CN116578111A - Unmanned aerial vehicle flight control method and system for gas pipeline inspection data acquisition - Google Patents
Unmanned aerial vehicle flight control method and system for gas pipeline inspection data acquisition Download PDFInfo
- Publication number
- CN116578111A CN116578111A CN202310327534.4A CN202310327534A CN116578111A CN 116578111 A CN116578111 A CN 116578111A CN 202310327534 A CN202310327534 A CN 202310327534A CN 116578111 A CN116578111 A CN 116578111A
- Authority
- CN
- China
- Prior art keywords
- aerial vehicle
- unmanned aerial
- network
- sensor
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 63
- 238000000034 method Methods 0.000 title claims abstract description 45
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 title claims abstract description 33
- 230000009471 action Effects 0.000 claims description 40
- 230000006870 function Effects 0.000 claims description 21
- 238000012549 training Methods 0.000 claims description 20
- 238000004422 calculation algorithm Methods 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 17
- 238000004891 communication Methods 0.000 claims description 15
- 238000013480 data collection Methods 0.000 claims description 12
- 230000005540 biological transmission Effects 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000012512 characterization method Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000015654 memory Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 238000004088 simulation Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 abstract description 24
- 239000007789 gas Substances 0.000 description 27
- 239000003345 natural gas Substances 0.000 description 8
- 238000010586 diagram Methods 0.000 description 3
- 230000032683 aging Effects 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 229910052744 lithium Inorganic materials 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention discloses an unmanned aerial vehicle flight control method and system for acquiring gas pipeline inspection data. The invention can rapidly and accurately collect the measured value of the methane gas concentration at the target site in real time and accurately detect the working state of the gas pipeline.
Description
Technical Field
The invention belongs to the field of unmanned aerial vehicle control, and relates to an unmanned aerial vehicle flight control method and system for gas pipeline inspection data acquisition.
Background
The natural gas pipeline is an important component of a natural gas system, and the safe and reliable operation of the natural gas pipeline is directly related to the stable development of national economy, and the problems of natural environment protection and pollution control. The pipeline of the gas pipeline is exposed to natural environment for a long time, so that the pipeline is not only subjected to normal mechanical load and internal pressure of gas transportation, but also subjected to external damage such as rainfall, snowfall, landslide, artificial perforation and the like, and the factors can promote the ageing of various elements of the pipeline, and if the ageing is not found and eliminated in time, various faults can be developed, and serious threats are formed on the safety and stability of a natural gas system.
The main component of the natural gas is methane, and the concentration value of methane near the natural gas pipeline is detected by a ground sensor to judge whether leakage exists in the natural gas pipeline. At present, the common natural gas pipeline inspection still depends on the inspection personnel to inspect along the pipeline line. The method mostly adopts a manual recording mode, inspection is not in place, inspection points are omitted, data storage is incomplete, irregular and inaccurate, and the problem of data loss is serious. In addition, the manual inspection has the problems that the reporting time of hidden danger points (such as pipeline corrosion and pipeline leakage) is long, and the concrete coordinates of the hidden danger points are not clearly described. The gas pipelines are distributed in various environments such as plain, mountain areas, hills, deserts and the like, and complicated terrain distribution also increases the difficulty of inspection.
Disclosure of Invention
The invention aims to solve the problems of incomplete manual inspection, missing inspection points, incomplete data storage, inaccuracy, data loss, long reporting time and difficult inspection in the prior art, and provides an unmanned aerial vehicle flight control method and system for gas pipeline inspection data acquisition.
In order to achieve the purpose, the invention is realized by adopting the following technical scheme:
an unmanned aerial vehicle flight control method for gas pipeline inspection data acquisition comprises the following steps:
modeling the unmanned aerial vehicle in a gas pipeline area inspection and ground sensor data collection task as a Markov process model to obtain a unmanned aerial vehicle flight control system state space;
setting up an actor network and a criticism network;
training an actor network and a criticizer network, and acquiring a current system state space of the unmanned aerial vehicle and an action space of an unmanned aerial vehicle inspection flight system based on the unmanned aerial vehicle flight control system state space;
collecting gas pipeline inspection data sent by a ground sensor, and constructing a sensor data state representation network;
based on the sensor data state representation network, historical data acquired by a ground sensor and current gas pipeline inspection data, optimizing and predicting are carried out, the state of not-transmitted data and complete sensor time sequence state information are obtained, and the state information and the action space of the unmanned aerial vehicle system are transmitted to the unmanned aerial vehicle for adjustment.
