阅读说明：本技术 基于bp神经网络的自适应可重构比例-积分-微分控制器 (Self-adaptive reconfigurable proportional-integral-derivative controller based on BP neural network ) 是由 刘建富 陈楠 魏廷存 于 2021-08-01 设计创作，主要内容包括：本发明涉及一种基于BP神经网络的自适应可重构比例-积分-微分(P-I-D)控制器,克服了现有PID控制器中积分项和微分项对控制性能的影响,以提高数字电源的瞬态性能。所述的控制器是基于数据驱动的,主要由BP神经网络以及可重构控制器组成。首先,分别采集PI、PD、PID控制方式下的数据,然后利用这三组数据分别线下训练BP神经网络,产生三组不同的控制参数(神经网络的权值和偏置),每一组控制参数代表一种控制方式。数字电源正常工作时,神经网络根据输入电压、电感电流以及误差电压等参数产生数字占空比,以调节输出电压使其稳定于基准电压。可重构控制器根据数字电源输出电压的实时状态,自适应地切换神经网络的控制参数,从而实现控制方式的重构。(The invention relates to a self-adaptive reconfigurable proportional-integral-derivative (P-I-D) controller based on a BP neural network, which overcomes the influence of an integral term and a derivative term in the existing PID controller on the control performance so as to improve the transient performance of a digital power supply. The controller is driven based on data and mainly comprises a BP neural network and a reconfigurable controller. Firstly, data under PI, PD and PID control modes are respectively collected, then the three groups of data are utilized to respectively train a BP neural network under a line, three groups of different control parameters (weight and bias of the neural network) are generated, and each group of control parameters represents a control mode. When the digital power supply works normally, the neural network generates a digital duty ratio according to parameters such as input voltage, inductive current, error voltage and the like so as to regulate output voltage to be stable at reference voltage. The reconfigurable controller adaptively switches the control parameters of the neural network according to the real-time state of the output voltage of the digital power supply, thereby realizing the reconfiguration of the control mode.)

1. A self-adaptive reconfigurable proportional-integral-derivative controller based on a BP neural network is characterized by comprising the BP neural network, a logic operation unit and a reconfigurable control unit; the logic operation unit carries out delay operation and addition and subtraction operation on the output voltage to generate voltage errors of the current period, the previous period and the second period and the voltage error change rate of the current period; the reconfigurable control unit adaptively switches control parameters of the BP neural network under three control modes of PI, PD and PID according to the real-time change state of the output voltage so as to realize the reconfiguration of the control mode; the BP neural network generates a digital duty ratio according to the input voltage, the inductive current and the voltage error change rate so as to regulate the output voltage to be stable at the reference voltage.

2. The adaptive reconfigurable proportional-integral-derivative controller based on the BP neural network as claimed in claim 1, wherein the reconfigurable control unit comprises a lookup table LUT for storing PI, PD and PID control parameters, a multiplexer and a timing control module; the time sequence control module generates a control signal of the multiplexer according to the state variable of the output voltage; the lookup table LUT stores PI, PD and PID control parameters of the neural network; and the multiplexer switches the control parameters of the BP neural network according to the control signals to realize control mode reconstruction.

3. The adaptive reconfigurable proportional-integral-derivative controller based on the BP neural network as claimed in claim 1, wherein the BP neural network adopts a neural network with a single hidden layer, and the structure of the neural network is 6-10-1, i.e. 6 input nodes, 10 hidden nodes and 1 output node, then the neural network is respectively trained by three groups of data in an offline training mode through a gradient descent method to obtain three groups of weights and offsets which respectively represent control parameters of three control modes of PI, PD and PID, and the trained three groups of control parameters are stored in a lookup table for later use.

