654JN chip, manufactured by Analog Devices (see website: www.analog.com).

The INV inverter, the first U1 and the second U2 amplifiers are designed to obtain two antiphase meanders F1 and F2 from a unipolar pulse signal at the output of the voltage-frequency converter PN-CH. These nodes can be implemented on operational amplifiers (see page 110, Fig. 3.26, [23]; Fig. 2.1, Fig. 2.2, pp. 31-33, [37]).

The RLD for determining the parameters of the movement of cuts on a hump works as follows.

After applying voltage to the RLD from the power source (not shown in Fig. 1) in the DSP BOS, the computing core, in accordance with the “Installation” subroutine [35], first configures the peripheral devices, distributes the internal memory, sets the values of internal variables, and copies the executable command code from low-performance ROM to high-performance RAM and issuing the command “Sampling from the ADC and storing results in memory,” by which the DSP goes into readiness mode for receiving digitized signals from the ADC with the subsequent formation of a data array in the RAM memory. After its completion, through the second port of the BOS DSP, a sequence of commands for programming its parameters is issued to the PBSU.

In the PBSU (see Fig. 3), after applying the supply voltage and receiving a sequence of programming commands from the BOS, the OTG reference clock generator is launched. The output signal of this generator is fed to a programmable frequency multiplier PUM, at the output of which clock pulses with a repetition period are formed $$T_{\text{you}}$$ , as well as to a programmable frequency divider of the MAP, at the output of which pulses with a duration of $$t_{\text{And}}$$ with repetition period $$T_{\text{P}}$$ to control the PPB transmitter (see diagram a in Fig. 4). At the same time, the duration $$t_{\text{And}}$$ probing radio pulses of microwave radiation are selected from the condition:

$$t_{\text{And}}\ge \frac{2}{c}\sqrt{{[l_{\text{RLD}}-(b/2)]}^2+l_{\max }^2}$$ ,

Where $$c$$ - speed of propagation of radio waves; $$l_{\text{RLD}}$$ - distance from the location of the RLD installation between tracks to the axis of the railway track; $$b$$ - carriage width $$l_{\text{max}}$$ - distance to the far border of the controlled section of the TP. These pulses, which also arrive at the first input of the SI pulse selector (see Fig. 3), form bursts of pulses from 0 to $$Z$$ (see diagram e for case $$Z=10$$ in Fig 4), filling the time interval $$t_{\text{And}}$$ , which are then fed to the clock input of the ADC. In this case, the number of clock pulses per time $$t_{\text{And}}$$ determines the range resolution of the radar: $$\Delta r=t_{\text{And}}c/2Z$$ , Where $$Z$$ - number of samples per time $$t_{\text{And}}$$ .

Pulses for starting the PPB transmitter from the first output of the PBSU $$u_{\text{them}}(t)$$ , (see diagram a in Fig. 4) are fed to the input of the pulse modulator IM, which at its output generates a sufficiently powerful pulse to trigger the microwave generator microwave-G, providing it with the conditions for excitation of microwave oscillations. Probing radio pulses generated by the microwave generator SHF-G $$u_{\text{probe}}(t)$$ with homodyne execution of the PPB (see Fig. 2, a ) they pass through the circulator C along the path first port - second port to antenna A. With autodyne execution of the PPB (see Fig. 2, b ) the sounding probes formed in the microwave generator radio pulses $$u_{\text{probe}}(t)$$ immediately enter antenna A. Expression for sounding radio signals $$u_{\text{probe}}(t)$$ (see diagram b in Fig. 4) with a rectangular envelope $$U_{\text{probe}}\text{(}k,t\text{)}$$ has the form [38]:

$$u_{\text{probe}}(t)=A_0{\displaystyle \sum _{k=1}^{\infty }{U}_{\text{probe}}\text{(}k,t{\text{)cos(ω}}_{\text{0}}t+{\phi }_{\text{k}}\text{)}}$$ , (1)

$$A_0$$ - amplitude of the probing radio pulse;

$$U_{\text{probe}}(k,t)=\{\begin{array}{l}\mathrm{1,}\text{at}k{T}_{\text{P}}<t<k{T}_{\text{P}}+t_{\text{And}}\\ \mathrm{0,}\text{at}k{T}_{\text{P}}+t_{\text{And}}<t<(k+1)T_{\text{P}}\end{array}$$ - unit function of the probing radio pulse;

$${\omega }_0=2\text{π}{f}_0$$ - circular filling frequency of the probing radio pulse;

$$t_{\text{And}}$$ And $$T_{\text{P}}=2\pi {\text{/Ω}}_{\text{P}}$$ - duration and repetition period of radio pulses;

$${\phi }_k$$ - random initial phase $$k$$ th radio pulse;

$$k$$ - integer, serial number of the probing radio pulse.

In antenna A, microwave oscillations are converted into electromagnetic radiation, which, in accordance with the radiation pattern of antenna A, is emitted into the controlled space of the TP.

It should be noted that with homodyne implementation of the PPB (see Fig. 2, a ) part of the power of the radio signal of the microwave generator microwave-G along the path first port - third port of the circulator C penetrates the port of the SM mixer as a heterodyne oscillation. We will call these oscillations “direct” (index “prm”) $$u_{\text{prm}}\text{(}t\text{)}$$ , are generally written as

$$u_{\text{prm}}\text{(}t\text{)}=A_{\text{prm}}{\displaystyle \sum _{k=1}^{\infty }{U}_{\text{prm}}\text{(}t,k\text{)}\text{cos(}{\omega }_{\text{0}}t+{\phi }_{\text{k}}\text{)}}$$ , (2)

Where

$$A_{\text{prm}}$$ - amplitude of “direct” heterodyne oscillations;

$$U_{\text{prm}}(t,k)=\{\begin{array}{l}\mathrm{1,}\text{at}k{T}_{\text{P}}<t<k{T}_{\text{P}}+t_{\text{And}}\\ \mathrm{0,}\text{at}k{T}_{\text{P}}+t_{\text{And}}<t<(k+1)T_{\text{P}}\end{array}$$ - unit function of “direct” heterodyne oscillations;

$${\omega }_0=2\text{π}{f}_0$$ - circular frequency of the sounding radio signal;

$$t_{\text{And}}$$ And $$T_{\text{P}}=2\pi {\text{/Ω}}_{\text{P}}$$ - duration and repetition period of radio pulses;

$${\phi }_k$$ - random initial phase $$k$$ th radio pulse;

$$k$$ - integer, serial number of the probing radio pulse.