The invention further improves that:
further, modeling the unmanned aerial vehicle in a gas pipeline area inspection and ground sensor data collection task as a Markov process model to obtain a state space of a flight control system of the unmanned aerial vehicle, specifically:
wherein b UAV (t) represents the state of the unmanned aerial vehicle battery, i represents the numbered i-th sensor among the N sensors on the ground; b i (t) and q i (t) represents the ground sensor battery status and the data size length transmitted by the sensor, respectively; g i (t) represents the data transmission channel gain of the ith sensor at time t; in order to estimate the energy and data arrival time consumed to acquire surface sensor data in an unplanned list, gamma is used i (t) represents a time span parameter, when takenGamma when the sensor data of the set is in mission planning i (t) has a value of 1, and when the acquired sensor data is not in the mission plan list, γ i (t) has a value of 0; s (t) is the selected sensor timing state; the unmanned aerial vehicle keeps flying at a fixed altitude, and the position coordinates of the unmanned aerial vehicle at the moment t are represented by (x (t), y (t) and z).
Further, an actor network and a criticism network are built, and the method specifically comprises the following steps:
establishing a current commentator network and a target commentator network which adopt the same structure and are used for simulating a system state action cost functionThe function represents a state space alpha k In the case of (1) perform an action->The value produced; establishing a current actor network and a target actor network, both using the same network structure, for simulating an action strategy function mu { alpha }, and k ∣w μ ' the function is represented in the state space alpha k Under the action executed by the unmanned aerial vehicle, w μ Is an actor network parameter matrix.
Further, training the actor network and the commentator network specifically comprises the following steps:
in an analog numerical simulation environment, calculating the battery state of the unmanned aerial vehicle at the time t of flight, and obtaining the system state space of the unmanned aerial vehicle at the current time; calculating an action space of the unmanned aerial vehicle inspection flight system, wherein the action space comprises the position and the speed of the unmanned aerial vehicle at the next moment and the sensor number of the acquired data;
the battery state algorithm for calculating the unmanned aerial vehicle at the time t is that the battery state of the unmanned aerial vehicle at the time t is calculated:
b UAV (t)＝b UAV (t-1)+Δb UAV (t)-ΔE UAV (t) (2)
wherein Δb UAV (t) represents the solar energy electric quantity collected by the unmanned plane at the time t-1 to t; use B UAV Representing the current position of the unmanned planeSetting a threshold value of the electric quantity required for returning to the ground charging station; in the flight inspection process, the unmanned aerial vehicle needs to keep b UAV (t)≥B UAV ；ΔE UAV And (t) represents the consumed electric quantity of the unmanned aerial vehicle at the time t, and the calculation mode is as follows:
wherein P is 0 And P' 0 Is a constant, ω (t) is the unmanned aerial vehicle motor speed, v 0 Is the average rotor running speed when hovering, v (t) is the instantaneous speed of the unmanned aerial vehicle, ζ drag And xi rotor Are respectively expressed as the resistance ratio of the fuselage and the firmness of the rotor wing, ρ air And S is rotor Air density and rotor disk area are shown, respectively.
Further, calculating an action space of the unmanned aerial vehicle inspection flight system, wherein the action space comprises a position and a speed of the unmanned aerial vehicle at the next moment and a sensor number of the acquired data, and specifically comprises the following steps:
a α ＝((x′(α),y′(α),z),(v x (α),v y (α)),i α ) (4)
wherein (x '(α), y' (α), z) is the next time unmanned plane position, (v) x (α),v y (alpha)) is the speed of the unmanned aerial vehicle in the horizontal plane direction at the next moment, i α Numbering the sensors to be acquired in the current state space; a, a α E a, a action space set, by collecting all actions that the drone can take to optimize the next location and speed of the drone, and selected ground sensors for data collection.