4. An adaptive method of a reconfigurable proportional-integral-derivative controller according to claim 1, characterized in that: the controller state switching has 6 conditions, the PI, the PD and the PID are switched mutually, and the specific state switching time sequence is designed as follows:

(1) the PI control is switched to the PD control, which indicates that the digital power supply enters a transient state from a steady state, and the controller needs to be switched to the PD control to regulate the output voltage so that the output voltage approaches to the reference voltage quickly; the trigger condition at this time is

(2) Switching from PD control to PID control indicates that the PD control reaches the limit and the voltage error cannot be reduced continuously, and the controller needs to switch from PD control to PID control to eliminate the voltage error, wherein the triggering condition (II) at the moment is

e_o[k]＝e_o[k-1]≠0

(3) Switching from PID control to PI control indicates that transient regulation of the digital power supply is completed and the digital power supply enters a steady state, the controller can maintain steady-state output only through PI control, and the triggering condition (c) at the moment is

e_o[k]＝e_o[k-1]＝0，

(4) Switching from PD control to PI control indicates that the digital power supply enters a steady state from a transient state, and in the transient state, the PD control just completely eliminates a voltage error, and the output voltage is stabilized at a reference voltage; the trigger condition at this time is

e_o[k]＝e_o[k-1]＝0

(5) The PI control is switched to the PID control, which also indicates that the digital power supply enters the transient state from the steady state, and the PI control is switched to the PD control, the former has small disturbance and can be quickly adjusted and finished only by the PID control, while the latter generally has large disturbance and more output voltage deviating from the reference voltage, and the controller needs to be switched to the PD control firstly to enable the output voltage to quickly approach the reference voltage, and then the PID is adopted to eliminate the voltage error; therefore, the trigger condition for switching from PI control to PID control is-

e_oc[k]＜e_oc[k-1]

(6) Switching from PID control to PD control indicates that one disturbance regulation of the digital power supply is not finished and a new disturbance is generated, namely superposition of the disturbances; under the regulation of PID control, the voltage error is gradually reduced and suddenly and newly disturbed, the voltage error is increased again, the controller needs to be switched from the original PID control to the PD control to quickly reduce the voltage error, and therefore the triggering condition is

{e_o[k]＞e_o[k-1]}||{e_o[k]＝e_o[k-1]≠0}。

Technical Field

The invention belongs to the field of power electronics, and relates to a self-adaptive reconfigurable proportional-integral-derivative (P-I-D) controller based on a BP neural network, which is particularly suitable for intelligently controlling a digital power supply.

Background

The structure of a digitally controlled DC-DC switching converter (hereinafter referred to as a digital power supply) is shown in FIG. 7. Analog output voltage V of load end_out(t) conversion to digital output V by ADC_out[k]Then V is added_out[k]And a reference voltage V_ref[k]Error signal e k between]Sent to a digital compensator. In the digital compensator, a specific digital control algorithm (such as PID algorithm) is adopted to calculate the digital duty ratio signal d [ k ]]Then, the Digital duty ratio signal d [ k ] is modulated by a Digital Pulse Width Modulator (DPWM)]Converting into analog duty ratio signal d (t), and finally driving power level switch S by Gate driver_pAnd S_nTo regulate the output voltage V_out(t) stabilizing the voltage at a reference voltage V_ref。

Aiming at a control algorithm of a digital power supply, a traditional PID control algorithm is based on a linear small signal model, adopts fixed gain and control coefficients, and has the advantages of simple structure and easiness in implementation, but the transient performance of the power supply is poor. The control algorithm combining the nonlinear control and the PID control, such as fuzzy PID control and neural network PID control, has better steady-state and transient performance. However, the adaptivity of the fuzzy PID control is limited by the number of fuzzy subsets, and although increasing the number of fuzzy subsets may improve the adaptivity of the control, the required hardware resources are greatly increased. The neural network PID control is usually based on data driving, theoretically can meet the requirements of each working point when training data are enough, and has strong adaptability. However, the neural network PID control is still a fixed structure based on PID control, and the influence of an integral term and a differential term which always exist in PID control on the transient response of the power supply cannot be overcome.

Documents "Improvement of Compensation efficiency of Neural Network Prediction for digital Controlled DC-DC Converter,2015IEEE International Telecommunications Energy Reference" and "A Reference Modification Model digital Controlled DC-DC Converter for Improvement of Transmission Response", IEEE Transmission on Power Elec, Vol.31, No.1, Jan.2016 ", successively propose a Neural Network PID controller applied to digital Power supply, the Neural Network adopts on-line training mode to adjust in real time according to state variables of the digital Power supplyControl coefficient (k) of PID controller_p,k_i,k_d) And a reference voltage V_refThe transient performance of the digital power supply is effectively improved. However, the method is limited by a fixed structure of the neural network PID controller, and the structure and control parameter reconfiguration of the controller cannot be realized, so that the influence of an integral term and a differential term on the transient response of the power supply is difficult to eliminate, and the further improvement of the transient performance is limited.

