CN111982207A - Tank car liquid temperature sensor based on artificial neural network algorithm and method thereof - Google Patents
Tank car liquid temperature sensor based on artificial neural network algorithm and method thereof Download PDFInfo
- Publication number
- CN111982207A CN111982207A CN202010978574.1A CN202010978574A CN111982207A CN 111982207 A CN111982207 A CN 111982207A CN 202010978574 A CN202010978574 A CN 202010978574A CN 111982207 A CN111982207 A CN 111982207A
- Authority
- CN
- China
- Prior art keywords
- neural network
- artificial neural
- liquid temperature
- liquid level
- probe
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65D—CONTAINERS FOR STORAGE OR TRANSPORT OF ARTICLES OR MATERIALS, e.g. BAGS, BARRELS, BOTTLES, BOXES, CANS, CARTONS, CRATES, DRUMS, JARS, TANKS, HOPPERS, FORWARDING CONTAINERS; ACCESSORIES, CLOSURES, OR FITTINGS THEREFOR; PACKAGING ELEMENTS; PACKAGES
- B65D90/00—Component parts, details or accessories for large containers
- B65D90/22—Safety features
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65D—CONTAINERS FOR STORAGE OR TRANSPORT OF ARTICLES OR MATERIALS, e.g. BAGS, BARRELS, BOTTLES, BOXES, CANS, CARTONS, CRATES, DRUMS, JARS, TANKS, HOPPERS, FORWARDING CONTAINERS; ACCESSORIES, CLOSURES, OR FITTINGS THEREFOR; PACKAGING ELEMENTS; PACKAGES
- B65D90/00—Component parts, details or accessories for large containers
- B65D90/48—Arrangements of indicating or measuring devices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention relates to the field of automatic detection, in particular to a tank car liquid temperature sensor based on an artificial neural network algorithm and a method thereof; the liquid level detection device comprises a probe and a terminal system, wherein the probe has an integrated function for detecting temperature and liquid level, and is electrically connected with the terminal; the probe comprises a bottom plate, a cylinder, an energy converter, an upper cover, a sealing ring, a shielding wire and a temperature sensor, wherein one end of the cylinder is connected with the bottom plate, and the other end of the cylinder is connected with the upper cover; the liquid temperature sensing detection unit produced by adopting the integrated micro-device manufacturing technology can obviously reduce the external dimension and the power consumption of the liquid temperature and liquid level sensor, expands the application range of the sensor and is beneficial to the miniaturization of instrument equipment and instrument systems; the liquid temperature and the liquid level sensor are synchronously monitored, and the diagnosis of the safety of the tank car from multiple dimensions is facilitated.
Description
Technical Field
The invention relates to the field of automatic detection, in particular to a tank car liquid temperature sensor based on an artificial neural network algorithm and a method thereof.
Background
Various flammable and explosive dangerous goods have huge potential safety hazards in the transportation process, and once leakage occurs, poisoning, fire and even explosion accidents can be caused, so that the life and property safety of people is seriously harmed. Therefore, in the transportation process of various liquid dangerous goods, the physical states of the tank transportation media such as liquid level, temperature and the like are mastered in time, and the method is an important measure for ensuring the safety in the transportation process;
at present, the monitoring means for the liquid level of the storage tank are divided into contact type and non-contact type, the contact type mainly comprises a manual scale detection method, a floater measuring device, a servo type, a capacitance type and a magnetostriction type liquid level meter, and the common characteristics of the liquid level meter are that an induction element is in contact with the liquid to be measured, and the risk of abrasion, easy liquid adhesion, corrosion and the like exists; the non-contact type mainly comprises a microwave radar, a ray, a laser and an ultrasonic liquid level meter, and has the common characteristics that an induction element is not in contact with a measured liquid and is not influenced by a medium, and because a transport vehicle often jolts in the running process, the liquid level can generate irregular fluctuation change, the data collected by the liquid level meter is basically a non-stable random signal, and the difficulty is brought to the liquid level meter to obtain accurate liquid level data.
