阅读说明：本技术 一种基于弱光栅的仿生温控神经网络系统及其工作方法 (Bionic temperature control neural network system based on weak grating and working method thereof ) 是由 王汉熙 梁恩邦 祁耀斌 黄鑫 于 2020-06-30 设计创作，主要内容包括：本发明涉及大规模温控领域,提供一种基于弱光栅的仿生温控神经网络系统及其工作方法,该系统包括基于弱光栅的测温神经网络、LED加热网络、嵌入式系统和供电系统；本发明中数万个甚至数十万个弱反射光纤光栅传感器和对应的LED,针对装备系统功能结构全面内外布局,共同组成定点测温-控温系统,分布于每个需要保温的点位,从而实现大规模密集集群传感。同时本发明系统和工作方法能够避免普通温控系统使用多台工控机所造成的信号延迟、冲突、误差等问题,提高了控制精度和控制效率,实现了精确及时的大量保温部位协同控制,具有类似于条件反射的实时测温控温、巡检测温能力,可满足高寒高海拔地区装备系统的工况保障需要。(The invention relates to the field of large-scale temperature control, and provides a bionic temperature control neural network system based on a weak grating and a working method thereof, wherein the system comprises a temperature measurement neural network based on the weak grating, an LED heating network, an embedded system and a power supply system; tens of thousands or even hundreds of thousands of weak reflection fiber grating sensors and corresponding LEDs in the invention are arranged inside and outside the equipment system in a comprehensive functional structure, and jointly form a fixed-point temperature measurement-control system which is distributed at each point position needing heat preservation, thereby realizing large-scale dense cluster sensing. Meanwhile, the system and the working method can avoid the problems of signal delay, conflict, error and the like caused by the use of a plurality of industrial personal computers in a common temperature control system, improve the control precision and the control efficiency, realize accurate and timely cooperative control of a large number of heat preservation parts, have the real-time temperature measurement and control and patrol detection capabilities similar to conditioned reflex, and can meet the working condition guarantee requirements of equipment systems in high-cold and high-altitude areas.)

1. A bionic temperature control neural network system based on weak gratings is characterized in that: the system comprises a temperature measurement neural network based on a weak grating, an LED heating network, an embedded system and a power supply system;

one or more temperature control nodes are arranged at each heat preservation part, and each temperature control node comprises a weak reflection fiber bragg grating sensor and an LED;

tens of thousands to hundreds of thousands of weak reflection fiber grating sensors form a temperature measurement neural network based on a weak grating, and the function of the temperature measurement neural network is to accurately acquire the temperature of each temperature control node;

tens of thousands to hundreds of thousands of LEDs form an LED heating network, and the function of the LED heating network is to accurately control the temperature of each temperature control node;

the embedded system is connected with the LED heating network and the temperature measurement neural network based on the weak grating, each weak reflection fiber grating sensor and each LED are accurately positioned in a pre-coding mode, and the pre-coding method comprises the following steps: each temperature control node is internally provided with a weak reflection fiber grating sensor and an LED, so that the embedded system can accurately position each weak reflection fiber grating sensor and LED only by encoding each temperature control node;

and the power supply system is connected with the embedded system, the LED heating network and the temperature measurement neural network based on the weak grating for power supply.

2. The weak grating-based bionic temperature control neural network system as claimed in claim 1, wherein: the weak reflection fiber grating sensor selects a low-reflectivity fiber grating with the reflectivity below 1%.

3. The weak grating-based bionic temperature control neural network system as claimed in claim 1, wherein: the remote monitoring system is in wireless connection with the embedded system and is used for remotely monitoring the working state of the temperature control neural network system.