If there is a detachment in the controlled space of the TP, then partial reflections of EM radiation arise from the “shiny” points on its surface, which are received by antenna A and converted into microwave oscillations of partial radio signals. With a homodyne design of the PPB (see Fig. 2, a ), these radio signals pass through the circulator C along the path second port - third port and arrive at the port of the SM mixer. With the autodyne design of the PPB (see Fig. 2, b ), the received partial radio signals immediately arrive at the port of the microwave generator SHF-G. Expression for radio signal $$u_{\text{negative}}\text{(}t\text{)}$$ , reflected from the set of “shiny” points of the cut, has the form:

$$u_{\text{negative}}\text{(}t\text{)}=A_0{\displaystyle \sum _{k=1}^{\infty }{\displaystyle \sum _{n=1}^N{\Gamma }_n{U}_{\text{negative}}\text{(}{\tau }_n,k,t\text{)}\text{cos}\text{[}{\omega }_0(t-{\tau }_n)+{\phi }_k+{\phi }_n\text{]}}}$$ , (3)

Where

$$A_0$$ - amplitude of the probing radio signal;

$$N$$ - the number of partial radio signals received from “shiny” points on the surface of the cut;

$${\text{Γ}}_n=\sqrt{\frac{P_{\text{probe}}{G}_{\text{A}}^2{\lambda }^2{\sigma }_n}{{(4\pi )}^3{P}_{\min }{R}_n^4}}$$ - dimensionless attenuation coefficient of the amplitude of the emitted signal along the propagation path up to $$n$$ -th “shiny” point and back, brought to antenna port A;

$$U_{\text{negative}}({\tau }_n,k,t)=\{\begin{array}{l}\mathrm{1,}\text{at}k{T}_{\text{P}}+{\tau }_n<t<k{T}_{\text{P}}+t_{\text{And}}+{\tau }_n\\ \mathrm{0,}\text{at}k{T}_{\text{P}}+t_{\text{And}}+{\tau }_n<t<(k+1)T_{\text{P}}+{\tau }_n\end{array}$$ - unit function of the reflected radio signal from $$n$$ -th “shiny” point;

$${\tau }_n=2R_n/c$$ ; - delay time of reflected radiation from $$n$$ -th “shiny” point;

$$R_n$$ - current distance to $$n$$ -th “shiny” point;

$${\phi }_k$$ - random initial phase $$k$$ th radio pulse;

$$k$$ - integer, serial number of the radio pulse;

$${\phi }_n$$ - constant phase shift associated with reflective properties $$n$$ -th “shiny” point;

$$P_{\text{probe}}$$ - average power of the sounding radio signal;

$$P_{\min }$$ - minimum detectable (threshold) radio signal;

$$G_{\text{A}}$$ - antenna gain A;

$$\text{λ}=c/f_0$$ - radiation wavelength;

$${\sigma }_n$$ - effective scattering area $$n$$ -th “shiny” point;

$$c$$ - speed of propagation of electromagnetic radiation.

In diagrams c and d of Figs. Figure 4 shows time diagrams of signals for the case of the presence in the radiation field of antenna A of two “shiny” points on the surface of the cut ( $$N=2$$ ), located respectively at distances $$R_{\text{1}}$$ And $$R_{\text{2}}$$ , when the delay times are $${\tau }_{\text{1}}=2R_{\text{1}}/c$$ And $${\tau }_{\text{2}}=2R_{\text{2}}/c$$ .

With a homodyne design of the PPB (see Fig. 2, a ), as a result of the nonlinear interaction of direct (2) and reflected (3) oscillations in the SM mixer, partial radio signals are converted to the low (Doppler) frequency region. At the same time, at the output of the CM mixer, the converted signals $$u_{\text{etc}}(t)$$ are formed in the form of video pulses (see diagrams and Fig. 4).

In the autodyne design of the PPB (see Fig. 2, b ), the received partial radio signals (3), mixing with natural oscillations (1), cause an autodyne effect in the microwave generator SHF-G [6], consisting of autodyne changes in the frequency and amplitude of oscillations , as well as the average value of current and/or voltage in the power supply circuit of the microwave generator SHF-G. These changes are almost equivalent to the conversion of the radio signal in the CM mixer $$u_{\text{etc}}(t)$$ , are also formed in the form of video pulses (see diagrams and Fig. 4). The expression obtained for video pulses from $$N$$ reflection elements has the form:

$$u_{\text{etc}}(t)=A_{\text{With}}{\displaystyle \sum _{k=1}^{\infty }{\displaystyle \sum _{n=1}^N{\Gamma }_n{U}_{\text{etc}}\text{(}{\tau }_n,k,t\text{)}\text{cos}[(4\pi /\lambda )R_n(t)+{\phi }_n]}}+u_{\text{w}\text{.}k}(t)+A_{\text{them}}{U}_k(t)$$ , (4)

Where

$$A_{\text{With}}=k_{\text{etc}}{A}_0{A}_{\text{prm}}$$ - amplitude factor of the output signal of the SM mixer or the autodyne microwave generator SHF-G;

$$A_0$$ - amplitude of the probing radio signal;

$$A_{\text{prm}}$$ - amplitude of “direct” heterodyne oscillations;

$$U_{\text{etc}}\text{(}{\tau }_n,k,t\text{)}=U_{\text{prm}}\text{(}k,t\text{)}\times U_{\text{negative}}({\tau }_n,k,t)=\{\begin{array}{l}\mathrm{1,}\text{at}k{T}_{\text{P}}+\tau <t<k{T}_{\text{P}}+t_{\text{And}}\\ \mathrm{0,}\text{at}k{T}_{\text{P}}<t<k{T}_{\text{P}}+t_{\text{And}}-\tau \end{array}$$ - unit function of the converted signal at the output of the SM mixer or the autodyne microwave generator SHF-G;

$$R_n(t)$$ - current distance to $$n$$ -th “shiny” point located on the surface of the cut relative to the RLD;

$$u_{\text{w}\text{.}k}$$ - component of the mixer’s own noise and the noise of the microwave generator, converted to the output of the SM mixer or the autodyne microwave generator SHF-G;

$$A_{\text{them}}=k_{\text{d}}{A}_0{A}_{\text{prm}}$$ - amplitude of the pulse received at the output of the SM mixer or the autodyne microwave generator SHF-G as a result of detecting a radio signal;

$$k_{\text{etc}}$$ And $$k_{\text{d}}$$ - conversion and amplitude detection coefficients of the SM mixer or the SHF-G autodyne microwave generator by voltage, respectively;

$$\lambda =2\pi c/{\omega }_0$$ - wavelength of microwave radiation.

