Neural network adaptive temperature control system and method based on DBD plasma reactor

Abstract

The invention provides a neural network adaptive temperature control system and method based on a DBD plasma reactor. In the system, a high-frequency and high-voltage alternating current power supply generates alternating current acting on the a end and the b end of a DBD plasma reactor; the infrared temperature sensor is used for detecting the temperature of a specified circular region in the DBD plasma reactor and transmitting the detected temperature data to the controller; the controller is used for receiving temperature data detected by the infrared temperature sensor, processing the data by adopting an RBF neural network, and outputting a control decision to the gas flow controller and the high-frequency controller to respectively realize flow control and flow power control; the gas flow controller supplies gas through a gas source, and outputs gas with corresponding flow parameters into the DBD plasma reactor through valve control. The temperature self-adaptive control method based on the RBF neural network is adopted, and the temperature of the DBD plasma reactor is effectively controlled by controlling the power of the high-frequency high-voltage alternating current and the flow velocity of the gas flow.

Description

Technical Field

[0001] The present invention relates to the technical field of temperature control, and in particular to a neural network adaptive temperature control system and method based on a DBD plasma reactor. Background Art

[0002] The energy consumption and heat generated by DBD non-equilibrium plasma under normal pressure is an important factor causing unnecessary reactor temperature rise, especially now that plasma reactors are more compact and use power supplies with higher energy and frequency, the problem of plasma reactor heating is more serious. From the perspective of particles, the huge temperature rise will affect the electronic and chemical processes. The appropriate temperature can regulate the reaction path of plasma active species, thereby achieving the treatment effect of the treated object. Traditional plasma reactor control can only control frequency, power, voltage amplitude, etc. For temperature control, due to the various reactors, temperature control has the characteristics of hysteresis, and the desired effect and control temperature range vary greatly, which makes temperature control difficult to achieve. Taking measures to reduce energy waste and using reasonable control decisions have become a broad consensus on DBD plasma reactor temperature control.

[0003] Temperature control has a high delay time. At the same time, the temperature characteristics of the DBD plasma reactor are affected by the change of the input voltage amplitude and the frequency change, which makes the temperature characteristics of the DBD plasma reactor present complex nonlinear characteristics. An adaptive system is a control system that can modify its own characteristics to adapt to the changes in the dynamic characteristics of the object and disturbance. This adaptive control method can achieve: during the operation of the system, it relies on the continuous collection of control process information to determine the current actual working state of the controlled object, optimize the performance criteria, and generate adaptive control laws, thereby adjusting the controller structure or parameters in real time, so that the system always automatically works in the optimal or suboptimal operating state. Summary of the invention

[0004] According to the technical problems raised above, a neural network adaptive temperature control system and method based on DBD plasma reactor is provided. The present invention adopts a temperature adaptive control method based on RBF neural network to effectively control the temperature of DBD plasma reactor by controlling the power of high-frequency high-voltage alternating current and the flow rate of gas flow.

[0005] The technical means adopted by the present invention are as follows:

[0006] A neural network adaptive temperature control system based on a DBD plasma reactor comprises: a high-frequency and high-voltage AC power supply, a gas flow controller, a gas source, a controller, an infrared temperature sensor and a DBD plasma reactor, wherein:

[0007] The high-frequency and high-voltage AC power supply is used to generate AC power, and the AC power acts on the a end and the b end of the DBD plasma reactor;

[0008] The infrared temperature sensor is used to detect the temperature of a specified circular area inside the DBD plasma reactor through a probe, and transmit the detected temperature data to the controller;

[0009] The controller is used to receive temperature data detected by the infrared temperature sensor, and use the RBF neural network to process the data, output control decisions to the gas flow controller to control the flow, and the high-frequency and high-voltage AC power supply to control the flow power;

[0010] The gas flow controller is supplied with gas by a gas source and controls the output of gas with corresponding flow parameters to enter the DBD plasma reactor through a valve.

[0011] Furthermore, the a-end and the b-end are electrode plates for dielectric barrier discharge. When working, the a-end and the b-end cycle and alternately perform two reaction processes of discharge and discharge recovery to generate a thermal effect.

[0012] Furthermore, the air source uses compressed air with a certain pressure so that the air has circulation power.