4. A method for operating a bionic temperature-controlled neural network system based on weak grating as claimed in claim 1, which comprises the following steps:

(1) laying a weak reflection fiber grating sensor and an LED at each heat preservation part to form a temperature measurement-heat preservation integrated temperature control node, installing an embedded system in each heat preservation subsystem under an equipment system, connecting with a temperature measurement neural network and an LED heating network based on a weak grating, and accurately positioning to each temperature control node in a pre-coding mode;

(2) the embedded system issues a routing inspection instruction according to a preset program, the temperature measurement neural network based on the weak grating collects temperature data of each temperature control node according to codes, and the codes and the temperature data of each temperature control node are sent to the embedded system;

(3) the embedded system judges whether each temperature control node reaches the lowest working condition temperature, then changes related control parameters according to a preset program, and changes or maintains the current temperature by controlling the LED heating power corresponding to each temperature control node;

(4) and (3) issuing a new round of inspection instruction by the embedded system, and repeating the step (2) and the step (3), so that the temperature of each temperature control node is kept stable near the preset temperature.

5. The working method of the bionic temperature-controlled neural network system based on the weak grating as claimed in claim 4, wherein the working method comprises the following steps: when the weak reflection fiber grating sensor monitors abnormal data, the embedded system sends alarm information to the remote monitoring system.

Technical Field

The invention relates to the field of large-scale temperature control, in particular to a bionic temperature control neural network system based on a weak grating and a working method thereof.

Background

Equipment working in alpine/high altitude areas, mechanical systems, power supply systems, automatic measurement and control systems, computer systems and the like are often incapable of working normally due to severe cold climate. The reasons include that the fuel oil pipeline of the power system is frozen and blocked/frozen and broken; the mechanical system loses the motion condition or freezes temporarily due to the change of the kinematic pair matching tolerance caused by high and cold; the power supply system causes the open circuit of the power supply plug connector due to high and cold or the power running of the battery system; the sensor of the automatic measurement and control system fails at high and low temperature, and the reference working point drifts; the computer system may not boot properly, etc.

The special cold insulation/anti-freezing/heat insulation design of the existing equipment system is essentially a passive high-cold/high-altitude special working condition design scheme, namely, the lowest operation basic temperature is provided for equipment through a physical/chemical/mechanical cold insulation and anti-freezing scheme. The design scheme of the passive high-cold/high-altitude special working condition generally causes two serious problems, namely once the design scheme is separated from the high-cold/high-altitude climatic environment, the equipment is difficult to normally operate, and the use efficiency of the equipment is reduced; and secondly, equipment designed under the special working condition of high cold/high altitude is designed, so that the volume/dead weight and the like are increased, and the manufacturing cost/running cost/maintenance cost and the like are increased.

Disclosure of Invention

The invention aims to overcome the defects in the prior art, and provides a bionic temperature control neural network system based on a weak grating and a working method thereof aiming at the real-time temperature measurement-control requirement of a ten-thousand-point-level operation node, so that the large-scale intensive cluster heat preservation and temperature control of equipment are realized, meanwhile, the traditional passive heat preservation is changed into fixed-point accurate temperature supplement, and heat preservation heat is actively applied in real time to maintain the minimum operation temperature required by the equipment system.

The object of the invention is achieved by the following technical measures.

A bionic temperature control neural network system based on a weak grating comprises a temperature measurement neural network based on the weak grating, an LED heating network, an embedded system and a power supply system;

one or more temperature control nodes are arranged at each heat preservation part, and each temperature control node comprises a weak reflection fiber bragg grating sensor and an LED;

tens of thousands to hundreds of thousands of weak reflection fiber grating sensors form a temperature measurement neural network based on a weak grating, and the function of the temperature measurement neural network is to accurately acquire the temperature of each temperature control node;

tens of thousands to hundreds of thousands of LEDs form an LED heating network, and the function of the LED heating network is to accurately control the temperature of each temperature control node;

the embedded system is connected with the LED heating network and the temperature measurement neural network based on the weak grating, each weak reflection fiber grating sensor and each LED are accurately positioned in a pre-coding mode, and the pre-coding method comprises the following steps: each temperature control node is internally provided with a weak reflection fiber grating sensor and an LED, so that the embedded system can accurately position each weak reflection fiber grating sensor and LED only by encoding each temperature control node;

and the power supply system is connected with the embedded system, the LED heating network and the temperature measurement neural network based on the weak grating for power supply.