Note that the initial phase $${\phi }_k$$ in (4) is absent, since under the condition $$\tau <t_{\text{And}}$$ direct and reflected radiation within $$k$$ th radio pulse are coherent and the phase $${\phi }_k$$ is subtracted in the converted signal.

The first term in (4), representing the result of conversion of the reflected radio signal in the SM mixer or the SHF-G autodyne microwave generator, contains information about the range to the “shiny” points on the surface of the cut and the speed of their movement relative to the RLD. At the same time, for the actual speeds of movement of the cut, the condition is valid that during the time $$t_{\text{And}}$$ Under the influence of the probing radio pulse, the distance between antenna A and the cut remains practically unchanged. Then, according to (4), the received video pulses at the outputs of the SM mixer or the autodyne microwave generator SHF-G remain practically constant during the action of these radio pulses. Therefore, these video pulses look in the form of step functions of time, while the “height” of the steps is proportional to the level of the reflected signal, and their sign (up or down) depends on the current phase difference of the emitted and reflected oscillations. The appearance of video pulse data on the oscilloscope screen is shown in Fig. 17 of article [39]. The timing diagrams of these signals for the case of two “shiny” points without taking into account noise are shown under letter d in Fig. 4. With the relative movement of the cut in the TP control zone, instantaneous changes in the height of the steps occur with the Doppler frequency [6].

Thus, reflected signals from “shiny” points, which are closer than subsequent elements on the surface of the cut, cause changes in the height of video pulses at the output of the mixer or autodyne from the time of arrival of the reflected radiation $${\tau }_{\text{1}}$$ until the end of this video pulse $$t_{\text{And}}$$ . At the same time, the formation of changes caused by the action of reflected radiation from the following “brilliant” points occurs against the background of previous changes from the moment of time $${\tau }_{\text{2}}$$ also until the end of the radio pulse $$t_{\text{And}}$$ . It follows that in the proposed RLD there is an additive addition of video pulses received from individual “shiny” points on the surface of the cut.

Second term $$u_{\text{w}\text{.}k}(t)$$ in (4) displays the result of converting the intrinsic noise of the SM mixer and the noise of the microwave generator SHF-G. The presence of these noises is expressed in noise modulation of the height of video pulses at the output of the SM mixer or the SHF-G autodyne microwave generator. It should be noted that this noise component $$u_{\text{w}\text{.}k}(t)$$ is a stationary normal process with zero mean.

The third term in (4) is due to the detection of the direct signal of the microwave generator SHF-G, acting on the input of the SM mixer, or the natural oscillations of the autodyne microwave generator SHF-G. Therefore, at the output of these devices, video pulses obtained as a result of receiving reflected radio signals are located on some “parasitic” pedestal (see diagram d in Figure 4).

From the output of the transceiver unit PPB (see Fig. 1), video pulses are then supplied to the signal input of the ADC, where the signal (4) is first sampled in time. Then, during the action of the clock pulses $$u_{\text{you}}(t)$$ (see diagram e in Fig. 4) in the ADC, instantaneous signal values (4) are sampled and stored in the form of pulses, the amplitude of which is equal to the instantaneous signal values (see diagram g in Fig. 4). The levels of these pulses are further converted into digital values in the ADC, which, in the form of a parallel code, enter the RAM of the DBSP as an array of data received for the received signal from $$k$$ -th video pulse:

$${\overrightarrow{u}}_{\text{etc}\text{.}kz}=[u_{\text{etc}\text{.}k0}{u}_{\text{etc}\text{.}k1}{u}_{\text{etc}\text{.}k2}\mathrm{...}{u}_{\text{etc}\text{.}kZ}]$$ , (5)

Where

$$u_{\text{etc}\text{.}kz}=u_{\text{etc}\text{.}k}{(t)}_{t=z{T}_{\text{you}}}$$ - digital samples of instantaneous values of the received signal from $$k$$ -th video pulse received for $$z$$ th clock pulse (here $$z=\mathrm{0,}\mathrm{1,}\mathrm{2,}\mathrm{...}Z$$ ).

Sequences digitized for each $$k$$ -th video pulse instantaneous values $$u_{\text{etc}\text{.}kz}$$ , received at points in time $$t=z{T}_{\text{you}}$$ , then enter the computing core of the DSP, where, in accordance with the program embedded in the ROM of the DSP, a subtraction operation is performed, in which from the current reading values, starting from $$z=1$$ , the values of previous samples are subtracted:

$$x_{kz}=(u_{\text{etc}\text{.}kz}-u_{\text{etc}\text{.}k(z-1)})$$ , (6)

The data array obtained as a result of subtraction (6) $$x_{kz}$$ enters the DSP RAM:

$${\overrightarrow{x}}_{kz}=[x_{k1}{x}_{k2}{x}_{k3}\mathrm{...}{x}_{kZ}]$$ , (7)

Where

$$x_{kz}$$ - digital samples of instantaneous values of differences from (6) for $$k$$ -th video pulse received for $$z$$ th clock pulse (here $$z=\mathrm{1,}\mathrm{2,}\mathrm{...}Z$$ ).

As a result of performing a subtraction operation on the resulting values $$x_{kz}$$ they are separated from the parasitic pedestal, and also, if there is a cut with $$N$$ “shiny” dots on its surface, the signals are separated. Signals received from $$n$$ -th “shiny” point are separated from the superposition of signals that are received from the previous “shiny” points. In diagram 3 of Fig. Figure 4, for the case of the presence of a cut with two “shiny” points in the control zone, shows the formation of difference signals that enter various memory cells of the DSP RAM.

Noise component $$u_{\text{w}\text{.}k}(t)$$ at the output of the SM mixer or the autodyne microwave generator SHF-G as a result of sampling and digitization of instantaneous sample values due to the ergodicity of the process on average across implementations $$k$$ and counts $$z$$ retains its rms value $${\sigma }_{\text{w}}$$ noise level. Noise level value $${\sigma }_{\text{w}}$$ can be calculated or measured experimentally and taken into account in the program of work of the DSP BOS.