Meanwhile, two sensors are needed for measuring liquid level and temperature at present, when the liquid level and the temperature are installed on a tank body, more hole sites need to be formed, and the traditional non-contact liquid level meter and the traditional thermometer are large in size and large in occupied space.
Disclosure of Invention
Aiming at the problems in the prior art, the invention discloses a tank car liquid temperature sensor based on an artificial neural network algorithm and a method thereof, which can solve the problems of inaccurate liquid level measurement caused by bumping in the running process of a tank car and the miniaturization and integration of the sensor.
The invention relates to a tank car liquid temperature sensor based on an artificial neural network algorithm, which comprises a probe and a terminal system, wherein the probe has an integrated function for detecting temperature and liquid level, and is electrically connected with the terminal; the probe comprises a bottom plate, a cylinder, an energy converter, an upper cover, a sealing ring, a shielding lead and a temperature sensor, wherein one end of the cylinder is connected with the bottom plate, the other end of the cylinder is connected with the upper cover, the energy converter is arranged in the cylinder, the temperature sensor is arranged on the bottom plate and faces the end face of the upper cover, and the shielding lead is connected with the cylinder through the sealing ring; the terminal comprises a cover body, a display screen, a processor, a battery, a shell, an electric connector, a circuit board, a switch and a sealing ring, wherein the top of the shell is connected with the upper cover, the bottom of the shell is connected with the electric connector, the processor is arranged in the shell, the bottom of the processor is connected with the circuit board, the switch is further installed on the circuit board, the battery and the display screen which are used for providing a power supply are further arranged on the processor, the battery is connected with a battery bin cover, and the processor is connected with the shell in a sealing mode through the sealing ring.
Preferably, the cover body is transparent, and the cover body can display the content of the display screen.
Preferably, the display screen is provided with the liquid temperature and the liquid level state of the tank body.
Preferably, when the probe is connected with the terminal, the probe is connected with an electric connector through a shielding wire, wherein the electric connector is in a socket structure, and a plug structure matched with the socket is arranged on the shielding wire.
The method for the liquid temperature sensor for the tank car based on the artificial neural network algorithm comprises the following steps:
(1) firstly, under the condition that liquid is loaded in a tank car and the tank car is in a running state, liquid level and liquid temperature data in the running process of a transport vehicle are obtained by using a probe, the probe takes a signal obtained by detection as the input of an artificial neural network, and a display screen outputs final display data as the artificial neural network;
(2) controlling the artificial neural network to be in a learning state, inputting training sample data acquired in the step one into the artificial neural network, responding to an input variable by the neural network to generate network output, comparing the network output with target output, training the artificial neural network, and when the error between the network output and the target output does not meet the preset precision, adjusting the weight of the network by the neural network until the error is smaller than the preset precision;
(3) the artificial neural network training comprises a signal forward propagation process and an error backward propagation process, wherein the signal forward propagation process is that signals are transmitted layer by layer sequentially through input neurons, are subjected to nonlinear processing of a hidden layer and an output layer, and are finally output by output neurons, and the network weight is unchanged in the process;
the error back propagation process is to compare the output of the neural network with the target output, when the error is large, the error signals of the neural network and the target output are used as input signals to be propagated forwards layer by layer from the output layer of the network, and the back propagation enables the network weight of the neural network to be continuously corrected towards the direction in which the error function takes effect until the error is reduced to the preset precision;
(4) when the neural network is in a working state, the neural network responds to the input liquid temperature and the liquid level according to the trained network weight values in the step three, so that the liquid level and the liquid temperature in a vehicle running state do not change along with the running process, and the measurement accuracy of the liquid temperature and the liquid level is obviously improved.
Compared with the prior art, the invention has the following technical effects:
the liquid temperature sensing detection unit produced by adopting the integrated micro-device manufacturing technology can obviously reduce the external dimension and the power consumption of the liquid temperature and liquid level sensor, expands the application range of the sensor and is beneficial to the miniaturization of instrument equipment and instrument systems; the liquid temperature and the liquid level sensor are synchronously monitored, so that the diagnosis of the safety of the tank car from multiple dimensions is facilitated; the method has the advantages that the artificial neural network algorithm is utilized to process the monitoring data, data errors caused by liquid level floating in the driving process are effectively avoided, the system can be more accurately and reasonably judge the running state of the tank car, and the possibility of misjudgment is reduced.