Disclosure of Invention

Technical problem to be solved

The transient performance of the digital power supply is improved by aiming at the influence of an integral term and a differential term on the control performance in the existing PID controller. The invention provides a self-adaptive reconfigurable P-I-D controller based on a BP neural network. The controller not only can realize the self-adaptive adjustment of the control coefficient, but also can realize the real-time reconstruction of the controller structure. The disturbance recovery time and overshoot of the digital power supply are effectively reduced, and the transient performance is remarkably improved.

Technical scheme

A self-adaptive reconfigurable proportional-integral-derivative controller based on a BP neural network is characterized by comprising the BP neural network, a logic operation unit and a reconfigurable control unit; the logic operation unit carries out delay operation and addition and subtraction operation on the output voltage to generate voltage errors of the current period, the previous period and the second period and the voltage error change rate of the current period; the reconfigurable control unit adaptively switches control parameters of the BP neural network under three control modes of PI, PD and PID according to the real-time change state of the output voltage so as to realize the reconfiguration of the control mode; the BP neural network generates a digital duty ratio according to the input voltage, the inductive current and the voltage error change rate so as to regulate the output voltage to be stable at the reference voltage.

The reconfigurable control unit comprises a lookup table LUT for storing PI, PD and PID control parameters, a multiplexer and a time sequence control module; the time sequence control module generates a control signal of the multiplexer according to the state variable of the output voltage; the lookup table LUT stores PI, PD and PID control parameters of the neural network; and the multiplexer switches the control parameters of the BP neural network according to the control signals to realize control mode reconstruction.

The BP neural network adopts a neural network with a single hidden layer, the structure of the BP neural network is 6-10-1, namely 6 input nodes, 10 hidden nodes and 1 output node are used, then the neural network is respectively trained by three groups of data in an offline training mode through a gradient descent method to obtain three groups of weights and offsets which respectively represent control parameters of three control modes of PI, PD and PID, and the trained three groups of control parameters are stored in a query table for later use.

An adaptive method of a reconfigurable proportional-integral-derivative controller, characterized by: the controller state switching has 6 conditions, the PI, the PD and the PID are switched mutually, and the specific state switching time sequence is designed as follows:

(1) the PI control is switched to the PD control, which indicates that the digital power supply enters a transient state from a steady state, and the controller needs to be switched to the PD control to regulate the output voltage so that the output voltage approaches to the reference voltage quickly; the trigger condition at this time is

(2) Switching from PD control to PID control indicates that the PD control reaches the limit and the voltage error cannot be reduced continuously, and the controller needs to switch from PD control to PID control to eliminate the voltage error, wherein the triggering condition (II) at the moment is

e_o[k]＝e_o[k-1]≠0

(3) Switching from PID control to PI control indicates that transient regulation of the digital power supply is completed and the digital power supply enters a steady state, the controller can maintain steady-state output only through PI control, and the triggering condition (c) at the moment is

e_o[k]＝e_o[k-1]＝0，

(4) Switching from PD control to PI control indicates that the digital power supply enters a steady state from a transient state, and in the transient state, the PD control just completely eliminates a voltage error, and the output voltage is stabilized at a reference voltage; the trigger condition at this time is

e_o[k]＝e_o[k-1]＝0

(5) The PI control is switched to the PID control, which also indicates that the digital power supply enters the transient state from the steady state, and the PI control is switched to the PD control, the former has small disturbance and can be quickly adjusted and finished only by the PID control, while the latter generally has large disturbance and more output voltage deviating from the reference voltage, and the controller needs to be switched to the PD control firstly to enable the output voltage to quickly approach the reference voltage, and then the PID is adopted to eliminate the voltage error; therefore, the trigger condition for switching from PI control to PID control is-

e_oc[k]＜e_oc[k-1]

(6) Switching from PID control to PD control indicates that one disturbance regulation of the digital power supply is not finished and a new disturbance is generated, namely superposition of the disturbances; under the regulation of PID control, the voltage error is gradually reduced and suddenly and newly disturbed, the voltage error is increased again, the controller needs to be switched from the original PID control to the PD control to quickly reduce the voltage error, and therefore the triggering condition is