Based on the obtained difference values $$x_{kz}$$ , using a sufficient number of samples in the process of changing the number $$k$$ video pulse, the DSP computing core performs a fast Fourier transform (FFT) operation using the “Radix2” algorithm using the Hanning window function. After applying the FFT, the resulting signal spectrum contains only components associated with reflections from “shiny” points on the surface of the moving cut. Moreover, for each count $$z$$ , beginning with $$z=1$$ , amplitude values are calculated $$A_z$$ and frequencies $${\Omega }_z$$ signals, which are stored in the RAM memory of the DSP in the form of data arrays:

$${\overrightarrow{A}}_z=[A_1{A}_2{A}_3\mathrm{...}{A}_Z]$$ And $${\overrightarrow{\Omega }}_z=[{\Omega }_1{\Omega }_2{\Omega }_3\mathrm{...}{\Omega }_Z]$$ . (8)

Next, for the obtained amplitudes $$A_z$$ (8) the DSP computing core calculates the signal-to-noise ratio values $$q_z=A_z/{\sigma }_{\text{w}}$$ , Where $${\sigma }_{\text{w}}$$ - root-mean-square value of the self-noise level at the output of the PPB, entered into the RAM memory as a constant. Received signal-to-noise ratio values $$q_z$$ are entered into the RAM memory of the DSP in the form of a data array:

$${\overrightarrow{q}}_z=[q_1{q}_2{q}_3\mathrm{...}{q}_Z]$$ . (9)

After this, the computing core of the DSP from a set of values $$q_z$$ signal-to-noise ratios, those values of amplitude serial numbers are selected $$A_z$$ and frequencies $${\Omega }_z$$ , at which the signal-to-noise level exceeds the threshold value: $$q_z\ge q_{\text{since then}}$$ , Where $$q_{\text{since then}}$$ - threshold signal-to-noise ratio, stored in RAM memory as a constant. This operation in radar is called solving the problem of detecting a signal from a location object. The amplitude values found in this case $$A_z$$ and frequencies $${\Omega }_z$$ based on number $$z$$ clock pulse, the computing core of the DSP brings into correspondence the serial numbers of the signals received from $$n$$ -th “shiny” point on the surface of the cut. Corresponding amplitude values $$A_n$$ and frequencies $${\Omega }_n$$ signals of a detected cut in the controlled section of the TP are entered into the RAM memory of the DSP in the form of data arrays:

$${\overrightarrow{A}}_n=[A_1{A}_2{A}_3\mathrm{...}{A}_N]$$ And $${\overrightarrow{\Omega }}_n=[{\Omega }_1{\Omega }_2{\Omega }_3\mathrm{...}{\Omega }_N]$$ , (10)

Where

$$A_n$$ And $${\Omega }_n$$ - amplitude and frequency of the Doppler signal spectrum obtained from $$n$$ -th “shiny” point detected on the surface of the moving cut (here $$n=\mathrm{1,}\mathrm{2,}\mathrm{...}N$$ ).

Next, the computational core of the DSP BOS calculates the current distance $$R_n$$ before $$n$$ -th “shiny” point on the surface of the detected cut according to the formula:

$$R_n=\frac{z_n{t}_{\text{And}}c}{2Z}$$ , (eleven)

Where

$$z_n$$ - number of samples $$z$$ , corresponding $$n$$ -th “brilliant” point, for which the amplitude $$A_n$$ signal exceeding the threshold level is detected;

$$t_{\text{And}}$$ - duration of the radio pulse;

$$c$$ - speed of propagation of microwave radiation;

$$Z$$ - total number of samples per time $$t_{\text{And}}$$ .

Obtained distance values $$R_n$$ are entered into the RAM memory of the DSP in the form of a data array:

$${\overrightarrow{R}}_n=[R_1{R}_2{R}_3\mathrm{...}{R}_N]$$ . (12)

Using array data (12), the computational core of the DSP finds the current positions of the “shiny” points on the cut relative to the reference line - the perpendicular from the place where the RLD is installed on the rail track:

$$l_{\text{father}\text{.}n}=\sqrt{R_n^2-{[l_{\text{RLD}}-(b/2)]}^2}$$ , (13)

Where

$$l_{\text{father}\text{.}n}$$ - current distance $$n$$ -th “shiny” point from the reference line (perpendicular from the location of the RLD installation on the rail track);

$$R_n$$ - current distance from the RLD installation location to $$n$$ -th “shiny” point on the surface of the cut;

$$l_{\text{RLD}}$$ - the distance from the location of the RLD installation in the intertrack to the axis of the railway track, which is stored in the RAM memory of the DSP as a constant;

$$b$$ - width of the car, stored in the RAM memory of the centralized transmission center (see Fig. 3.28, [23]).

After this, the computational core of the DSP from the set of current distance values $$l_{\text{father}\text{.}n}$$ “shiny” points, those values for which the two-sided inequality holds are selected:

$$l_{\min }\le l_{\text{father}\text{.}n}\le l_{\max }$$ , (14)

Where

$$l_{\text{father}\text{.}n}$$ - current distance $$n$$ -th “shiny” point from the reference line (perpendicular from the place where the RLD is installed on the rail track);

$$l_{\text{min}}$$ , $$l_{\text{max}}$$ - the near and far boundaries of the controlled section of the TP, which are defined as the distance from the reference line (see Fig. 3.28, [23]).

Signals that do not satisfy condition (14), which also includes interference from vehicles moving on adjacent tracks, are ignored in further processing and excluded from the DSP RAM. The fulfillment of inequality (14) means the presence of a cut-off in the controlled section of the TP. For “shiny” points satisfying this inequality, from the data array (10) for frequencies $${\Omega }_n$$ and (12) for distances $$R_n$$ The computing core of the DSP compiles and stores new data arrays in the RAM memory of the DSP:

$${\overrightarrow{\Omega }}_m=[{\Omega }_1{\Omega }_2{\Omega }_3\mathrm{...}{\Omega }_M]$$ , $${\overrightarrow{R}}_m=[R_1{R}_2{R}_3\mathrm{...}{R}_M]$$ , (15)

Where

$${\Omega }_m$$ - frequency of the Doppler signal spectrum obtained from $$m$$ -th “brilliant” point satisfying inequality (14) (here $$m=\mathrm{1,}\mathrm{2,}\mathrm{...}M$$ );

$$R_m$$ - current distance from the radar to $$m$$ -th “shiny” point, satisfying inequality (14).