Further, training the actor network and the critic network further includes:
training a current evaluation home network by adopting a Belman algorithm, wherein a loss function is expressed as follows:
wherein the method comprises the steps ofDelta is discount factor, Q ′ For target criticism network output, Q is current criticism network output, and the current criticism network parameter w is updated by using a back propagation algorithm according to a loss function Q The method comprises the steps of carrying out a first treatment on the surface of the Periodically updating the target critics network parameter w by adopting a soft updating algorithm Q′ ：
w Q′ ←τw Q +(1-τ)w Q′ (6)
Where τ is the soft update operator;
and (3) performing optimization calculation on the above method by using a gradient ascent method, and updating the network parameter gradient of the current actor:
periodically updating the target critics network parameter w by adopting a soft updating algorithm Q′ ：
w μ′ ←τw μ +(1-τ)w μ′ (8)。
Further, collecting the gas pipeline inspection data sent by the ground sensor, and constructing a sensor data state representation network, wherein the method specifically comprises the following steps:
the sensor state representation network consists of a long-period memory network module, and the output of the last block in the network is used as the output of the state representation network; processing the input sequence by adding sensor new information to the memory, calculating a state characterization network output by controlling the gate to which new information is stored, old information is discarded, and the extent to which current information is utilized
Where σ is a sigmoid function, { W o ,W c ,W f ,W p The weight matrix, { e } is o ,e c ,e f ,e p -an offset matrix; whenever a timeWhen the unmanned plane selects one sensor for communication, the device reports the past and unreported states of the unmanned plane, and reports the states as time sequence signals to be input into a sensor state characterization network to obtain the complete sensor time sequence information state.
Further, the unmanned aerial vehicle further comprises information interaction with the ground station, the ground station receives flight data of the unmanned aerial vehicle and analyzes the flight data, and the ground station obtains a sensor state of the pipeline inspection area.
An unmanned aerial vehicle flight control system for gas line inspection data acquisition, comprising:
the construction module is used for modeling the unmanned aerial vehicle inspection and ground sensor data collection tasks in the gas pipeline area into a Markov process model to obtain a state space of a flight control system of the unmanned aerial vehicle;
the construction module is used for constructing an actor network and a criticism network;
the training module is used for training the actor network and the criticism network, and acquiring the current system state space of the unmanned aerial vehicle and the action space of the unmanned aerial vehicle inspection flight system based on the unmanned aerial vehicle flight control system state space;
the acquisition module is used for acquiring the gas pipeline inspection data sent by the ground sensor and constructing a sensor data state representation network;
the acquisition module performs optimization prediction based on the sensor data state representation network, the historical data acquired by the ground sensor and the current gas pipeline inspection data, acquires the state of the data which is not transmitted and complete sensor time sequence state information, and sends the state information and the complete sensor time sequence state information to the unmanned aerial vehicle to adjust the unmanned aerial vehicle system state space and the action space.
Compared with the prior art, the invention has the following beneficial effects:
according to the invention, the unmanned aerial vehicle is modeled as a Markov process model in the gas pipeline area inspection and ground sensor data collection tasks, a state space of an unmanned aerial vehicle flight control system is obtained, an actor network and a criticism network are built to guide the unmanned aerial vehicle to execute an optimal action strategy, and the sensor historical data and the current data are optimally predicted by building a sensor data state representation network, so that a complete sensor time sequence information state is obtained. The invention can collect the methane gas concentration measurement value of the target site rapidly and accurately in real time, and can accurately detect the working state of the gas pipeline.
Drawings
For a clearer description of the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of the method of the present invention;
FIG. 2 is a schematic diagram of a sensor state characterization network model according to the present invention;
fig. 3 is a structural diagram of an unmanned aerial vehicle flight control system for collecting gas pipeline inspection data.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
In the description of the embodiments of the present invention, it should be noted that, if the terms "upper," "lower," "horizontal," "inner," and the like indicate an azimuth or a positional relationship based on the azimuth or the positional relationship shown in the drawings, or the azimuth or the positional relationship in which the inventive product is conventionally put in use, it is merely for convenience of describing the present invention and simplifying the description, and does not indicate or imply that the apparatus or element to be referred to must have a specific azimuth, be configured and operated in a specific azimuth, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and the like, are used merely to distinguish between descriptions and should not be construed as indicating or implying relative importance.
Furthermore, the term "horizontal" if present does not mean that the component is required to be absolutely horizontal, but may be slightly inclined. As "horizontal" merely means that its direction is more horizontal than "vertical", and does not mean that the structure must be perfectly horizontal, but may be slightly inclined.