{e_o[k]＞e_o[k-1]}||{e_o[k]＝e_o[k-1]≠0}。

Advantageous effects

Compared with the prior art, the self-adaptive reconfigurable P-I-D controller based on the neural network is data-driven and can meet the requirements of various working points. The neural network controller based on data driving not only has the adaptivity of a control coefficient, but also can be adaptively switched among three control modes of PI, PD and PID according to the state of output voltage so as to realize the real-time reconstruction of a proportional-integral-derivative controller structure. By adopting the self-adaptive reconfigurable P-I-D controller based on the BP neural network, the influence of an integral term and a differential term in a PID controller on the transient performance of a power supply is reduced, and the transient performance of the digital power supply is effectively improved.

Drawings

The drawings are only for purposes of illustrating particular embodiments and are not to be construed as limiting the invention, wherein like reference numerals are used to designate like parts throughout.

FIG. 1 is a diagram of a BP neural network-based adaptive reconfigurable P-I-D controller according to the present invention;

FIG. 2 illustrates a reconfigurable mechanism for a controller according to the present invention;

FIG. 3 illustrates the operation of the reconfigurable controller according to the present invention during a load disturbance;

FIG. 4 shows a process for obtaining neural network parameters;

FIG. 5 shows a hardware circuit architecture of a neural network employed in the present invention: (a) a normalization module of the neural network; (b) a circuit from BP neural network input layer to hidden layer; (c) a hidden layer to output layer circuit;

FIG. 6 is a comparison of the transient performance of the digital power supply using the present invention technique and the prior art neural network-PID (NN-PID) technique: (a) a start-up state; (b) the load current abruptly changes from 1.2A to 0.2A; (c) the load current abruptly changes from 0.2A to 1.2A; (d) the input voltage is suddenly changed from 5.0V to 6.0V; (e) the input voltage is suddenly changed from 5.0V to 4.0V;

fig. 7 shows a structure of a digitally controlled DC-DC switching converter (simply referred to as a digital power supply).

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

Referring to the attached figure 1, the adaptive reconfigurable P-I-D controller based on the BP neural network provided by the invention is composed of the BP neural network, a logic operation unit and a reconfigurable control unit, wherein the reconfigurable control unit is composed of a lookup table (LUT) for storing PI, PD and PID control parameters, a multiplexer and a time sequence control module. The input signal of the self-adaptive reconfigurable P-I-D controller based on the BP neural network is { V_out[k],I_ind[k],V_in[k]In which V is_out[k]Is the output voltage of the current cycle, I_ind[k]Is the inductor current for the current cycle,V_in[k]for the input voltage of the current cycle, the output signal is a digital duty cycle d_nn[k]. Logic operation unit pair output voltage V_out[k]Performing delay operation and addition/subtraction operation to generate { e_o[k],e_oc[k],e_o[k-1],e_o[k-2]In which e_o[k]Is the voltage error of the current cycle, e_o[k-1]And e_o[k-2]Voltage error of the previous and second periods, respectively, e_oc[k]＝e_o[k]-e_o[k-1]And is the voltage error change rate of the current cycle. The reconfigurable control unit adaptively switches the control parameters of the neural network according to the real-time change state of the output voltage so as to realize the reconfiguration of a control mode. Wherein the time sequence control module is used for controlling the time sequence according to the state variable { e) of the output voltage_o[k],e_oc[k]Generates the control signal ctrl for the multiplexer. The look-up table stores three sets of weights and offsets of the neural network, i.e., PI, PD and PID control parameters. And the multiplexer switches the weight and the bias of the BP neural network according to the control signal ctrl so as to realize control mode reconstruction. BP neural network according to input voltage V_inInductor current I_indAnd voltage error information e_o[k],e_o[k-1],e_o[k-2],e_oc[k]D, yielding a digital duty cycle d_nn[k]To regulate the output voltage. The specific design process of the time sequence control module is as follows:

1. and acquiring parameters of the neural network. The process of obtaining neural network parameters is shown in fig. 4, and first, data of three control modes (PI, PD, PID) are respectively collected, i.e. input information { V of the proposed controller_in,I_ind,e_o[k],e_o[k-1],e_o[k-2],e_oc[k]And output information d_nn[k]}. In order to avoid the influence of different dimensionalities of the data on the control precision of the neural network, the collected data is subjected to normalization processing. Input signal X_m[k]The algorithm for (m ═ 1, …,6) normalization is as follows:

in the invention, a neural network with a single hidden layer is adopted, and the structure of the neural network is 6-10-1, namely 6 input nodes, 10 hidden nodes and 1 output node are adopted. And then, respectively training the neural network by using three groups of data in an offline training mode and a gradient descent method to obtain three groups of weights and offsets which respectively represent control parameters of three control modes of PI, PD and PID, and storing the trained three groups of control parameters into a query table for later use. Then, a circuit is built as shown in fig. 5, wherein fig. 5a is a normalization module of the neural network, fig. 5b is a circuit from the input layer to the hidden layer of the BP neural network, and fig. 5c is a circuit from the hidden layer to the output layer.

2. Design of the reconfigurable controller. The reconfigurable controller adaptively switches the control parameters of the neural network according to the state of the output voltage, and is mainly used for selecting the control parameters of the neural network, so that the reconfiguration of a control mode is realized. As shown in fig. 1, the reconfigurable controller is composed of a look-up table (LUT), a multiplexer and a timing control module. The reconfigurable control state is shown in fig. 2. In a starting stage or an initial stage corresponding to a transient state of a power supply, because the output voltage deviates from the reference voltage more, the reconfigurable controller firstly selects the PD control parameter to realize PD control, so that the output voltage quickly approaches the reference voltage (the PD control parameter refers to a parameter trained by a neural network under PD control data, and the PI control parameter and the PID control parameter are the same as the PD control parameter). The conditions under which the controller is maintained in PD control are as follows:

{e_o[k]≠e_o[k-1]}&{e_o[k]＞e_PD[k]}&{e_o[k-1]＞e_PD[k]}， (4)

this means that the voltage error is changing in the PD control mode, which means that the PD control does not reach the limit, and the voltage error can be reduced continuously. Wherein e is_o[k]As a voltage error of the current cycle, e_o[k-1]Is the voltage error of the previous cycle, e_PD[k]Indicating a voltage error that cannot be eliminated when the PD control reaches a limit.

Because the PD control can not completely eliminate the voltage error, when the PD control can not continuously reduce the voltage error, the reconfigurable controller outputs PID control parameters to realize PID control so as to eliminate the voltage error, and the controller maintains the conditions of the PID control as follows:

{e_o[k]≠e_o[k-1]}&{0＜e_o[k]＜e_PD[k]}&{0＜e_o[k-1]＜e_PD[k]}， (5)

also indicating that the voltage error is changing all the time, which means that the voltage error has not been completely eliminated.

When the voltage error is eliminated, the output voltage is equal to the reference voltage, the controller is switched to PI control, a steady state is entered, and the output voltage is maintained to be stable at the reference voltage. In a steady state, the neural network adopts PI control parameters to realize PI control, so that the output voltage is maintained to be stable at the reference voltage. The conditions under which the controller maintains the PI control are:

{e_oc[k]＝e_oc[k-1]＝0}&{e_o[k]＝0}， (6)

the output voltage is stabilized at the reference voltage at the time when the change rate indicating the voltage error is always zero and the voltage error is also zero. Wherein eoc [ k ] represents the error change rate of the current cycle, and eoc [ k-1] represents the error change rate of the previous cycle.

The controller state switching has 6 conditions, the PI, the PD and the PID are switched mutually, and the specific state switching time sequence is designed as follows:

(1) and the PI control is switched to the PD control, which indicates that the digital power supply enters a transient state from a steady state, and the controller needs to be switched to the PD control to regulate the output voltage so that the output voltage is quickly close to the reference voltage. The trigger condition at this time is

Wherein e_oc[k]>e_oc[k-1]Indicating that the output voltage deviates from the reference voltage more and more quickly; e.g. of the type_oc[k]＝e_oc[k-1]Not equal to 0 indicates that the output voltage is deviating from the reference voltage at a constant speed; e.g. of the type_oc[k]＝e_oc[k-1]Not equal to 0 and e_o[k]＝e_o[k-1]Not equal to 0 indicates an output voltage deviationThe reference voltage, no longer changes.