Simultaneously with the detection of the presence of a cut in the controlled section of the TP, the identification number of the controlled cut, received from the computer when the RLD is operating in the GAC system, or manually entered by the operator when the RLD is operating autonomously, is entered into the RAM memory of the DSP. In addition, the data of the current time of the GAC unified time system or from the BOS timer built into the DSP is automatically entered into the RAM memory of the DSP.

Based on the obtained Doppler frequency values $${\Omega }_m$$ and distances (15) taking into account the values of the constants $$l_{\text{RLD}}$$ And $$b$$ stored in the RAM memory of the DSP, the ground speed is calculated by the computing core of the DSP $$V_m$$ movement $$m$$ -th “shiny” point located on the surface of the controlled cut, according to the following formula:

$$V_m=\frac{c{\Omega }_m}{4\pi f_0}\times \frac{R_m}{\sqrt{R_m^2-{[l_{\text{RLD}}-(b/2)]}^2}}$$ , (16)

Where

$$V_m$$ - ground speed $$m$$ -th “shiny” point located on the surface of the controlled cut at the current moment in time;

$$c$$ - speed of propagation of radio waves;

$${\Omega }_m$$ - value of circular Doppler frequency $$m$$ -th “shiny” point located on the surface of the controlled cut;

$$f_0$$ - cyclic frequency of RLD radiation;

$$R_m$$ - current distance from the RLD installation location to $$m$$ -th “shiny” point on the surface of the controlled release;

$$l_{\text{RLD}}$$ - distance from the location of the RLD installation between tracks to the axis of the railway track;

$$b$$ - width of the car (see Fig. 3.28, [23]).

Based on the obtained ground speed values $$V_m$$ movement of “shiny” points located on the surface of the controlled cut, the computational core of the DSP BOS calculates the speed value $$V_{\text{father}}$$ movement of the cut at the current time as the average of the implementation $$M$$ measurements:

$$V_{\text{father}}=\frac{V_1+V_2+V_3+\mathrm{...}+V_M}{M}=\frac{1}{M}{\displaystyle \sum _{m=1}^M{V}_m}$$ . (17)

Speed calculation results $$V_{\text{father}}$$ release sequentially when changing the video pulse number $$k$$ smoothed, for example, using the “moving average” operation or the Kalman filtering algorithm and entered into the DSP RAM memory in the form of an array of speed data $${\widehat{V}}_{\text{father}}(t_k)$$ controlled release, reduced to $$k$$ -th video pulse. Then, using the smoothed values of the velocity data array $${\widehat{V}}_{\text{father}}(t_k)$$ , when changing the video pulse number $$k$$ The computational core of the DSP performs numerical integration according to the expression:

$$L_{\text{father}}={\displaystyle \underset{t_{\text{input}}}{\overset{t_{\text{out}}}{\int }}{\widehat{V}}_{\text{father}}(t_k)d{t}_k}-L_{\text{ku}}$$ , (18)

Where

$$L_{\text{father}}$$ - cut length;

$${\widehat{V}}_{\text{father}}(t_k)$$ - smoothed value of the ground speed of the controlled trailer at the current time $$t_k$$ , tied to $$k$$ -th video pulse;

$$t_{\text{input}}$$ , $$t_{\text{out}}$$ - respectively, the moments of entry and exit of the controlled cut to the controlled section of the TP;

$$L_{\text{ku}}=l_{\text{max}}-l_{\text{min}}$$ - length of the controlled section of the transformer substation;

$$l_{\text{min}}$$ , $$l_{\text{max}}$$ - near and far boundaries of the controlled section of the TP.

Numerical integration (18) can be performed using various methods. For example, using the method of rectangles, trapezoids, parabolas (Simpson's method) or others. The resulting length calculation result $$L_{\text{father}}$$ the controlled cut according to (18) is entered into the RAM memory of the DSP and transferred to the GAC computer.

Next, using the smoothed values of the speed data array $${\widehat{V}}_{\text{father}}(t_k)$$ , when changing the video pulse number $$k$$ The computational core of the DSP performs numerical differentiation, for example, using the finite difference method, according to the expression:

$$a_{\text{father}}(t_k)=\frac{d}{d{t}_k}{\widehat{V}}_{\text{father}}(t_k)=\frac{{\widehat{V}}_{\text{father}}(t_k+\Delta t_{\text{change}})-{\widehat{V}}_{\text{father}}(t_k-\Delta t_{\text{change}})}{2\Delta t_{\text{change}}}$$ , (19)

Where

$$a_{\text{father}}(t_k)$$ - acceleration of the movement of the cutter in the controlled section of the TP at the current time $$t_k$$ , tied to $$k$$ -th video pulse;

$${\widehat{V}}_{\text{father}}(t_k)$$ - smoothed value of the ground speed of the controlled trailer at the current time $$t_k$$ , tied to $$k$$ -th video pulse;

$$\Delta t_{\text{change}}$$ - time interval for measuring acceleration.

Minimum value of the measuring time interval $$\Delta t_{\text{change}}$$ , corresponding to the repetition period $$T_{\text{P}}$$ probing radio pulses does not provide satisfactory accuracy in determining the acceleration of the release. Its optimal value, at which the error in determining acceleration is minimal, depends on the value of the measured uncoupling speed [40]. Therefore, the optimal values of the measuring time interval are stored in the RAM memory of the DSP in the form of a table for individual speed ranges $$\Delta t_{\text{change}}$$ .

Results of calculation of current speed values $${\widehat{V}}_{\text{father}}(t_k)$$ and acceleration $$a_{\text{father}}(t_k)$$ controlled cut, calculated according to (18) and (19) for each smoothed implementation, enter the RAM and ROM of the DSP, and are also transmitted through the bus transceiver PP-1 of the DSP, the SCD bus, the RS-485 module, which provides galvanic isolation of the RLD and two-wire communication line to the GAC computer. The GAC system manages the dismantling of railway trains, displays information about the movement of cuts and their location on the hump, as well as documenting the dismantling process.

In addition, the current speed values $${\widehat{V}}_{\text{father}}(t_k)$$ cut from the third port of the DSP through the frequency output bus of the ShChV are supplied to the input of the DAC, the output of which generates a voltage directly proportional to the speed of the controlled cut. This voltage is then supplied to the input of the voltage-frequency converter PN-CH, at the output of which a pulse signal of the meander type is generated, having a frequency directly proportional to the cut speed. After this signal passes through the INV inverter and amplifiers U-1 and U-2, two paraphase frequency signals necessary for the “manual” control system of the moderator operation are formed at outputs F1 and F2.