In the technical scheme, the weak reflection fiber grating sensor selects a low-reflectivity fiber grating with the reflectivity of less than 1%.

In the technical scheme, the bionic temperature control neural network system based on the weak grating further comprises a remote monitoring system, wherein the remote monitoring system is in wireless connection with the embedded system and is used for remotely monitoring the working state of the temperature control neural network system.

The invention also provides a working method of the bionic temperature control neural network system based on the weak grating, which comprises the following steps:

(1) laying a weak reflection fiber grating sensor and an LED at each heat preservation part to form a temperature measurement-heat preservation integrated temperature control node, installing an embedded system in each heat preservation subsystem under an equipment system, connecting with a temperature measurement neural network and an LED heating network based on a weak grating, and accurately positioning to each temperature control node in a pre-coding mode;

(2) the embedded system issues a routing inspection instruction according to a preset program, the temperature measurement neural network based on the weak grating collects temperature data of each temperature control node according to codes, and the codes and the temperature data of each temperature control node are sent to the embedded system;

(3) the embedded system judges whether each temperature control node reaches the lowest working condition temperature, then changes related control parameters according to a preset program, and changes or maintains the current temperature by controlling the LED heating power corresponding to each temperature control node;

(4) and (3) issuing a new round of inspection instruction by the embedded system, and repeating the step (2) and the step (3), so that the temperature of each temperature control node is kept stable near the preset temperature.

In the technical scheme, when the weak reflection fiber grating sensor monitors abnormal data, the embedded system sends alarm information to the remote monitoring system.

Tens of thousands or even hundreds of thousands of weak reflection fiber grating sensors and corresponding LEDs are arranged inside and outside aiming at the complete function structure of the equipment system, and jointly form a fixed-point temperature measurement-control system which is distributed at each point position needing heat preservation, each temperature control node simulates the neuron cells of a human body, and a temperature control neural network similar to the human body is built.

The invention can increase the multiplexing number of the weak reflection fiber grating sensor by a plurality of multiplexing modes such as time division multiplexing, wavelength division multiplexing, space division multiplexing and the like, thereby realizing large-scale dense cluster sensing. According to the existing technology, the temperature measurement and control of hundreds of thousands of temperature control nodes can be realized, and the requirements of large-scale equipment systems can be completely met. Meanwhile, the problems of signal delay, conflict, errors and the like caused by the fact that a common temperature control system uses a plurality of industrial personal computers can be avoided, the control precision and the control efficiency are improved, accurate and timely cooperative control of a large number of heat preservation parts is achieved, the bionic temperature control neural network is a temperature measurement neural network, an LED heating network and a heating automatic control system based on a weak grating, has the real-time temperature measurement and control and routing inspection temperature measurement capabilities similar to conditioned reflex, and can meet the working condition guarantee requirements of equipment systems in high-cold and high-altitude areas.

The invention overcomes the defects of the conventional temperature control system and has the following advantages:

(1) the traditional passive high-cold/high-altitude special working condition is changed into an active working condition, so that manpower and material resources required by special working condition design are saved, and meanwhile, the use scene of a common equipment system is expanded, so that the passive high-cold/high-altitude special working condition can be suitable for high-cold high-altitude areas.

(2) The multi-temperature-control-node quasi-distributed temperature measurement is realized through the weak reflection fiber bragg grating, and the fixed-point accurate temperature compensation is realized through the small LED, so that the precision of heat preservation and temperature control is ensured.

(3) The temperature measurement of one equipment system is realized by using one optical fiber, and each equipment system is independently controlled automatically. The remote monitoring system only needs to monitor whether the equipment system returns an abnormal signal or not, and the problems of signal delay, conflict, error and the like of the conventional temperature control system when the number of temperature control nodes is extremely large are solved.

(4) The temperature control nodes are extremely large in number, high in response speed and high in control precision, and the temperature control system is a novel temperature control system of a bionic neural network.

Drawings

FIG. 1 is a schematic diagram of a temperature controlled neural network of the present invention.