[0013] Furthermore, the DBD plasma reactor is provided with an air inlet and an air outlet, and the air inlet and the air outlet are connected through an air pipe.

[0014] The present invention also provides a neural network adaptive temperature control method based on a DBD plasma reactor, which is implemented based on the neural network adaptive temperature control system based on the DBD plasma reactor, comprising:

[0015] S1, set the input temperature Td through the controller;

[0016] S2, the controller sets the control law according to the set input temperature Td and the output of the RBF neural network, and outputs the control decision u1 to the gas flow controller, and outputs the control decision u2 to the high-frequency and high-voltage AC power supply;

[0017] S3, the gas flow controller receives the control decision u1, and converts the digital quantity of the control decision u1 into a physical quantity to change the gas flow;

[0018] S4, the high-frequency high-voltage AC power supply receives the control decision u2, and converts the digital quantity of the control decision u2 into a physical quantity, thereby changing the power of the high-frequency high-voltage AC power supply;

[0019] S5, changing the flow rate of the gas and the power of the high-frequency and high-voltage AC power supply through steps S3 and S4, thereby affecting the temperature of the DBD plasma reactor;

[0020] S6, infrared temperature sensor detects the temperature of DBD plasma reactor in real time;

[0021] S7, comparing the temperature of the DBD plasma reactor detected in real time by the infrared temperature sensor with the set input temperature Td to obtain an error value e;

[0022] S8. Input the error value e into the RBF neural network to perform adaptive control law calculations and adjust the controller parameters online to adapt to the uncertainty of the system and ensure the stability and performance of the system.

[0023] Furthermore, the mathematical models of the control decision u1 and the control decision u2 are as follows:

[0024]

[0025] In the above formula, represents the adaptive rate function of the control power; represents the adaptive rate function for controlling the gas flow rate; An estimation function representing a nonlinear function of high voltage AC power and temperature; An estimation function representing a nonlinear function of gas flow rate and temperature; Represents the second derivative of temperature; K 1 T represents the frequency control parameter matrix; K 2 T represents the gas flow control parameter matrix; E represents the error matrix.

[0026] Furthermore, the RBF neural network has a built-in input layer, a hidden layer and an output layer, wherein:

[0027] The input layer serves as the input matrix of the RBF neural network;

[0028] The hidden layer uses Gaussian basis function to improve the approximation accuracy of nonlinear functions;

[0029] The hidden layer is output to the output layer after operating with the built-in weights, which is the estimation function of f1(x) and f2(x).

[0030] Furthermore, the input layer includes an error value e, a derivative of the error value e, an output frequency f of a high-frequency and high-voltage AC power supply, an output voltage V of a high-frequency and high-voltage AC power supply, and an output flow rate of a gas flow controller.

[0031] Compared with the prior art, the present invention has the following advantages:

[0032] 1. The present invention provides a neural network adaptive temperature control system and method based on a DBD plasma reactor. The adaptive algorithm based on the RBF neural network can play a good role in controlling the temperature of the DBD plasma reactor, improving the ozone production efficiency, optimizing the reaction path of plasma active species, etc. Based on the advantages of the adaptive control algorithm itself, the DBD non-equilibrium plasma reaction can achieve parameter self-correction under changes in different environments. It is a control design with strong adaptability and good stability.

[0033] 2. The neural network adaptive temperature control system and method based on the DBD plasma reactor provided by the present invention can reduce energy waste and avoid excessive temperature rise.

[0034] Based on the above reasons, the present invention can be widely promoted in the fields of temperature control and the like. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0036] Figure 1 It is the basic circuit diagram of the existing temperature control system based on DBD plasma reactor.

[0037] Figure 2 It is the basic physical structure of temperature control of DBD plasma reactor of the present invention.

[0038] Figure 3 This is the structural diagram of the neural network adaptive temperature control system based on DBD plasma reactor.

[0039] Figure 4 This is a flow chart of the neural network adaptive temperature control method based on DBD plasma reactor.