It should be noted that the possibility of automatic self-monitoring of the performance of the radar, provided in the prototype, is not the subject of the present invention. This function can also be implemented in the proposed device.

Thus, the proposed technical solutions allowed the mixture of Doppler signals at the output of the PPB, obtained from “shiny” points on the surface of a moving cut, to be divided into separate independent partial signals from point reflectors. This eliminates the main drawback of the prototype and known analogues - fading (disappearance) of signals, which increases the accuracy of determining the parameters of the movement of cuts, and also eliminates the need to use complex signal loss restorers and processing algorithms in the RLD.

Processing of individual partial signals makes it possible to separately determine the relative location of “shiny” points on the surface of the cut and their ground speed of movement. Since the “shiny” points are located on the surface of the same cut, applying the averaging operation to the obtained values further increases the accuracy of determining the parameters of the cut’s movement.

In addition, the proposed RLD, thanks to the selection of cuts by range, has new properties. This is the ability to separately measure the speed of passing trailers on different retarders and increased resistance to interference from vehicles moving on adjacent tracks. These properties were obtained by selecting signals for processing those “shiny” points that are located within only a controlled section of the TP path.

The limited control area using RLD makes it possible to solve the problem of autonomous detection of cuts at TP without the use of track circuits or axle counting sensors. In addition to measuring the speed of movement of cuts at the TP, the proposed device provides the ability to determine the length of cuts and their acceleration. Taking into account the known data on the length of cars, it is possible to control the number of cars in the cuts. The specified expansion of the functionality of the claimed device provides its additional advantages in relation to the prototype and analogues.

Literature

1. Grigorin-Ryabov V.V., Verigo V.M., Shelukhin O.I., Shelukhin V.I. Radio engineering railway devices. - M.: Transport, 1986. - 161 p.

2. Noskov V.Ya., Smolsky S.M., Ignatkov K.A. and others. Modern hybrid-integrated autodyne generators of the microwave and millimeter ranges and their application. Part 11. Basics of autodyne implementation // Advances in modern radio electronics. 2019. No. 2. P. 5-33.

3. Noskov V.Ya., Ignatkov K.A. Application of a stabilized two-diode autodyne in a radar sensor for hump humps // 22nd International Crimean Conference “Microwave Engineering and Telecommunication Technologies” (CriMiKo’2012). Sevastopol, 2012. pp. 893-896.

4. Patent US3710385, publ. 01/09/1973. Vehicle initial speed and stopping distance indicator / Howard et al.

5. Ermak G.P., Varavin A.V., Popov I.V., Vasilyev O.S., Usov L.S. Radar sensor for monitoring the visibility and fluidity of dry warehouses in the territories of sorting warehouses // Science and Innovation. 2009. T. 5. No. 5. P. 9-16 (Ukr). (Ermak G.P., Varavin A.V., Popov I.V., Vasilyev A.S., Usov L.S. Radar sensor for monitoring the presence and speed of rolling stock in hump areas // Science and Innovation. 2009. T. 5. No. 5. P. 9-16) Internet access at: http://dspace.nbuv.gov.ua/xmlui/handle/123456789/ 27893.

6. Noskov V.Ya., Smolsky S.M., Ignatkov K.A., Mishin D.Ya., Chupakhin A.P. Modern hybrid-integrated autodyne generators of the microwave and millimeter ranges and their application. Part 10. Fundamentals of analysis and calculation of autodyne parameters taking into account noise // Advances in modern radio electronics. 2018. No. 3. P. 18-52.

7. Umansky G.M. Electronic speed meter ES-TsNII // Automation, telemechanics and communications. 1962. No. 5. P. 13-19.

8. Auto. date SU305414, publ. 06/04/1971. IPC G01P3/42. Radar speed meter for wagons / Yu.V. Vavanov and others.

9. Grigorin-Ryabov V.V. (ed). Radar devices (theory and principles of construction). M.: Soviet radio, 1970. - 680 p.

10. Shelukhin V.I. Sensors for measuring and monitoring railway transport devices. - M.: Transport, 1990. - 119 p.

11. Efimov L.L., Kronov V.I., Prudovsky I.I. Doppler velocity meter DIS-74. Automation, telemechanics and communications. 1980. No. 4. pp. 21-22.

12. Kogan I.M. Short-range radar (theoretical foundations). - M.: Sov. radio, 1973. - 272 p.

13. German patent DE2126663, publ. 12/07/1971. IPC B61K7/12, G01S9/44, G01S13/583(EP). Verfahren zum Ermitteln der Geschwindigkeit von Fahrzeugen / Grafinger

14. Auto. date SU410314, publ. 01/05/1974, bulletin. No. 1. IPC G01P3/42. Measuring signal converter / B.T. Anashkin, L.L. Efimov.

15. Vavanov Yu.V., Verigo A.M., Dagaeva N.Kh., Peregonov S.A., Kernov Yu.P. Radar meter for the speed of movement of cuts for hump humps // Automation, telemechanics and communications. 1977. No. 9. P. 3-5.

16. Radar speed meter for railway cars at a hump / Yu.N. Kuznetsov, L.F. Mikhailov, V.A. Parilov et al. // Electronic technology. Ser. Microwave electronics. 1982. No. 7. P. 56-58.

17. Verigo A.M., Vavanov Yu.V., Tenn F.A., Podorov N.A. Radar speed meter for wagons // Automation, telemechanics and communications. 1983. No. 3. P. 7-9.

18. Auto. date SU1437284A1, publ. 11/15/1988, bulletin. No. 42. IPC ⁴ B61L17/00. Device for measuring the acceleration of movement of cuts on a hump / V.I. Shelukhin, O.I. Shelukhin.

19. US Patent US4172256, publ. 10/23/1979. IPC G01S13/92 (EP), G01S9/44. Circuit for speed measurement of vehicles according to the Doppler-radar principle / Pacozzi.

20. Shelukhin V.I., Kolesnichenko N.N., Pyzhyanov V.G. Modified radar cut speed meter // Automation, telemechanics and communications. 1988. No. 8. P. 8-10.

21. Patent RU2003543С1, publ. 30.11.1993, bulletin. No. 43-44. IPC ⁵ B61L17/00. Device for measuring the parameters of the movement of cuts on a hump / V.I. Shelukhin and others.

