CN112947500B

CN112947500B - Underwater vehicle water flow monitoring system

Info

Publication number: CN112947500B
Application number: CN202110185590.XA
Authority: CN
Inventors: 马顺利; 曾超凡; 吴天祥; 许诺
Original assignee: Fudan University
Current assignee: Fudan University
Priority date: 2021-02-10
Filing date: 2021-02-10
Publication date: 2022-05-20
Anticipated expiration: 2041-02-10
Also published as: CN112947500A

Abstract

The invention belongs to the field of integrated circuit technology and electronic information, and particularly relates to a water flow monitoring system of an underwater vehicle. The main part of the monitoring system is a flow meter chip array, and the flow rate of fluid on the surface of a chip is converted into a frequency signal. The device also comprises a neural network operation module behind the chip array, and the recognition of the surrounding environment condition is realized. By using the flow velocity meter chip to sense the temperature and outputting the pulse signals with different frequencies, the high power consumption of the output signals and the high area overhead of the traditional flow velocity meter are avoided. The distributed characteristic can be widely covered on the surfaces of surface underwater vehicles such as ships, submarines and the like, and the comprehensiveness of information collection is improved.

Description

Underwater vehicle water flow monitoring system

Technical Field

The invention belongs to the technical field of underwater vehicles, and particularly relates to a water flow monitoring system of an underwater vehicle.

Background

Since the invention of sonar in the last century, the navigation and ship industries have stepped into modernization in terms of information detection technology. The technology for navigation and distance measurement through electroacoustic conversion and information processing by utilizing the propagation and reflection characteristics of sound waves in water can also be utilized to measure the speed of a water flow environment. According to the comprehensive comparison of the water flow speed of the ship when the ship is static and the water flow speed of the ship in all directions during the operation of the ship, the navigation information such as the attitude of the ship can be determined. Sonar is the most widely and important device in water acoustics.

However, the large sound receiving device affects the space volume of the ship, reducing its concealment and manufacturing costs. Factors influencing the working performance of the sonar are the technical conditions of the sonar, and the influence of external conditions is serious, so that the sound waves are influenced and restricted by uneven distribution of seawater media, sea surfaces and sea bottoms in the process of transmission, refraction, scattering, reflection and interference can be generated, sound ray bending, signal fluctuation and distortion can be generated, the change of a transmission path is caused, and the acting distance and the measurement precision of the sonar are seriously influenced. The self-noise of the carrying platform is mainly related to the speed of the ship, the larger the speed of the ship is, the larger the self-noise is, the closer the sonar working distance is, and otherwise, the farther the sonar working distance is; the larger the target reflectivity is, the farther the distance is found by the other side active sonar; the greater the intensity of the target radiation noise, the farther the opposite party finds by the passive sonar. The motion compensation, imaging process required to be performed subject to the absolute moving speed of modern ships having approached the speed of sound is more complicated.

The electromagnetic flowmeter is a novel flow measuring instrument which is rapidly developed along with the development of electronic technology in 50-60 years of the 20 th century. The electromagnetic flowmeter measures the flow of a conductive fluid according to the electromotive force induced when the conductive fluid passes through an external magnetic field by applying the electromagnetic induction principle. However, the electromagnetic flowmeter has extremely high requirements on the external working environment, the equipment needs strict maintenance, and the space overhead required by the pipeline type electromagnetic flowmeter similar to the airspeed head is only small compared with the traditional sonar component.

The many drawbacks of sonar and electromagnetic flow meters have increased the demand for integrated circuit chips for detecting ship speed and submarine attitude. The identification speed measurement principle of the traditional electromagnetic flowmeter and sonar is separated. The speed measuring device based on the chip array has the following advantages:

1. the ship surface distribution characteristic of the chip can enable information collection in the space category to be more comprehensive;

2. available frequency bandwidth and large information capacity. The electromagnetic signal only needs to be transmitted between a computer and a flow velocity meter chip in the ship and is not limited by the severe attenuation of the electromagnetic signal in water and the low resolution of the sound wave signal beyond a certain distance;

3. the low power consumption and the low use area enable the ship design to consider more macroscopic functions without sacrificing more important functional modules for the speed measuring array module.

Disclosure of Invention

In view of the above, the present invention is directed to a water flow monitoring system for an underwater vehicle, which is small and light, and can monitor the water flow speed, the water flow direction and the surrounding water flow.