In the description of the embodiments of the present invention, it should also be noted that, unless explicitly specified and limited otherwise, the terms "disposed," "mounted," "connected," and "connected" should be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.
The invention is described in further detail below with reference to the attached drawing figures:
referring to fig. 1 and 2, the invention discloses a gas pipeline inspection data acquisition unmanned aerial vehicle flight control method, which comprises the following steps:
step 1: modeling the unmanned aerial vehicle in a gas pipeline area inspection and ground sensor data collection task as a Markov process model, and firstly, representing a system state space as:
wherein b UAV (t) represents the state of the unmanned aerial vehicle battery, i represents the numbered i-th sensor among the N sensors on the ground; b i (t) and q i (t) represents the ground sensor battery status and the data size length transmitted by the sensor, respectively; g i (t) represents the data transmission channel gain of the ith sensor at time t; in order to estimate the energy and data arrival time consumed to acquire surface sensor data in an unplanned list, gamma is used i (t) represents a time span parameter, γ when the acquired sensor data is in the mission plan i (t) has a value of 1, and when the acquired sensor data is not in the mission plan list, γ i (t) has a value of 0; s (t) is the selected sensor timing state; the unmanned aerial vehicle keeps flying at a fixed altitude, and the position coordinates of the unmanned aerial vehicle at the moment t are represented by (x (t), y (t) and z).
Preferably, the unmanned aerial vehicle adopts eight rotor unmanned aerial vehicle, and unmanned aerial vehicle battery adopts lithium cell series connection mode, adopts the imperial reaches TX2 integrated circuit board as airborne data processing computer.
Step 2: starting a training process, wherein a current system state space of the unmanned aerial vehicle is obtained in an analog numerical simulation environment, and in the training process, various parameter values of the system state space are measured and calculated by a sensor in the simulation environment, and particularly, the battery state of the unmanned aerial vehicle at the time of t flight is calculated by adopting the following formula, wherein the battery state belongs to a part of the system state space content:
b UAV (t)＝b UAV (t-1)+Δb UAV (t)-ΔE UAV (t) (2)
wherein Δb UAV (t) represents the solar energy electric quantity collected by the unmanned plane at the time t-1 to t; use B UAV Indicating the power threshold required for the drone to return from the current location to the ground charging station. In the flight inspection process, the unmanned aerial vehicle needs to keep b UAV (t)≥B UAV 。ΔE UAV (t) represents an unmanned personThe electricity consumed by the machine at the time t is calculated by the following steps:
wherein P is 0 And P 0 ' is a constant, ω (t) is the unmanned aerial vehicle motor speed, v 0 Is the average rotor running speed when hovering, v (t) is the instantaneous speed of the unmanned aerial vehicle, ζ drag And xi rotor Are respectively expressed as the resistance ratio of the fuselage and the firmness of the rotor wing, ρ air And S is rotor Air density and rotor disk area are shown, respectively.
Step 3: in a numerical simulation environment, calculating an action space of the unmanned aerial vehicle inspection flight system, wherein the action space comprises the position and the speed of the unmanned aerial vehicle at the next moment and the sensor number of the acquired data:
a α ＝((x′(α),y′(α),z),(v x (α),v y (α)),i α ) (4)
wherein (x '(α), y' (α), z) is the next time unmanned plane position, (v) x (α),v y (alpha)) is the speed of the unmanned aerial vehicle in the horizontal plane direction at the next moment, i α The sensors to be acquired in the current state space are numbered. a, a α E a, a action space set, by collecting all actions that the drone can take to optimize the next location and speed of the drone, and selected ground sensors for data collection.
Using Lβ.alpha.a α The expression system adopts action a in the current state space α And the packet loss amount of the data transmission of the ground sensor.
Step 4: the unmanned aerial vehicle flight control system consists of a critic network and an actor network, wherein the actor network function is used for outputting an action space used by the unmanned aerial vehicle for executing a flight measurement task at the current moment, and the critic network function is used for evaluating the output of the actor network in the training process so as to guide the network updating process. Establishing a current commentator network and a target commentator network which adopt the same structure and are used for simulating a systemState action cost functionThe function represents a state space alpha k In the case of (1) perform an action->The value produced; establishing a current actor network and a target actor network, both using the same network structure, for simulating an action strategy function mu { alpha }, and k ∣w μ ' the function is represented in the state space alpha k Under the action executed by the unmanned aerial vehicle, w μ Is an actor network parameter matrix. The commentator network input is a system state space and an action space, and the output is an action value Q function; the input to the actor network is the system state space and the output is the optimal action space currently to be performed by the drone.