22. Patent RU2229404С2, publ. 05/27/2004, bulletin. No. 15. IPC ⁷ B61L17/00. A method for measuring the speed of movement of cuts on a hump and a device that implements it / P.V. Chernov, V.V. Chekurskov and others.

23. Shelukhin V.I. Automation and mechanization of hump humps: Textbook for technical schools and railway colleges. transport. - M.: Route, 2005. - 240 p.

24. Komarov V.M., Zalichev N.N., Vorotnikov V.N. Collision avoidance systems for ground vehicles / Foreign radio electronics. 1981. No. 4. P. 54-72.

25. Shelukhin V.I., Shelukhin O.I. Multifunctional hump motion sensor // Automation, telemechanics and communications. 1991. No. 4. P. 10-14.

26. Maryukhnenko V.S. Radar systems in railway transport. Prospects for application. - Irkutsk: IrGUPS, 2017. - 146 p.

27. Shelukhin O.V. Short-range radio systems. - M.: Radio and communication, 1989. - 240 p.

28. Handbook on radar / Ed. M.I. Skolnik. In 2 books. Book 1. Moscow: Tekhnosphere, 2014.

29. Kostenko A.A., Khlopov G.I. Coherent short- and ultra-short-range millimeter-wave radar systems. - Kharkov: IPC “Contrast”, 2015. - 352 p.

30. Voskresensky D.I. Antennas and microwave devices. Calculation and design of antenna arrays and their radiating elements. - M.: Soviet radio, 1972. - 162 p.

31. Baranov A.V., Krevsky M.A. Transistor generators of harmonic microwave oscillations. - M.: Hotline - Telecom, 2021. - 276 p.

32. Malyshev V.A. Onboard active microwave devices. - L.: Shipbuilding, 1990, 264 p.

33. Grushvitsky R.I., Mursaev A.Kh., Gloomy E.P. Design of systems on programmable logic chips. - St. Petersburg: BHV-Petersburg, 2002.

34. Guzik V.F., Kalyaev I.A., Levin I.I. Reconfigurable computing systems. Rostov-on-Don: Southern Federal University Publishing House, 2016.

35. “User Guide: MSP430X1XX Family of Microcontrollers.” Per. from English M.: ZAO Kompel, 2004.

36. Lebedev O.N., Marcinkevichyus A.-J.K., Bagdanskis E.-A.K. and others. Memory chips, DAC and ADC: Handbook. - M.: KUBK, 1996. 384 p.

37. Shcherbakov V.I., Grezdov G.I. Electronic circuits using operational amplifiers: Handbook. - K.: Technology, 1983, 213 p.

38. Noskov V.Ya., Bogatyrev E.V., Ignatkov K.A.. The principle of constructing an onboard radar sensor for detecting fast-moving targets. Advances in modern radio electronics. 2019. No. 12. pp. 16-22.

39. Noskov V.Ya., Smolsky S.M. Modern hybrid-integrated autodyne generators of the microwave and millimeter ranges and their applications. Part 6. Research of radio pulse autodynes. Advances in modern radio electronics. 2009. No. 6. P. 3-51.

40. Shelukhin V.I. Measuring the acceleration of railway vehicles // Automation, telemechanics and communications. 1987. No. 1. P. 13-15.

Claims (25)

1. A method for radar determination of the parameters of the movement of cuts on a hump, which consists in the fact that by means of a radar sensor antenna (RSD), installed at the beginning or end of the controlled section of the braking position (TP), probing microwave radiation in the form of a pencil is emitted at the height of the car automatic coupler directional patterns coinciding in direction with the diagonal of the controlled section of the TP, they receive reflected radiation from the “shiny” points located within the antenna radiation pattern on the surface of the moving cut, mix them with part of the energy of the probing radiation, converting the resulting mixture of reflected and probing radiation, characterized in that that the probing microwave radiation is formed in the form of radio pulses, while the radiation reflected from the “shiny” points in the form of radio pulses is received and mixed with part of the radiation energy of the probing radio pulses, converting the overlapping parts of these radio pulses into video pulses, the resulting video pulses are sampled in time and stored in multiple points in time $$t_{kz}$$ samples of the instantaneous values of these video pulses $$u_{\text{etc}.kz}$$ , Where $$k=\mathrm{0,}\mathrm{1,}\mathrm{2,}\mathrm{...}$$ – serial number of the video pulse; $$z=\mathrm{0,}\mathrm{1,}\mathrm{2,...}Z$$ – serial number of the sample inside the video pulse, then from the current samples $$u_{\text{etc}.kz}$$ , beginning with $$z=1$$ , subtract the previous sample values $$u_{\text{etc}.k(z-1)}$$ : $$x_{kz}=(u_{\text{etc}\text{.}kz}-u_{\text{etc}\text{.}k(z-1)})$$ , and according to the obtained difference values $$x_{kz}$$ , using many samples along $$k$$ video pulses are obtained for each sample $$z$$ spectrum of the Doppler signal, determining its amplitude $$A_z$$ and frequency $${\Omega }_z$$ , then for the obtained amplitude samples $$A_z$$ calculate signal-to-noise ratio values $$q_z=A_z/{\sigma }_{\text{w}}$$ , Where $${\sigma }_{\text{w}}$$ – root mean square value of the noise level at the receiver output, then from a set of values $$q_z$$ signal-to-noise ratios select those sequence number values $$z$$ for amplitudes $$A_z$$ and frequencies $${\Omega }_z$$ , for which the signal-to-noise level exceeds the threshold: $$q_z\ge q_{\text{since then}}$$ , Where $$q_{\text{since then}}$$ – threshold signal-to-noise ratio, then for selected amplitudes $$A_z$$ and frequencies $${\Omega }_z$$ in ascending order of number $$z$$ determine the number of readings $$z_n$$ for signals received from $$n$$ th “shiny” point, after which the current distance is calculated $$R_n$$ to each of $$n$$ "shiny" points on the surface of the cut according to the formula $$R_n=\frac{z_n{t}_{\text{And}}c}{2Z}$$ , Where $$z_n$$ – number of samples $$z$$ For $$n$$ th “shiny” point; $$t_{\text{And}}$$ – radio pulse duration; $$c$$ – speed of propagation of microwave radiation; $$Z$$ – total number of samples per time $$t_{\text{And}}$$ , then calculate the current distances of the “shiny” points on the cut relative to the reference line - the perpendicular from the place where the RLD is installed on the rail track - using the following formula: $$l_{\text{father}\text{.}n}=\sqrt{R_n^2-{[l_{\text{RLD}}-(b/2)]}^2}$$ , Where $$l_{\text{father}\text{.}n}$$ – current distance $$n$$ th “shiny” point from the reference line; $$R_n$$ – current distance from the RLD installation location to $$n$$ th “shiny” point; $$l_{\text{RLD}}$$ – distance from the location of the RLD installation between tracks to the axis of the railway track; $$b$$ – the width of the car, then from the set of current distance values $$l_{\text{father}\text{.}n}$$ “shiny” points are chosen those values for which the two-sided inequality is satisfied $$l_{\min }\le l_{\text{father}\text{.}n}\le l_{\max }$$ , Where $$l_{\text{father}\text{.}n}$$ – current distance $$n$$ th “shiny” point from the reference line; $$l_{\text{min}}$$ , $$l_{\text{max}}$$ – the near and far boundaries of the measuring section of the TP, which are determined as the distance from the reference line, then for “shiny” points that satisfy the specified inequality, from the data array for frequencies $${\Omega }_n$$ and distances $$R_n$$ compile and store data arrays: $${\overrightarrow{\Omega }}_m=[{\Omega }_1{\Omega }_2{\Omega }_3\mathrm{...}{\Omega }_M]$$ , $${\overrightarrow{R}}_m=[R_1{R}_2{R}_3\mathrm{...}{R}_M]$$ , Where $${\Omega }_m$$ – frequency of the Doppler signal spectrum obtained from $$m$$ th “shiny” point; $$R_m$$ – current distance from the radar to $$m$$ th “brilliant” point, after that, based on the obtained values of Doppler frequencies $${\Omega }_m$$ and distances $$R_m$$ calculate ground speed $$V_{\text{father}}$$ movement of the release according to the following formula: $$V_{\text{father}}=\frac{c}{4\pi f_{0}M}{\displaystyle \sum _{m=1}^M\frac{{\Omega }_m{R}_m}{\sqrt{R_m^2-{[l_{\text{RLD}}-(b/2)]}^2}}}$$ , Where $$c$$ – speed of propagation of radio waves; $$f_0$$ – cyclic frequency of radiation; $$M$$ – number of “shiny” points on the cut surface for which Doppler frequency values were obtained $${\Omega }_m$$ and current distance $$R_m$$ ; $${\Omega }_m$$ – Doppler frequency value $$m$$ th “shiny” point; $$R_m$$ – current distance from the RLD installation location to $$m$$ th “shiny” point; $$l_{\text{RLD}}$$ – distance from the location of the RLD installation between tracks to the axis of the railway track; $$b$$ – width of the car, then the obtained ground speed values $$V_{\text{father}}$$ movement of the release sequentially when the number of the video pulse changes $$k$$ smooth, then using the smoothed values $${\widehat{V}}_{\text{father}}$$ release speeds, calculate acceleration $$a_{\text{father}}$$ movement of the release according to the formula $$a_{\text{father}}=\frac{{\widehat{V}}_{\text{father}\text{.2}}-{\widehat{V}}_{\text{father}\text{.1}}}{\Delta t_{\text{change}}}$$ , Where $${\widehat{V}}_{\text{father}\text{.1}}$$ And $${\widehat{V}}_{\text{father}\text{.2}}$$ – values of the speed of movement of the cut in the first and second sections of the TP, separated in time by the measuring time interval $$\Delta t_{\text{change}}$$ . 2. The method according to claim 1, characterized in that the microwave radiation of the RLD is formed in the form of periodic radio pulses, duration $$t_{\text{And}}$$ which are selected from the condition $$t_{\text{And}}\ge \frac{2}{c}\sqrt{{[l_{\text{RLD}}-(b/2)]}^2+l_{\max }^2}$$ , Where $$c$$ – speed of propagation of radio waves; $$l_{\text{RLD}}$$ – distance from the location of the RLD installation between tracks to the axis of the railway track; $$b$$ – width of the car; $$l_{\text{max}}$$ – distance to the far border of the controlled section of the TP. 3. Method according to paragraphs. 1 and 2, characterized in that the video pulses of the converted signal are separated based on the delay time of the reflected radiation from individual “shiny” points on the surface of the cut, located within the controlled area of the TP. 4. Method according to paragraphs. 1 – 3, characterized in that the spectrum of the reflected signal is obtained by performing a fast Fourier transform operation on the data array of the converted signal using the “Radix2” algorithm using the Hanning window function. 5. Method according to paragraphs. 1 – 4, characterized in that the ground speed values $$V_{\text{father}}$$ movement of the release sequentially when the number of the video pulse changes $$k$$ smoothed using the “moving average” operation or the Kalman filtering algorithm. 6. A device for determining the parameters of the movement of cuts on a hump, containing a series-connected antenna, a transceiver unit (RPB), an analog-to-digital converter (ADC) and a signal processing unit (SPU), characterized in that a pulse modulator and a programmable unit are additionally introduced into it synchronization and control (PBSU), and the output of the pulse modulator is connected to the first output of the PPB, and the signal input of the ADC is connected to its second output, the first output of the PBSU is connected to the input of the pulse modulator, its second output is connected to the BOS through the ShKP programming command bus, and the third the pin is connected to the clock input of the ADC. 7. The device according to claim 6, characterized in that the PPB is made according to a homodyne circuit and contains a microwave generator, a mixer and a circulator, while the microwave generator is connected with its high-frequency port to the first port of the circulator, and the second - to the antenna, and to The third port of the circulator is connected to the input of the mixer, and the modulation input of the microwave generator and the output of the mixer are connected to the first and second terminals of the PPB, respectively. 8. The device according to claim 6, characterized in that the PPB is made according to an autodyne circuit based on a microwave generator connected to the antenna, containing in its oscillatory system or connected to it through a transmission line an amplitude detector, the modulation input of the microwave generator and the amplitude output The detector is connected to the first and second terminals of the PPB, respectively. 9. Device according to any one of paragraphs. 6 – 8, characterized in that the PBSU contains a reference clock generator connected to the inputs of the programmable frequency multiplier and frequency divider, while the output of the programmable frequency divider is the first output of the PBSU, the second output is a programming command bus, the outputs of the programmable frequency divider and frequency multiplier are connected to the inputs a pulse selector, the output of which is the third output of the PBSU.