The invention provides a water flow monitoring system of an underwater vehicle, which comprises a flow meter chip array and an SNN pulse neural network sensor; the flow velocity meter chip array consists of a plurality of flow velocity meter chip units, and the flow velocity meter chip units are distributed on all positions on the surface of the underwater vehicle in space; the flow meter chip unit specifically comprises a flow meter sensor (namely a temperature sensor) and a power amplifier; wherein:

the flow velocity meter sensor comprises a negative resistor formed by an LC resonance network and an NMOS transistor; specifically, the on-chip passive inductor comprises two on-chip passive inductors L1 and L2, and 16 NMOS transistors M1-M16; one ends of two inductors L1 and L2 are connected with a power supply voltage VDD in common, the other ends of the inductors are connected with the grids of two different NMOS tubes, the sources of the two NMOS tubes are connected, and the drains of the two NMOS tubes are connected with a bias voltage in common, so that a capacitor is formed; thus, the total number of the capacitors formed by the two NMOS tubes is seven, and the capacitors are connected in parallel, so that a capacitor array is formed; the two inductors L1 and L2 and seven capacitors form a coupling network; two NMOS tubes are in cross coupling at the lowest end of the capacitor array, namely the drain electrode of one NOMS is connected with the grid electrode of the other NMOS, and meanwhile, the two source electrodes are grounded in common.

The power amplifier comprises three-stage power amplifiers and four transformer matching networks in total between the input end and the output end of each amplifier and each cascaded amplifier; the matching network consists of a parallel capacitor and four on-chip passive inductors. The amplifier part (at each stage) consists of a group of two NMOS differential pair transistors and two capacitors, wherein the capacitors are formed by cross connection of the drains of the two NMOS transistors to the body end of the other transistor; the transistor grid DC bias of the amplifier is introduced from the tap of the secondary coil of the transformer of the matching network of the previous stage; after the signals are output from the sensor of the flow velocity meter, power amplification is carried out, so that electromagnetic signals can be wirelessly transmitted in space, and the electromagnetic signals can be transmitted to a remote information processing end for further processing and analysis.

The impulse neural network sensor comprises 15 dimensionality inputs, and the frequency values of input signals are subjected to synapse operation and classification to obtain the environmental water flow conditions corresponding to each group of input signals, so that the navigation attitude of the underwater vehicle is judged.

The invention relates to a water flow monitoring system of an underwater vehicle, which comprises the following working procedures:

the temperature of the chip in the environment is directly determined by the flow rate of fluid on the surface of the chip, and the temperature of the chip correspondingly changes along with the change of the water speed, so that the capacitance value in the chip changes, different capacitance values can enable the sensor of the flow meter to output signals with different frequencies, and the influence of the temperature change of the chip area on the capacitance value of the NMOS capacitor array is converted into the frequency of the output signal, so that the water speed information is contained in the signal frequency;

the chip unit of the current meter converts the temperature signal of the chip area into an electric signal with different frequencies for outputting, the temperature of the chip is directly influenced by the water speed on the surface of the chip, and the information of the output signal is contained in the signal frequency; after power amplification, the signal is transmitted to a receiving end for signal processing through an antenna; the receiving end collects and synthesizes water speed information in all directions, so that the information is transmitted to the impulse neural network for operation; the sensor of the impulse neural network receives the impulse signal frequency value vector output by the array, and ionization grouping is carried out after operation, so that the surrounding environment condition of the object is detected, and the navigation attitude of the aircraft is judged.

In the invention, the frequency of the output signal of the current meter sensor can be linearly modulated and changed in different ranges according to the adjustment of the control voltage.

In the invention, the flow velocity meter sensor can finally expand a target detection area from one point to a full-coverage surface in any shape by building a honeycomb network taking a hexagon as a basic topological structure.

Preferably, the flowmeter sensor outputs the temperature of the chip through electric signals with different frequencies by adopting an on-chip passive inductor and a temperature-sensitive NMOS tube capacitor array resonance technology.

Preferably, the impulse neural network is a sensor structure and is composed of two stages of neurons connected in the forward direction, and the first stage is an input stage and is composed of 15 neurons, and the impulse neural network is used for transmitting impulse frequencies of 15 flow velocity sensors from different positions. The second stage is an output stage consisting of 4 neurons, which correspond to the four poses of the aircraft respectively: head up, head down, left tilt, and right tilt. The operation is divided into four groups of neuron synapse weights with different settings for carrying out convolution and operation with input signals, then appropriate activation function processing is carried out, and whether one of four output neurons is activated or not is determined according to four groups of operation results, so that the attitude of the underwater vehicle represented by the group of input signals is determined. Wherein the synaptic weights are obtained after repeated training.

According to the flow meter without the flow meter of the embodiment, the temperature sensing chip array is used for receiving the temperature information to obtain the flow rate information, so that the high area overhead and low detection accuracy of the traditional flow meter are avoided; meanwhile, the rear end of the flow meter is provided with an off-chip pulse neural network system, so that the processing capacity of receiving information is greatly improved.

Drawings

Fig. 1 is a block diagram of information receiving, converting and processing flow of a flow rate monitoring system.

FIG. 2 is a schematic diagram of submarine surface current meter chip distribution.