Step 5: training a current evaluation home network by adopting a Belman algorithm, wherein a loss function is expressed as follows:
wherein delta is a discount factor, Q' is a target criticist network output, Q is a current criticist network output, and a back propagation algorithm is used for updating current criticist network parameters w according to a loss function Q . Periodically updating the target critics network parameter w by adopting a soft updating algorithm Q′ ：
w Q′ ←τw Q +(1-τ)w Q′ (6)
Where τ is the soft update operator.
And (3) performing optimization calculation on the above method by using a gradient ascent method, and updating the network parameter gradient of the current actor:
periodically updating the target critics network by adopting a soft updating algorithmParameters w of the collaterals Q′ ：
w μ′ ←τw μ +(1-τ)w μ′ (8)
Preferably, the commentator network comprises a 5-layer hidden layer network and the actor network structure comprises a 7-layer hidden layer network.
And according to the algorithm iteration steps, finishing the training process of the critics network and the actors network, and obtaining model network parameters.
Step 6: deploying a critic network and an actor network obtained through training on an unmanned aerial vehicle, wherein the unmanned aerial vehicle flies above a gas pipeline area, a ground sensor issues pipeline state data into a communication channel, the unmanned aerial vehicle receives sensor data in a current communication channel, a ground sensor state representation network is established, the state representation network predicts a state of not-transmitted data by capturing long-term representation features of time sequence data, the predicted state is obtained, and the sensor is complete in time sequence state informationAnd providing the information to the unmanned aerial vehicle. The sensor state representation network consists of long-term and short-term memory network modules, and the output of the last block in the network is used as the output of the state representation network. Processing the input sequence by adding new sensor information to the memory, calculating a state characterization network output by controlling the gate of the extent to which new information is stored, old information is discarded and current information is utilized>To be a select sensor timing state, it is part of the system state space:
wherein σ is a sigmoid function, { W o ,W c ,W f ,W p The weight matrix, { e } is o ,e c ,e f ,e p And is an offset matrix. Each time a drone selects a sensorIn communication, the device reports its past and unreported status (associated with each time slot since the device last reported) to the sensor status characterization network as a timing signal to obtain a complete sensor timing information status.
Preferably, the state characterization network is comprised of 50 long and short duration memory blocks.
Step 7: the unmanned aerial vehicle and the ground station are communicated by adopting a 5G airborne network communication system. The system of the space terminal uses a single board computer as a core, and communicates with the unmanned aerial vehicle flight control computer through a serial port, and adopts a bar Long Jidai chip 5G module. The ground terminal system runs a bottom communication algorithm, can restore a sending sequence from a plurality of channels of messages, and according to the characteristics of unmanned aerial vehicle-ground station data transmission, the performance of the system is improved as much as possible so as to meet the transmission requirements between unmanned aerial vehicle and ground stations, space terminal data are analyzed, and the ground stations acquire the sensor state of a pipeline inspection area.
Preferably, the unmanned aerial vehicle and the ground communication system adopt MH 5000-315 p 5G communication modules, DIGI XBER data transmission modules, and the 5G communication modules need external independent power supply, the maximum value of input voltage is 4.2V, the minimum value of input voltage is 3.7V, and the typical value is 4.0V. The data communication protocol algorithm divides the data stream into data packets of a size not exceeding 200 bytes, and assigns a unique sequence number (seq) to each data packet. The sequence number uniquely identifies the data packet and is incremented in the order of transmission time. The data output of the communication fusion algorithm must follow the time sequence of data transmission. Specifically, if the sequence number corresponding to the packet sequence output by the communication data fusion algorithm is { seq1, seq2, & gt, seqn }, it is necessary to ensure that seq1< seq2< & gt, & lt seqn. And in a certain time window, receiving the data packets but not outputting the data packets, and outputting the data packets according to a specified sequence (seq sequence) only when the data packets exceed the time window until the data packets exceeding the time window do not exist. The size of the time window is adjusted to find a balance between low latency and reliability, thereby restoring the transmitted sequence with high accuracy.