Drawings
Fig. 1 is a schematic diagram of a terminal structure according to the present invention.
FIG. 2 is a schematic view of the probe of the present invention.
FIG. 3 is a schematic flow chart of the method of the present invention.
Reference numerals: 1-a cover body; 2-a display screen; 3-battery compartment cover; 4-a battery; 5-a shell; 6-an electrical connector; 7-a circuit board; 8-a switch; 9-sealing ring; 10-a base plate; 11-a barrel body; 12-a transducer; 13-upper cover; 14-a sealing ring; 15-shielded wire; 16-temperature sensor.
Detailed Description
Examples
The invention relates to a tank car liquid temperature sensor based on an artificial neural network algorithm, which comprises a probe and a terminal system, wherein the probe has an integrated function for detecting temperature and liquid level, and is electrically connected with the terminal;
the probe comprises a bottom plate 10, a cylinder 11, a transducer 12, an upper cover 13, a sealing ring 14, a shielding lead 15 and a temperature sensor 16, wherein one end of the cylinder 11 is connected with the bottom plate 10, the other end of the cylinder 11 is connected with the upper cover 13, the transducer 12 is arranged in the cylinder 11, the temperature sensor 16 is arranged on the bottom plate 10 and faces the end face of the upper cover 13, and the shielding lead 15 is connected with the cylinder 11 through the sealing ring 14; the terminal includes lid 1, display screen 2, treater, battery 4, casing 5, electric connector 6, circuit board 7, switch 8 and sealing washer 9, 5 tops of casing link to each other with upper cover 13, and 5 bottoms of casing link to each other with electric connector 6, are equipped with the treater in the casing 5, and its treater bottom links to each other with circuit board 7, still installs switch 8 on the circuit board 7, still is equipped with battery 4 and display screen 2 that are used for providing the power on the treater, and battery 4 links to each other with battery storehouse lid 3, link to each other through sealing washer 9 is sealed between treater and the casing 5. The cover body 1 is transparent, and the cover body 1 can display the content of the display screen 2. The display screen 2 is provided with the liquid temperature and the liquid level state of the tank body. When the probe is connected with the terminal, the probe is connected with the electric connector 6 through the shielding wire 15, wherein the electric connector 6 is in a socket structure, and the shielding wire 15 is provided with a plug structure matched with the socket.
The method for the liquid temperature sensor for the tank car based on the artificial neural network algorithm comprises the following steps:
(1) firstly, under the condition that liquid is loaded in a tank car and the tank car is in a running state, liquid level and liquid temperature data in the running process of a transport vehicle are obtained by using a probe, the probe takes a signal obtained by detection as the input of an artificial neural network, and a display screen outputs final display data as the artificial neural network;
(2) controlling the artificial neural network to be in a learning state, inputting training sample data acquired in the step one into the artificial neural network, responding to an input variable by the neural network to generate network output, comparing the network output with target output, training the artificial neural network, and when the error between the network output and the target output does not meet the preset precision, adjusting the weight of the network by the neural network until the error is smaller than the preset precision;
(3) the artificial neural network training comprises a signal forward propagation process and an error backward propagation process, wherein the signal forward propagation process is that signals are transmitted layer by layer sequentially through input neurons, are subjected to nonlinear processing of a hidden layer and an output layer, and are finally output by output neurons, and the network weight is unchanged in the process;
the error back propagation process is to compare the output of the neural network with the target output, when the error is large, the error signals of the neural network and the target output are used as input signals to be propagated forwards layer by layer from the output layer of the network, and the back propagation enables the network weight of the neural network to be continuously corrected towards the direction in which the error function takes effect until the error is reduced to the preset precision;
(4) when the neural network is in a working state, the neural network responds to the input liquid temperature and the liquid level according to the trained network weight values in the step three, so that the liquid level and the liquid temperature in a vehicle running state do not change along with the running process, and the measurement accuracy of the liquid temperature and the liquid level is obviously improved.
Claims (5)
1. The liquid temperature sensor for the tank car based on the artificial neural network algorithm is characterized by comprising a probe and a terminal system, wherein the probe has an integrated function for detecting temperature and liquid level, and the probe is electrically connected with the terminal;
the probe comprises a bottom plate (10), a cylinder body (11), a transducer (12), an upper cover (13), a sealing ring (14), a shielding wire (15) and a temperature sensor (16), wherein one end of the cylinder body (11) is connected with the bottom plate (10), the other end of the cylinder body is connected with the upper cover (13), the transducer (12) is arranged in the cylinder body (11), the temperature sensor (16) is arranged on the bottom plate (10) and faces the end face of the upper cover (13), and the shielding wire (15) is connected with the cylinder body (11) through the sealing ring (14);
the terminal comprises a cover body (1), a display screen (2), a processor, a battery (4), a shell (5), an electric connector (6), a circuit board (7), a switch (8) and a sealing ring (9), the top of the shell (5) is connected with an upper cover (13), the bottom of the shell (5) is connected with the electric connector (6), the processor is arranged in the shell (5), the bottom of the processor is connected with the circuit board (7), the switch (8) is further installed on the circuit board (7), the battery (4) and the display screen (2) which are used for providing power are further arranged on the processor, the battery (4) is connected with a battery bin cover (3), and the processor is connected with the shell (5) in a sealing mode through the sealing ring (9).
2. The tank car liquid temperature sensor based on the artificial neural network algorithm is characterized in that the cover body (1) is transparent, and the cover body (1) can display the content of the display screen (2).
3. The tank car liquid temperature sensor based on the artificial neural network algorithm as claimed in claim 2, wherein the display screen (2) is provided with the liquid temperature and the liquid level state of the tank body.
4. The tank car liquid temperature sensor based on the artificial neural network algorithm according to claim 1, wherein the probe is connected with the terminal through a shielding wire (15) and is connected with an electric connector (6), wherein the electric connector (6) is in a socket structure, and a plug structure matched with the socket is arranged on the shielding wire (15).
5. The method for the liquid temperature sensor for the tank car based on the artificial neural network algorithm according to any claim 1-4, characterized by comprising the following steps:
(1) firstly, under the condition that liquid is loaded in a tank car and the tank car is in a running state, liquid level and liquid temperature data in the running process of a transport vehicle are obtained by using a probe, the probe takes a signal obtained by detection as the input of an artificial neural network, and a display screen outputs final display data as the artificial neural network;
(2) controlling the artificial neural network to be in a learning state, inputting training sample data acquired in the step one into the artificial neural network, responding to an input variable by the neural network to generate network output, comparing the network output with target output, training the artificial neural network, and when the error between the network output and the target output does not meet the preset precision, adjusting the weight of the network by the neural network until the error is smaller than the preset precision;
(3) the artificial neural network training comprises a signal forward propagation process and an error backward propagation process, wherein the signal forward propagation process is that signals are transmitted layer by layer sequentially through input neurons, are subjected to nonlinear processing of a hidden layer and an output layer, and are finally output by output neurons, and the network weight is unchanged in the process;
the error back propagation process is to compare the output of the neural network with the target output, when the error is large, the error signals of the neural network and the target output are used as input signals to be propagated forwards layer by layer from the output layer of the network, and the back propagation enables the network weight of the neural network to be continuously corrected towards the direction in which the error function takes effect until the error is reduced to the preset precision;
(4) when the neural network is in a working state, the neural network responds to the input liquid temperature and the liquid level according to the trained network weight values in the step three, so that the liquid level and the liquid temperature in a vehicle running state do not change along with the running process, and the measurement accuracy of the liquid temperature and the liquid level is obviously improved.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010978574.1A CN111982207A (en) | 2020-09-17 | 2020-09-17 | Tank car liquid temperature sensor based on artificial neural network algorithm and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010978574.1A CN111982207A (en) | 2020-09-17 | 2020-09-17 | Tank car liquid temperature sensor based on artificial neural network algorithm and method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111982207A true CN111982207A (en) | 2020-11-24 |
Family
ID=73450849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010978574.1A Withdrawn CN111982207A (en) | 2020-09-17 | 2020-09-17 | Tank car liquid temperature sensor based on artificial neural network algorithm and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111982207A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103499374A (en) * | 2013-09-05 | 2014-01-08 | 江苏大学 | Method and system for ultrasonic dynamic liquid level measurement based on neural network |
| CN107607176A (en) * | 2017-10-16 | 2018-01-19 | 厦门伍迪电子科技有限公司 | A kind of sensor |
| CN108302330A (en) * | 2018-02-06 | 2018-07-20 | 韩俊 | A kind of domestic water monitoring device |
| CN108318107A (en) * | 2018-05-11 | 2018-07-24 | 中国电子科技集团公司第五十八研究所 | A kind of liquid-level switch based on TOF technologies |
| CN111238603A (en) * | 2018-11-29 | 2020-06-05 | 南通中集罐式储运设备制造有限公司 | Tank body and non-contact liquid level measuring device thereof |
-
2020
- 2020-09-17 CN CN202010978574.1A patent/CN111982207A/en not_active Withdrawn
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103499374A (en) * | 2013-09-05 | 2014-01-08 | 江苏大学 | Method and system for ultrasonic dynamic liquid level measurement based on neural network |
| CN107607176A (en) * | 2017-10-16 | 2018-01-19 | 厦门伍迪电子科技有限公司 | A kind of sensor |
| CN108302330A (en) * | 2018-02-06 | 2018-07-20 | 韩俊 | A kind of domestic water monitoring device |
| CN108318107A (en) * | 2018-05-11 | 2018-07-24 | 中国电子科技集团公司第五十八研究所 | A kind of liquid-level switch based on TOF technologies |
| CN111238603A (en) * | 2018-11-29 | 2020-06-05 | 南通中集罐式储运设备制造有限公司 | Tank body and non-contact liquid level measuring device thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2720006B1 (en) | Fuel tank monitoring system | |
| US5802910A (en) | Measuring system for liquid volumes and liquid levels of any type | |
| US10006895B2 (en) | System and method for monitoring atmospheric pollution | |
| US20040093943A1 (en) | Impedance level meter for liquids in tanks | |
| CN108871498A (en) | Condenser type adaptive universal liquid level emasuring device and method | |
| CN102183283A (en) | Capacitance-type liquid level measuring probe | |
| US11181403B2 (en) | Process variable sensor testing | |
| Santhosh et al. | Design of an instrument for liquid level measurement and concentration analysis using multisensor data fusion | |
| CN111982207A (en) | Tank car liquid temperature sensor based on artificial neural network algorithm and method thereof | |
| CN112540106A (en) | Metal concentration detection device and method for mixed oil | |
| CN207881790U (en) | A kind of and chain liquid level gauge of liquid-level switch | |
| CN113324623A (en) | Capacitance guided wave magnetic composite liquid level meter and measuring device | |
| CN208847287U (en) | Condenser type adaptive universal liquid level emasuring device | |
| CN103323075A (en) | Fuel level sensor | |
| CN201392335Y (en) | Dangerous liquid detector | |
| CN111198305A (en) | Liquid level potential sensor | |
| CN105784074A (en) | High-accuracy array type liquid level measurement device | |
| CN215114746U (en) | Capacitance/guided wave magnetic composite liquid level meter and measuring device | |
| CN110646061A (en) | Double-path redundant intelligent calibration radio frequency admittance anti-overflow liquid level meter and measurement method thereof | |
| CN201122646Y (en) | Accumulator with checking device | |
| CN108007525A (en) | A kind of level monitoring device and vehicle fuel level monitor system | |
| CN217900955U (en) | Integrated external measuring liquid level meter | |
| CN220871981U (en) | Pressure sensor calibration system | |
| CN206496816U (en) | A kind of Non-contact type liquid level detecting device | |
| Yahya | A multi-level storage tank gauging and monitoring system using a nanosecond pulse |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20201124 |