FIG. 3 is a schematic diagram of the flow meter chip.

Fig. 4 is an electrical schematic diagram of a flow meter sensor (temperature sensor).

Fig. 5 is an electrical schematic diagram of a power amplifier.

Fig. 6 is a schematic diagram of an output frequency and control voltage characteristic and a cellular network.

FIG. 7 is a schematic view of an odometer chip.

FIG. 8 is a diagram of a spiking neural network.

FIG. 9 is a schematic diagram of the flowmeter array principle.

Detailed Description

The underwater vehicle water flow detection system is further described with reference to the accompanying drawings. Like elements in the various figures are denoted by like reference numerals. For purposes of clarity, the various features in the drawings are not necessarily drawn to scale. Moreover, some well-known elements may not be shown in the figures.

Numerous specific details of the invention are set forth in the following description in order to provide a more thorough understanding of the invention. However, as will be understood by those skilled in the art, the present invention may be practiced without these specific details.

Fig. 1 shows a flow chart of information receiving, converting and processing of a flow rate monitoring system.

As shown in FIG. 1, the chip of the flow rate meter converts the temperature signal of the chip area into an electric signal with different frequencies for output, and the temperature of the chip is directly influenced by the flow rate of the fluid on the surface of the chip, and the information of the output signal is contained in the signal frequency. After power amplification, the signal can be transmitted to a receiving end for signal processing through an antenna. The impulse neural network perceptron carries out ionization grouping after operation by receiving the impulse signal frequency value vector output by the array. Thereby detecting the environmental conditions around the object.

FIG. 2 shows a schematic distribution diagram of submarine surface anemometer chips.

As shown in fig. 2, the flow rate meter chips can be distributed on each part of the submarine surface randomly or regularly to obtain the flow rate information in each spatial direction.

Fig. 3 shows a schematic view of the flow meter chip.

As shown in fig. 3, the odometer chip includes two parts, an odometer sensor and a power amplifier. The former converts the flow velocity vector information into a frequency signal, and the latter amplifies the power of the frequency signal.

Fig. 4 shows an electrical schematic of the odometer sensor (temperature sensor).

As shown in fig. 4, the odometer sensor includes a negative resistance formed by an LC resonant network and an NMOS transistor. The influence of the temperature change of the chip area on the capacitance value of the NMOS capacitor array is converted into the frequency of an output signal, and all information is contained in the signal frequency because the temperature of the chip in the environment is directly determined by the flow rate of fluid on the surface of the chip, so that low-power-consumption emission and processing are facilitated. The on-chip passive inductor comprises two on-chip passive inductors L1 and L2, and 16 NMOS transistors M1-M16. One end of each of the two inductors is connected with a power voltage VDD in common, the other end of each of the two inductors is connected with the grids of two different NMOS tubes, the sources of the two NMOS tubes are connected, the drains of the two NMOS tubes are connected with a bias voltage in common, so that a capacitor is formed, the capacitors formed by the two NMOS tubes are seven in total and are connected in parallel, so that a capacitor array is formed, the two cross-coupled NMOS tubes are arranged at the lowest end, namely the drain of the NOMS is connected with the grid of the other NMOS tube, and meanwhile, the two sources of the two inductors are grounded in common.

Fig. 5 shows an electrical schematic of a power amplifier.

As shown in fig. 5, the power amplifier performs power amplification on the signal output by the sensor of the flow meter so as to perform wireless propagation in space as an electromagnetic signal, so as to transmit the electromagnetic signal to the telematics terminal for further processing and analysis. The power amplifier comprises three-stage power amplifiers and four transformer matching networks in total between input and output ends and the cascaded amplifiers. The matching network consists of parallel capacitors C1, C4, C7 and C10 and four on-chip passive inductors K1, K2, K3 and K4. The amplifier section consists of a pair of differentially connected transistors M1-M6 and capacitors C2 and C3, C5 and C6, C8 and C9 with their respective drains cross-connected to the body terminals of the pair.

Fig. 6 shows an output frequency and control voltage characteristic and a cellular network diagram.

As shown in fig. 6, the frequency of the output signal may vary linearly with the input voltage over several different frequency bands. The coverage range of each flow meter chip unit is regarded as a honeycomb regular hexagon, namely, the monitoring area can be popularized to the whole surface of the target to be detected in a mode shown in the figure, so that the effect of complete coverage is achieved.

FIG. 7 shows a schematic of an odometer chip.

As shown in fig. 7, the flow velocity information on the chip surface is processed by the chip to obtain an amplified frequency signal, and the amplified frequency signal is transmitted through the antenna.

Fig. 8 shows a diagram of a spiking neural network.