Referring to fig. 3, the invention discloses an unmanned aerial vehicle flight control system for gas pipeline inspection data acquisition, which comprises:
the construction module is used for modeling the unmanned aerial vehicle inspection and ground sensor data collection tasks in the gas pipeline area into a Markov process model to obtain a state space of a flight control system of the unmanned aerial vehicle;
the construction module is used for constructing an actor network and a criticism network;
the training module is used for training the actor network and the criticism network, and acquiring the current system state space of the unmanned aerial vehicle and the action space of the unmanned aerial vehicle inspection flight system based on the unmanned aerial vehicle flight control system state space;
the acquisition module is used for acquiring the gas pipeline inspection data sent by the ground sensor and constructing a sensor data state representation network;
the acquisition module performs optimization prediction based on the sensor data state representation network, the historical data acquired by the ground sensor and the current gas pipeline inspection data, acquires the state of the data which is not transmitted and complete sensor time sequence state information, and sends the state information and the complete sensor time sequence state information to the unmanned aerial vehicle to adjust the unmanned aerial vehicle system state space and the action space.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. The unmanned aerial vehicle flight control method for gas pipeline inspection data acquisition is characterized by comprising the following steps of:
modeling the unmanned aerial vehicle in a gas pipeline area inspection and ground sensor data collection task as a Markov process model to obtain a unmanned aerial vehicle flight control system state space;
setting up an actor network and a criticism network;
training an actor network and a criticizer network, and acquiring a current system state space of the unmanned aerial vehicle and an action space of an unmanned aerial vehicle inspection flight system based on the unmanned aerial vehicle flight control system state space;
collecting gas pipeline inspection data sent by a ground sensor, and constructing a sensor data state representation network;
based on the sensor data state representation network, historical data acquired by a ground sensor and current gas pipeline inspection data, optimizing and predicting are carried out, the state of not-transmitted data and complete sensor time sequence state information are obtained, and the state information and the action space of the unmanned aerial vehicle system are transmitted to the unmanned aerial vehicle for adjustment.
2. The unmanned aerial vehicle flight control method for acquiring the gas pipeline inspection data according to claim 1, wherein the unmanned aerial vehicle inspection and ground sensor data collection tasks in the gas pipeline area are modeled as a markov process model, and a unmanned aerial vehicle flight control system state space is obtained, specifically:
wherein b UAV (t) represents the state of the unmanned aerial vehicle battery, i represents the numbered i-th sensor among the N sensors on the ground; b i (t) and q i (t) represents the ground sensor battery status and the data size length transmitted by the sensor, respectively; g i (t) represents the data transmission channel gain of the ith sensor at time t; in order to estimate the energy and data arrival time consumed to acquire surface sensor data in an unplanned list, gamma is used i (t) represents a time span parameter, γ when the acquired sensor data is in the mission plan i (t) has a value of 1, and when the acquired sensor data is not in the mission plan list, γ i (t) has a value of 0; s (t) is the selected sensor timing state; the unmanned aerial vehicle keeps flying at a fixed altitude, and the position coordinates of the unmanned aerial vehicle at the moment t are represented by (x (t), y (t) and z).
3. The unmanned aerial vehicle flight control method for collecting gas pipeline inspection data according to claim 2, wherein the construction of the actor network and the commentator network is specifically as follows:
establishing a current commentator network and a target commentator network which adopt the same structure and are used for simulating a system state action cost functionThe function represents a state space alpha k In the case of (1) perform an action->The value produced; establishing a current actor network and a target actor network, both using the same network structure, for simulating an action strategy function mu { alpha }, and k ∣w μ ' the function is represented in the state space alpha k Under the action executed by the unmanned aerial vehicle, w μ Is an actor network parameter matrix.