[0040] In the figure: 101, 220V AC power supply; 102, 24V DC power supply; 103, high-frequency and high-voltage AC power supply in the existing temperature control system; 104, DBD plasma reactor in the existing temperature control system; 105, controller in the existing temperature control system; 106, gas flow controller in the existing temperature control system; 107, switch in the existing temperature control system; 108, switch; 109, first network cable; 110, second network cable; 111, third network cable; 112, fourth network cable; 201, high-frequency and high-voltage AC power supply; 202, gas flow controller; 203, gas source; 204, controller; 205, infrared temperature sensor; 206, DBD plasma reactor; 207, end a; 208, end b; 209, air outlet; 210, air inlet. DETAILED DESCRIPTION

[0041] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0042] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0043] like Figure 1 As shown, it is the basic circuit diagram of the existing temperature control system based on DBD plasma reactor. Figure 1In the embodiment, a 220v AC power supply 101 is connected to a 24v DC power supply 102 and a high-voltage high-frequency AC power supply 103 through electric wires. The output end of the high-voltage high-frequency AC power supply 103 is connected to a DBD plasma reactor 104, and the negative output port of the high-voltage high-frequency AC power supply 103 is grounded. At the same time, the negative input port of the DBD plasma reactor 104 is grounded. The two ends of the DBD plasma reactor 104 are connected to two electrode plates where dielectric barrier discharge occurs. The 24v DC power supply 102 outputs 24V DC power, which is connected to a controller 105, a gas flow controller 106, an infrared sensor 107 and a switch 108 to realize power supply. The controller 105 is connected to the switch 108 using a first network cable 109; the gas flow controller 106 is connected to the switch 108 through a third network cable 111; the infrared temperature sensor 107 is connected to the switch 108 through a second network cable 110; the high-voltage high-frequency AC power supply 103 is connected to the switch 108 through a fourth network cable 112, and the network connection adopts the PROFINET communication protocol. However, this type of plasma reactor control can only control frequency, power, voltage amplitude, etc., and it is difficult to achieve temperature control.

[0044] In view of the problems of the existing plasma reactor control in the above embodiments, the present invention provides a neural network adaptive temperature control system based on a DBD plasma reactor, such as Figure 2 As shown, it includes: a high-frequency high-voltage AC power supply 201, a gas flow controller 202, a gas source 203, a controller 204, an infrared temperature sensor 205 and a DBD plasma reactor 206, wherein:

[0045] The high-frequency and high-voltage AC power supply 201 is used to generate AC power, and the AC power acts on the a-end 207 and the b-end 8 of the DBD plasma reactor 206;

[0046] The infrared temperature sensor 205 is used to detect the temperature of a specified circular area inside the DBD plasma reactor 206 through a probe, and transmit the detected temperature data to the controller 204;

[0047] The controller 204 is used to receive the temperature data detected by the infrared temperature sensor 205, and use the RBF neural network to process the data, and output the control decision to the gas flow controller 202 to control the flow, and the high-frequency high-voltage AC power supply 201 to control the flow power;

[0048] The gas flow controller 202 is supplied with gas by the gas source 203 and controls the output of gas with corresponding flow parameters into the DBD plasma reactor 206 through valve control.

[0049] In specific implementation, as a preferred embodiment of the present invention, the a-end 207 and the b-end 208 are electrode plates for dielectric barrier discharge. When working, the a-end 207 and the b-end 208 cyclically and alternately perform two reaction processes of discharge and discharge recovery to generate a thermal effect.

[0050] In specific implementation, as a preferred embodiment of the present invention, the air source 203 uses compressed air with a certain pressure so that the air has circulation power.

[0051] In specific implementation, as a preferred embodiment of the present invention, the DBD plasma reactor 206 is provided with a gas inlet 210 and a gas outlet 209, and the gas inlet 210 and the gas outlet 209 are connected by a gas pipe.

[0052] like Figure 3 As shown, an embodiment of the present invention further provides a neural network adaptive temperature control method based on a DBD plasma reactor implemented based on the above neural network adaptive temperature control system based on a DBD plasma reactor, comprising:

[0053] S1, setting the input temperature Td through the controller 204;

[0054] S2, the controller 204 sets the control rule according to the set input temperature Td and the output of the RBF neural network, and outputs the control decision u1 to the gas flow controller 202, and outputs the control decision u2 to the high-frequency and high-voltage AC power supply 201;

[0055] S3, the gas flow controller 202 receives the control decision u1, and converts the digital quantity of the control decision u1 into a physical quantity to change the gas flow;

[0056] S4, the high-frequency high-voltage AC power source 201 receives the control decision u2, and converts the digital quantity of the control decision u2 into a physical quantity, thereby changing the power of the high-frequency high-voltage AC power source 201;

[0057] S5, changing the flow rate of the gas and the power of the high-frequency and high-voltage AC power supply through steps S3 and S4, thereby affecting the temperature of the DBD plasma reactor 206;

[0058] S6, the infrared temperature sensor 205 detects the temperature of the DBD plasma reactor 206 in real time;

[0059] S7, performing an error comparison between the temperature of the DBD plasma reactor 206 detected in real time by the infrared temperature sensor 205 and the set input temperature Td to obtain an error value e;

[0060] S8. Input the obtained error value e into the RBF neural network to perform adaptive control law calculation and adjust the parameters of the controller 204 online to adapt to the uncertainty of the system and ensure the stability and performance of the system.

[0061] In specific implementation, as a preferred embodiment of the present invention, the mathematical models of the control decision u1 and the control decision u2 are as follows:

[0062]

[0063] In the above formula, represents the adaptive rate function of the control power; represents the adaptive rate function for controlling the gas flow rate; An estimation function representing a nonlinear function of high voltage AC power and temperature; An estimation function representing a nonlinear function of gas flow rate and temperature; Represents the second derivative of temperature; K 1 T represents the frequency control parameter matrix; K 2 T represents the gas flow control parameter matrix; E represents the error matrix.

[0064] In specific implementation, as a preferred embodiment of the present invention, the RBF neural network has a built-in input layer, a hidden layer and an output layer, wherein:

[0065] The input layer serves as the input matrix of the RBF neural network;

[0066] The hidden layer uses a Gaussian basis function. The hidden layer in the present invention uses 20 neurons to improve the approximation accuracy of nonlinear functions.

[0067] The hidden layer is output to the output layer after operating with the built-in weights, which is the estimation function (approximation function) of f1(x) and f2(x).

[0068] In specific implementation, as a preferred embodiment of the present invention, the input layer includes the error value e, the derivative of the error value e, the output frequency f of the high-frequency and high-voltage AC power supply 201, the output voltage V of the high-frequency and high-voltage AC power supply 201, and the output flow rate of the gas flow controller 202.

[0069] like Figure 4 As shown, it is a flow chart of the method of the present invention, Figure 4 middle:

[0070] Initialization: assign variables to default values, set controls to default states, prepare the unprepared, and ensure that the entire system can function normally when it starts running for the first time.

[0071] Set the initial temperature value: After the system is initialized, you need to confirm the temperature value that needs to be controlled in the system.

[0072] Data reception and processing: In the entire system, the temperature value measured by the infrared temperature sensor, as well as the frequency and voltage parameters of the high-frequency and high-voltage alternating current need to be received, updated, and processed so that subsequent processes can use the processed data normally.

[0073] Error comparison e: The temperature of the DBD plasma reactor is compared with the set temperature value to obtain the error value e.

[0074] Adaptive control law operation: After receiving the error value e, the adaptive control law operation is performed to adjust the controller parameters online to adapt to the uncertainty of the system and ensure the stability and performance of the system.

[0075] RBF neural network algorithm: by receiving frequency, voltage, temperature, gas flow and other data, the nonlinear function of the system is approximated and learned.

[0076] Control law calculation and control decision output: After the system has processed the adaptive control law algorithm and the RBF neural network algorithm, it will perform algorithm calculations based on the control law and output the control decision. The control decision is the digital quantity that controls the high-frequency and high-voltage AC power and the gas flow of the gas flow controller. After receiving the digital quantity of the control decision, the two actuators need to convert it into physical quantity.

[0077] The above steps are repeated to run in a cycle, and the end of the entire system is set by the program or controlled manually.

[0078] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0079] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0080] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0081] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A neural network adaptive temperature control system based on a DBD plasma reactor, characterized in that: include: A high-frequency and high-voltage AC power supply (201), a gas flow controller (202), a gas source (203), a controller (204), an infrared temperature sensor (205) and a DBD plasma reactor (206), wherein: The high-frequency and high-voltage AC power source (201) is used to generate AC power, and the AC power acts on the a end (207) and the b end (8) of the DBD plasma reactor (206); The infrared temperature sensor (205) is used to detect the temperature of a specified circular area inside the DBD plasma reactor (206) through a probe, and transmit the detected temperature data to the controller (204); The controller (204) is used to receive temperature data detected by the infrared temperature sensor (205), process the data using an RBF neural network, and output a control decision to the gas flow controller (202) for flow control, and the high-frequency and high-voltage AC power supply (201) for flow power control; The gas flow controller (202) is supplied with gas through the gas source (203), and controls the output of gas with corresponding flow parameters into the DBD plasma reactor (206) through valve control.

2. The neural network adaptive temperature control system based on DBD plasma reactor according to claim 1, characterized in that: The a end (207) and the b end (208) are electrode plates for dielectric barrier discharge. When working, the a end (207) and the b end (208) cycle and alternately perform two reaction processes of discharge and discharge recovery to generate a thermal effect.

3. The neural network adaptive temperature control system based on DBD plasma reactor according to claim 1, characterized in that: The air source (203) uses compressed air of a certain pressure so that the air has circulation power.

4. The neural network adaptive temperature control system based on DBD plasma reactor according to claim 1, characterized in that: The DBD plasma reactor (206) is provided with an air inlet (210) and an air outlet (209), and the air inlet (210) and the air outlet (209) are connected via an air pipe.

5. A method for adaptive temperature control of a DBD plasma reactor using a neural network, which is implemented based on the neural network adaptive temperature control system of a DBD plasma reactor according to any one of claims 1 to 4, characterized in that: include: S1. Setting the input temperature Td through the controller (204); S2, the controller (204) sets the control rule according to the set input temperature Td and the output of the RBF neural network, and outputs the control decision u1 to the gas flow controller (202), and outputs the control decision u2 to the high-frequency and high-voltage AC power supply (201); S3, the gas flow controller (202) receives the control decision u1, and converts the digital value of the control decision u1 into a physical value, thereby changing the flow rate of the gas; S4, the high-frequency high-voltage AC power source (201) receives the control decision u2, and converts the digital quantity of the control decision u2 into a physical quantity, thereby changing the power of the high-frequency high-voltage AC power source (201); S5, changing the flow rate of the gas and the power of the high-frequency and high-voltage AC power supply through steps S3 and S4, thereby affecting the temperature of the DBD plasma reactor (206); S6. The infrared temperature sensor (205) detects the temperature of the DBD plasma reactor (206) in real time; S7, performing an error comparison between the temperature of the DBD plasma reactor (206) detected in real time by the infrared temperature sensor (205) and the set input temperature Td to obtain an error value e; S8. Input the obtained error value e into the RBF neural network to perform adaptive control law calculation and adjust the parameters of the controller (204) online to adapt to the uncertainty of the system and ensure the stability and performance of the system.

6. The neural network adaptive temperature control method based on DBD plasma reactor according to claim 5, characterized in that: The mathematical models of the control decision u1 and the control decision u2 are as follows: In the above formula, represents the adaptive rate function of the control power; represents the adaptive rate function for controlling the gas flow rate; An estimation function representing a nonlinear function of high voltage AC power and temperature; An estimation function representing a nonlinear function of gas flow rate and temperature; Represents the second derivative of temperature; K1 T represents the frequency control parameter matrix; K2 T represents the gas flow control parameter matrix; E represents the error matrix.

7. The neural network adaptive temperature control method based on DBD plasma reactor according to claim 5, characterized in that: The RBF neural network has a built-in input layer, a hidden layer and an output layer, wherein: The input layer serves as the input matrix of the RBF neural network; The hidden layer uses Gaussian basis function to improve the approximation accuracy of nonlinear functions; The hidden layer is output to the output layer after operating with the built-in weights, which is the estimation function of f1(x) and f2(x).

8. The neural network adaptive temperature control method based on DBD plasma reactor according to claim 7, characterized in that: The input layer includes an error value e, a derivative of the error value e, an output frequency f of a high-frequency and high-voltage AC power source (201), an output voltage V of the high-frequency and high-voltage AC power source (201), and an output flow rate of a gas flow controller (202).

Images