As shown in fig. 8, the impulse neural network is a sensor structure, and is formed by forward connection of two stages of neurons, and the first stage is an input stage composed of 15 neurons, and is used for transmitting impulse frequencies of 15 flow rate sensors from different positions. The second stage is an output stage consisting of 4 neurons, which correspond to the four poses of the aircraft respectively: head up, head down, left tilt, and right tilt. The operation is divided into four groups of neuron synapse weights with different settings for performing convolution and operation with input signals, and then the appropriate activation function processing is performed, and whether one of four output neurons is activated or not is determined according to four groups of operation results. Thereby determining the attitude of the underwater vehicle represented by the set of input signals.

Fig. 9 shows a schematic view of an array of flow meters.

Each set of data for the impulse neural network shown in fig. 9 is from an array of spatially distributed flow meter chip units. Which is transmitted via radio electromagnetic waves. And calculating by the neural network according to the input data to obtain water flow information data.

Claims

1. The underwater vehicle water flow monitoring system is characterized by comprising a flow meter chip array and an SNN pulse neural network sensor; the flow meter chip array consists of a plurality of flow meter chip units, and each flow meter chip unit specifically comprises a flow meter sensor and a power amplifier; each flow velocity meter chip unit in the flow velocity meter chip array is spatially distributed on each position of the surface of the underwater vehicle in a honeycomb shape; wherein:

the flow velocity meter sensor comprises two on-chip passive inductors L1 and L2, and 16 NMOS tubes M1-M16; one end of each of the two inductors is connected with a power supply voltage VDD in common, the other end of each of the two inductors is connected with the grids of two different NMOS tubes, the sources of the two NMOS tubes are connected, and the drains of the two NMOS tubes are connected with a bias voltage in common, so that a capacitor is formed; thus, the total number of the capacitors formed by the two NMOS tubes is seven, and the capacitors are connected in parallel, so that a capacitor array is formed; the two inductors L1 and L2 and seven capacitors form a coupling network; the last two NMOS tubes are in cross coupling, namely the drain electrode of one NOMS is connected with the grid electrode of the other NMOS, and the two source electrodes are grounded in common;

the power amplifier comprises three-stage power amplifiers and four transformer matching networks in total between the input end and the output end of each amplifier and each cascaded amplifier; the matching network consists of a parallel capacitor and four on-chip passive inductors; the amplifier part consists of a group of two NMOS differential pair transistors and two capacitors, wherein the capacitors are formed by cross connection of respective drains of two NMOS transistors to a body terminal of the other NMOS transistor; the transistor grid voltage direct current bias of the amplifier is introduced from a tap of a secondary coil of a transformer of the matching network of the previous stage; after the signal is output from the sensor of the current meter, power amplification is carried out so as to facilitate wireless transmission of the electromagnetic signal in space, and the electromagnetic signal is transmitted to a remote information processing end for further processing and analysis;

the impulse neural network sensor comprises 15 input neurons for inputting impulse signals, and the frequency values of the input signals are subjected to synapse operation and are classified to obtain the environmental water flow conditions corresponding to each group of input signals, so that the navigation attitude of the underwater vehicle is judged.

2. The underwater vehicle current monitoring system of claim 1, wherein the workflow is as follows:

the temperature of the chip in the environment is directly determined by the flow velocity of fluid on the surface of the chip, and the temperature of the chip is correspondingly changed along with the change of the water velocity, so that the capacitance value in the chip is changed, the different capacitance values enable the sensor of the current meter to output signals with different frequencies, and the influence of the temperature change of the chip area on the capacitance value of the NMOS capacitor array is converted into the frequency of the output signal, so that the water velocity information is contained in the signal frequency;

3. The underwater vehicle current monitoring system of claim 1, wherein the odometer sensor, in response to adjustment of the control voltage, outputs a signal having a frequency that varies linearly with the control voltage over a range of different frequencies.

4. The underwater vehicle current monitoring system of claim 1, wherein the anemometer sensors can finally expand a target detection area from one point to a full coverage surface of any shape by building a cellular network with a hexagonal basic topology.

5. The underwater vehicle water flow monitoring system of claim 1, wherein the NMOS transistors are all field effect transistors.

6. The underwater vehicle current monitoring system of claim 1, wherein said impulse neural network is a perceptron structure comprised of two stages of neurons connected in a forward direction, the first stage being an input stage comprised of 15 neurons for delivering the impulse frequency of 15 flow rate sensors from different locations; the second stage is an output stage consisting of 4 neurons and corresponding to four postures of the aircraft respectively: head up, head down, left tilt and right tilt; the operation is divided into four groups of neuron synaptic weights which are arranged differently and are used for carrying out convolution and operation on input signals, then the four groups of neuron synaptic weights are processed through proper activation functions, and whether one of four output neurons is activated or not is determined according to the four groups of operation results, so that the attitude of the underwater vehicle represented by the group of input signals is determined; wherein the synaptic weights are obtained through training.