4. The unmanned aerial vehicle flight control method for collecting gas line inspection data according to claim 3, wherein the training of the actor network and the criticism network is specifically as follows:
in an analog numerical simulation environment, calculating the battery state of the unmanned aerial vehicle at the time t of flight, and obtaining the system state space of the unmanned aerial vehicle at the current time; calculating an action space of the unmanned aerial vehicle inspection flight system, wherein the action space comprises the position and the speed of the unmanned aerial vehicle at the next moment and the sensor number of the acquired data;
the battery state algorithm for calculating the unmanned aerial vehicle at the time t is that the battery state of the unmanned aerial vehicle at the time t is calculated:
b UAV (t)＝b UAV (t-1)+Δb UAV (t)-ΔE UAV (t) (2)
wherein Δb UAV (t) represents the solar energy electric quantity collected by the unmanned plane at the time t-1 to t; use B UAV A power threshold value required for indicating that the unmanned aerial vehicle returns to the ground charging station from the current position; in the flight inspection process, the unmanned aerial vehicle needs toHold b UAV (t)≥B UAV ；ΔE UAV And (t) represents the consumed electric quantity of the unmanned aerial vehicle at the time t, and the calculation mode is as follows:
wherein P is 0 And P' 0 Is a constant, ω (t) is the unmanned aerial vehicle motor speed, v 0 Is the average rotor running speed when hovering, v (t) is the instantaneous speed of the unmanned aerial vehicle, ζ drag And xi rotor Are respectively expressed as the resistance ratio of the fuselage and the firmness of the rotor wing, ρ air And S is rotor Air density and rotor disk area are shown, respectively.
5. The unmanned aerial vehicle flight control method for acquiring the gas pipeline inspection data according to claim 4, wherein the calculation of the action space of the unmanned aerial vehicle inspection flight system comprises the position and the speed of the unmanned aerial vehicle at the next moment and the sensor number of the acquired data, and specifically comprises the following steps:
a α ＝((x′(α),y′(α),z),(v x (α),v y (α)),i α ) (4)
wherein (x '(α), y' (α), z) is the next time unmanned plane position, (v) x (α),v y (alpha)) is the speed of the unmanned aerial vehicle in the horizontal plane direction at the next moment, i α Numbering the sensors to be acquired in the current state space; a, a α E a, a action space set, by collecting all actions that the drone can take to optimize the next location and speed of the drone, and selected ground sensors for data collection.
6. The unmanned aerial vehicle flight control method of claim 5, wherein training the actor network and the commentator network further comprises:
training a current evaluation home network by adopting a Belman algorithm, wherein a loss function is expressed as follows:
wherein delta is a discount factor, Q' is a target criticist network output, Q is a current criticist network output, and a back propagation algorithm is used for updating current criticist network parameters w according to a loss function Q The method comprises the steps of carrying out a first treatment on the surface of the Periodically updating the target critics network parameter w by adopting a soft updating algorithm Q′ ：
w Q′ ←τw Q +(1-τ)w Q′ (6)
Where τ is the soft update operator;
and (3) performing optimization calculation on the above method by using a gradient ascent method, and updating the network parameter gradient of the current actor:
periodically updating the target critics network parameter w by adopting a soft updating algorithm Q′ ：
w μ′ ←τw μ +(1-τ)w μ′ (8)。
7. The unmanned aerial vehicle flight control method for collecting gas pipeline inspection data according to claim 6, wherein the collecting of the gas pipeline inspection data sent by the ground sensor constructs a sensor data state representation network, specifically:
the sensor state representation network consists of a long-period memory network module, and the output of the last block in the network is used as the output of the state representation network; processing the input sequence by adding sensor new information to the memory, calculating a state characterization network output by controlling the gate to which new information is stored, old information is discarded, and the extent to which current information is utilized
Where σ is a sigmoid function, { W o ,W c ,W f ,W p The weight matrix, { e } is o ,e c ,e f ,e p -an offset matrix; whenever the drone selects one of the sensor communications, the device reports its past and unreported status, reporting as a timing signal to the sensor status characterization network, resulting in a complete sensor timing information status.
8. The method for controlling the flight of the unmanned aerial vehicle for collecting the inspection data of the gas pipeline according to claim 7, wherein the unmanned aerial vehicle further comprises information interaction with a ground station, the ground station receives and analyzes the flight data of the unmanned aerial vehicle, and the ground station obtains the sensor state of the inspection area of the pipeline.