FIG. 2 is a flowchart illustrating temperature control of an embedded system according to the present invention.

FIG. 3 is a hardware schematic diagram of the temperature controlled neural network system of the present invention.

FIG. 4 is a schematic view of the temperature measurement-thermal insulation integrated structure of the present invention.

Wherein: 1. the LED light source comprises a protective shell, 2 parts of a bracket, 3 parts of a weak reflection fiber grating sensor and 4 parts of an LED.

Detailed Description

The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings.

As shown in fig. 1 and fig. 3, the bionic temperature control neural network system based on the weak grating of the present embodiment includes a temperature measurement neural network based on the weak grating, an LED heating network, an embedded system, a remote monitoring system, and a power supply system.

One or more temperature control nodes are arranged at each heat preservation part, and each temperature control node comprises a weak reflection fiber grating sensor and an LED.

Tens of thousands to hundreds of thousands of weak reflection fiber grating sensors form a temperature measurement neural network based on a weak grating, and the function of the temperature measurement neural network is to accurately acquire the temperature of each temperature control node; the weak reflection fiber grating can reach micron level, so that it may be laid easily in the positions needing heat preservation, such as kinematic pair, etc. and has excellent multiplexing capacity, so that one fiber may be used in precise temperature measurement of tens of thousands of heat preservation positions and several temperature control nodes in the same heat preservation position in one equipment system.

Tens of thousands to hundreds of thousands of LEDs form an LED heating network, and the function of the LED heating network is to accurately control the temperature of each temperature control node; the size of the LED can be controlled to be 1-2mm, so that the volume of a temperature measurement-heat preservation integrated structure (shown in figure 4) is extremely small, the temperature measurement-heat preservation integrated structure can be sequentially installed at a plurality of positions of a heat preservation part, and a plurality of one-to-one temperature control nodes are formed between each position needing heat preservation and the weak reflection fiber grating sensor, so that the temperature measurement-temperature control is more accurate.

Each heat preservation subsystem under the equipment system is provided with a set of embedded system, the embedded system is connected with an LED heating network and a temperature measurement neural network based on weak grating, each weak reflection fiber grating sensor and each LED are accurately positioned in a pre-coding mode, and the pre-coding method comprises the following steps: each heat preservation subsystem is provided with an independent embedded system and is connected with a remote monitoring system; each temperature control node is internally provided with a weak reflection fiber grating sensor and an LED, so that the embedded system can accurately position each weak reflection fiber grating sensor and the LED by only encoding each temperature control node, thereby accurately controlling each temperature control node, constructing a self-adaptive temperature control neural network by temperature measurement-heat preservation cooperative control of large-scale nodes, and realizing tens of thousands and even hundreds of thousands of points of heating self-control.

The remote monitoring system is wirelessly connected with the embedded system through technologies such as Bluetooth and wifi, and is used for remotely monitoring the working state of the temperature control neural network system; the system can be integrated in a remote control system as a module, and can also be independently developed into an APP to be installed in a mobile phone or a computer.

And the power supply system is connected with the embedded system, the LED heating network and the temperature measurement neural network based on the weak grating for power supply. The power supply system has two options, namely, the power supply system is directly integrated into the equipment system and is powered by the power supply of the equipment system; the other is to supply power by adopting an independent power supply mode of an external rechargeable battery.

The temperature control node is a position node needing heat preservation and comprises a sensing node (weak reflection fiber bragg grating sensor) and a heat preservation node (LED).

The heat preservation subsystem under the equipment system refers to a subsystem (such as a kinematic pair, an engine, a computer and the like) needing heat preservation under a certain equipment system (such as a mechanical system, a power supply system, a computer system and the like).

The heat preservation part refers to a part needing heat preservation in the equipment system, and can be obtained by consulting a product manual or performing analog simulation and then comparing with an actual use scene. The number of temperature control nodes of each heat preservation part is determined by parameters such as required heating value, required heat preservation volume, surface area and the like.

The weak reflection fiber grating is a low-reflectivity fiber grating having a reflectivity of 1% or less.

The temperature measurement-heat preservation integrated structure is a temperature control node structure, is composed of a weak reflection fiber grating sensor with a small volume and an LED, and can be directly installed at a position needing heat preservation. Fig. 4 provides a temperature measurement-thermal insulation integrated structure, wherein a protective shell 1 is used for preventing dust or impact from affecting a weak reflection fiber grating sensor 3 and an LED, and the weak reflection fiber grating sensor 3 and the LED are fixedly arranged on a bracket 2 and are electrically connected with an embedded system and a power supply system.

The embodiment also provides a working method of the bionic temperature control neural network system based on the weak grating, which comprises the following steps:

(1) laying a weak reflection fiber grating sensor and an LED at each heat preservation part to form a temperature measurement-heat preservation integrated temperature control node, installing an embedded system in a heat preservation subsystem under each equipment system, connecting the embedded system with a temperature measurement neural network and an LED heating network based on a weak grating, and accurately positioning each temperature control node in a pre-coding mode;

(2) the embedded system issues a routing inspection instruction according to a preset program, the temperature measurement neural network based on the weak grating collects temperature data of each temperature control node according to codes, and the codes and the temperature data of each temperature control node are sent to the embedded system;

(3) the embedded system judges whether each temperature control node reaches the lowest working condition temperature, then changes related control parameters according to a preset program, and changes or maintains the current temperature by controlling the LED heating power corresponding to each temperature control node;

(4) the embedded system issues a new round of inspection instructions, and repeats the step (2) and the step (3), so as to keep the temperature of each temperature control node stable near the preset temperature; the temperature control flow chart is shown in FIG. 2;

(5) when the weak reflection fiber grating sensor monitors abnormal data, the embedded system sends alarm information to the remote monitoring system. Namely, the remote monitoring system running on the industrial personal computer only needs to monitor whether each equipment system sends abnormal data or not, and a program built in the remote monitoring system in advance reacts to the abnormal data, so that the pressure of the remote monitoring system is reduced, the heat preservation and temperature control precision is improved, and the problems of signal conflict, delay and the like when the number of nodes of the common temperature control system is too large are solved.

In the above embodiment, the multiplexing mode of the thermometric neural network based on the weak grating may adopt a mode combining wavelength division multiplexing and time division multiplexing.

In the above embodiment, the temperature control mode in the preset program of the embedded system may further improve the temperature control accuracy by adopting a mode of combining proportional control and PID control. And E is set as a threshold value, and a proper E value is selected according to an experimental result. When | E | ≧ E, proportional control is adopted, because the deviation is large at the beginning, the adoption of proportional control is favorable for accelerating the adjustment speed, thereby improving the response speed of the system; when | E | < E, PID control is adopted, and the PID control is favorable for eliminating static errors of the system and suppressing interference, so that the control precision of the system is improved.

In the above embodiment, the precoding manner may adopt one of the following coding manners:

1. respectively encoding a mechanical system, a power supply system, an automatic measurement and control system and a computer system into JX, GD, ZK and JSJ;

2. the thermal insulation subsystems under each system are respectively coded, for example, an automatic measurement and control system can respectively code a measurement system and a control system as ZK.1 and ZK.2, and a mechanical system can respectively code a kinematic pair and an actuating system as JX.1 and JX.2;

3. and respectively encoding each temperature control node under each heat preservation subsystem, for example, the temperature control nodes in the kinematic pairs can be encoded into JX.1.1, JX.1.2 and JX.1.3.

The temperature control neural network system is a closed loop system, consists of tens of thousands or even hundreds of thousands of weak reflection fiber grating sensors to form a temperature measurement system, is comprehensively distributed inside and outside the functional structure of the equipment system, measures the temperature in real time, feeds the temperature back to the embedded system in real time, and implements autonomous temperature adjustment through the LED, thereby ensuring that the equipment system can be self-adaptive to the environmental temperature, constructing the working condition and not needing external participation; the remote monitoring system is added to play a monitoring role only.

Details not described in the present specification belong to the prior art known to those skilled in the art.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.