(2) Switching from PD control to PID control indicates that the PD control reaches the limit and the voltage error cannot be reduced continuously, and the controller needs to switch from PD control to PID control to eliminate the voltage error, wherein the triggering condition (II) at the moment is

e_o[k]＝e_o[k-1]≠0， (8)

Indicating that the voltage error is no longer decreasing, i.e. the PD control reaches a limit.

(3) Switching from PID control to PI control indicates that transient regulation of the digital power supply is completed and the digital power supply enters a steady state, the controller can maintain steady-state output only through PI control, and the triggering condition (c) at the moment is

e_o[k]＝e_o[k-1]＝0， (9)

That is, the voltage error and the error change rate are both 0, indicating that the output voltage has stabilized at the reference voltage.

(4) Switching from PD control to PI control indicates that the digital power supply will go from transient to steady state, and in transient, the PD control just completely eliminates the voltage error, and the output voltage is stabilized at the reference voltage. The trigger condition at this time is

e_o[k]＝e_o[k-1]＝0， (10)

The trigger condition for switching PID control to PI control is the same.

(5) The PI control is switched to the PID control, which also indicates that the digital power supply enters the transient state from the steady state, and the PI control is switched to the PD control, the former has small disturbance and can be quickly adjusted and completed only by the PID control, while the latter generally has large disturbance and more output voltage deviating from the reference voltage, and the controller needs to be switched to the PD control firstly to enable the output voltage to quickly approach the reference voltage, and then the PID is adopted to eliminate the voltage error. Therefore, the trigger condition for switching from PI control to PID control is-

e_oc[k]＜e_oc[k-1]， (11)

Indicating that the rate of change of the error gradually decreases, i.e., the output voltage deviates from the reference voltage more and more slowly.

(6) Switching from PID control to PD control indicates that a disturbance adjustment of the digital power supply is not completed and a new disturbance, namely a superposition of disturbances, is generated. Under the regulation of PID control, the voltage error is gradually reduced and suddenly and newly disturbed, the voltage error is increased again, the controller needs to be switched from the original PID control to the PD control to quickly reduce the voltage error, and therefore the triggering condition is

{e_o[k]＞e_o[k-1]}||{e_o[k]＝e_o[k-1]≠0}， (12)

Wherein e_o[k]>e_o[k-1]Indicating that the voltage error is gradually increasing, i.e. the output voltage is deviating from the reference voltage; e.g. of the type_o[k]＝e_o[k-1]Not equal to 0 indicates that the output voltage deviates from the reference voltage and no longer varies.

Example (b):

here, the operation principle and the timing control of the reconfigurable control will be described by taking the load disturbance as an example. As shown in fig. 3, when the load current suddenly changes, the output voltage deviates from the reference voltage, and once the trigger condition (i) is satisfied, the neural network is switched from PI control to PD control so that the output voltage approaches the reference voltage as soon as possible, as shown in equation (1).

Wherein e is_oc[k]>e_oc[k-1]Indicating that the output voltage deviates from the reference voltage more and more quickly; e.g. of the type_oc[k]＝e_oc[k-1]Not equal to 0 indicates that the output voltage is deviating from the reference voltage at a constant speed; e.g. of the type_oc[k]＝e_oc[k-1]Not equal to 0 and e_o[k]＝e_o[k-1]Not equal to 0 indicates that the output voltage deviates from the reference voltage and does not change any more.

With the gradual reduction of the voltage error, when the voltage error meets the triggering condition II, as shown in the formula (2), the PD control reaches the limit, and the voltage error cannot be continuously eliminated, at the moment, the neural network is switched to PID control from the PD control, so that the voltage error is eliminated and the disturbance recovery time is reduced.

e_o[k]＝e_o[k-1]≠0， (2)

When e [ k ] ═ e [ k-1] ═ 0, namely satisfy the trigger condition (c), mean state error that represents the output voltage is eliminated, the neural network is switched from PID control to PI control in order to maintain the steady state.

Aiming at a Buck type digital power supply, an attached figure 6 is a comparison of transient response curves of the digital power supply in a starting state, load current jump +/-1A and input voltage jump +/-1V by adopting the technology of the invention and the existing neural network-PID (NN-PID) control technology respectively.

While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and various equivalent modifications or substitutions can be easily made by those skilled in the art within the technical scope of the present disclosure.