9. Unmanned aerial vehicle flight control system of data acquisition is patrolled and examined to gas line, a serial communication port, include:
the construction module is used for modeling the unmanned aerial vehicle inspection and ground sensor data collection tasks in the gas pipeline area into a Markov process model to obtain a state space of a flight control system of the unmanned aerial vehicle;
the construction module is used for constructing an actor network and a criticism network;
the training module is used for training the actor network and the criticism network, and acquiring the current system state space of the unmanned aerial vehicle and the action space of the unmanned aerial vehicle inspection flight system based on the unmanned aerial vehicle flight control system state space;
the acquisition module is used for acquiring the gas pipeline inspection data sent by the ground sensor and constructing a sensor data state representation network;
the acquisition module performs optimization prediction based on the sensor data state representation network, the historical data acquired by the ground sensor and the current gas pipeline inspection data, acquires the state of the data which is not transmitted and complete sensor time sequence state information, and sends the state information and the complete sensor time sequence state information to the unmanned aerial vehicle to adjust the unmanned aerial vehicle system state space and the action space.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310327534.4A CN116578111A (en) | 2023-03-24 | 2023-03-24 | Unmanned aerial vehicle flight control method and system for gas pipeline inspection data acquisition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310327534.4A CN116578111A (en) | 2023-03-24 | 2023-03-24 | Unmanned aerial vehicle flight control method and system for gas pipeline inspection data acquisition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116578111A true CN116578111A (en) | 2023-08-11 |
Family
ID=87532935
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310327534.4A Pending CN116578111A (en) | 2023-03-24 | 2023-03-24 | Unmanned aerial vehicle flight control method and system for gas pipeline inspection data acquisition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116578111A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116736891A (en) * | 2023-08-15 | 2023-09-12 | 众芯汉创（北京）科技有限公司 | Autonomous track planning system and method for multi-machine collaborative inspection power grid line |
-
2023
- 2023-03-24 CN CN202310327534.4A patent/CN116578111A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116736891A (en) * | 2023-08-15 | 2023-09-12 | 众芯汉创（北京）科技有限公司 | Autonomous track planning system and method for multi-machine collaborative inspection power grid line |
| CN116736891B (en) * | 2023-08-15 | 2023-10-20 | 众芯汉创（北京）科技有限公司 | Autonomous track planning system and method for multi-machine collaborative inspection power grid line |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210255158A1 (en) | Emissions estimate model algorithms and methods | |
| CN108107911B (en) | Solar airplane autonomous optimization flight path planning method | |
| CN108614274B (en) | Cross type crossing line distance measuring method and device based on multi-rotor unmanned aerial vehicle | |
| Chung et al. | Placement and routing optimization for automated inspection with unmanned aerial vehicles: A study in offshore wind farm | |
| CN105093130A (en) | Unmanned aerial vehicle cruising capacity monitoring system and method | |
| CN109886396A (en) | A kind of transmission line galloping on-line prediction system and method | |
| CN107943078A (en) | More rotor dual systems unmanned plane inspection fault diagnosis systems and method | |
| CN102722178A (en) | Electric field measuring obstacle avoidance system and method for live wire routing inspection of unmanned aerial vehicle | |
| CN205120933U (en) | Duration monitoring system and unmanned vehicles with accuse platform flies thereof | |
| CN202632111U (en) | Electric field measurement obstacle avoidance system for polling live wire by unmanned aerial vehicle | |
| CN116578111A (en) | Unmanned aerial vehicle flight control method and system for gas pipeline inspection data acquisition | |
| CN110082498A (en) | A kind of landslide monitoring data unmanned plane acquisition system based on wireless sensor Internet of Things | |
| CN103713305B (en) | A kind of location neck ring and localization method | |
| CN107153424A (en) | A kind of full-automatic unmanned machine of energy anti-strong wind patrols winged system | |
| CN207965138U (en) | A kind of unmanned plane radar measuring flow system | |
| CN112306092A (en) | Unmanned aerial vehicle system of patrolling and examining | |
| Daponte et al. | DronesBench: An innovative bench to test drones | |
| CN108609202A (en) | Flight is jolted prediction model method for building up, prediction technique and system | |
| CN109829534A (en) | Evaluation method, the device and system of opencast coal mine dump stability of community | |
| CN106603002B (en) | A kind of photovoltaic plant fault detection system | |
| CN114153224A (en) | Micro-meteorology-based unmanned aerial vehicle flight path intelligent planning control system and method | |
| CN114637220A (en) | Semi-physical test system and method for application verification of robot operating system | |
| CN110798351A (en) | Power grid fault detection point deployment method based on PSO and ant colony-genetic algorithm | |
| CN114610070A (en) | Unmanned aerial vehicle-cooperated wind power plant intelligent inspection method | |
| CN109471452A (en) | Remote engineering and geological disaster safety monitoring system and method based on unmanned plane |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |