An event-triggered control system and method with high-order unmodeled dynamics

Abstract

The present invention discloses an event-triggered control system and method with high-order unmodeled dynamics, including: using a sensor to measure the output signal of a controlled object; an estimator to calculate parameter estimates; a controller to solve the control signal according to a robust adaptive control algorithm; an event trigger to determine whether to trigger an event according to the control signal using an event trigger condition; a communication network module to determine whether to open the communication network according to the event trigger result, and when the communication network is opened, the controller sends a control signal to the actuator, and the actuator acts according to the latest control signal received. The present invention designs an event-triggered robust adaptive control method for discrete uncertain systems with high-order unmodeled dynamics, and is suitable for discrete systems that cannot obtain high-order dynamics due to physical reasons, such as DC motor control systems. At the same time, the introduction of the event trigger mechanism reduces the system communication energy consumption and the actuator action frequency, avoids excessively frequent actuator actions, and increases the actuator service life.

Description

Technical Field

[0001] The present invention belongs to the technical field of uncertain network physical system control, and in particular, relates to an event-triggered control system and method with high-order unmodeled dynamics. Background Art

[0002] Uncertainties are usually unavoidable in the modeling process of actual network-physical control systems, such as unknown parameters, high-order unmodeled dynamics, external disturbances, etc. High-order unmodeled dynamics exist in control system models as a typical uncertainty. For example, in order to simplify the controller design, some high-order dynamic parts are deliberately ignored to obtain a reduced-order model, or high-order models cannot be obtained for physical reasons. For example, in a DC motor control system, a second-order discrete system is used to describe the DC motor controlled object model, but the actual DC motor control model is higher than the second order, and the high-order dynamic part is difficult to obtain directly for physical reasons. Therefore, it is of great significance to study the control method of systems with high-order unmodeled dynamics.

[0003] In cyber-physical control systems, sensor sampling, controller calculations, and communication network transmissions need to be limited within the system load range. The classic control strategy is based on periodic signal sampling and transmission. In order to ensure system control performance, the signal transmission period usually needs to be set very small, resulting in a waste of communication resources. The idea of ​​event-triggered control is to design events to characterize the state of the system, and then determine whether to transmit the system signal based on whether the event triggering conditions are met, in order to replace the fixed-periodic communication method. As a non-periodic communication strategy, event-triggered communication has the advantage of reducing the frequency of communication. How to design event-triggered control methods for systems with high-order unmodeled dynamics is an urgent problem to be solved. Summary of the invention

[0004] In order to solve the problems in the prior art, the present invention provides an event-triggered control system and method with high-order unmodeled dynamics; by introducing an event trigger mechanism in the communication network from the controller module to the actuator module, the communication frequency and the actuator action frequency are reduced, and the estimator is used to estimate the high-order and low-order unknown parameters of the controlled object, and the robust adaptive controller ensures the stability of the system's bounded input and bounded output.

[0005] The technical solution adopted by the present invention is as follows:

[0006] An event-triggered control system with high-order unmodeled dynamics, comprising:

[0007] The sensor module is used to measure the output signal of the controlled object and send it to the estimator module;

[0008] An estimator module is used to estimate unknown parameters of the controlled object according to the input and output signals of the controlled object, and send the parameter estimation results to the controller module;

[0009] A controller module, used for calculating a control signal according to the received estimator data and sending the control signal to the event trigger module;

[0010] An event trigger module, used to determine whether to trigger an event according to the received control signal, and when the event trigger condition is met, the event trigger sends a flag signal 1 to the communication network module, and when the event trigger condition is not met, the event trigger sends a flag signal 0 to the communication network module;

[0011] The communication network module is used to determine whether to open the communication network according to the received event trigger flag signal 1 or 0; when the communication network is opened, the controller module sends data to the actuator module, otherwise the controller module does not send data to the actuator module;

[0012] The actuator module is used to perform actions on the controlled object according to the received control signal.

[0013] Furthermore, the controlled object is an n+N order discrete uncertain system, in which the parameters of the N order high-order dynamics are unknown; the parameters of the n order low-order dynamics are unknown, but the set of parameters of the n order low-order dynamics is known; at the same time, the system is subject to unknown time-varying interference, and the upper bound of the interference is unknown.

[0014] Furthermore, the estimator module estimates unknown parameters of the controlled object based on an improved projection algorithm according to the received historical output signals of the controlled object, historical input signals, and a set of n-order low-order dynamic parameters of the controlled object.

[0015] Furthermore, in the improved projection algorithm, the historical The variables composed of the input and output data of the order are used as the variables of the estimator module, where It indicates the order of information of historical input and output data used by the improved projection algorithm compared with the projection algorithm.

[0016] Furthermore, the controller module is used to calculate the control signal based on the pole configuration method according to the received estimation result of the estimator, the historical output signal of the controlled object, and the tracking reference signal.

[0017] Furthermore, the event trigger condition of the event trigger module is: to determine whether the absolute value of the difference between the currently received control signal and the control signal at the last event trigger exceeds the set trigger threshold; if the absolute value of the difference between the current control signal and the control signal at the last event trigger exceeds the set trigger threshold, the trigger condition is met; otherwise, the trigger condition is not met; when the event trigger condition is met, the event trigger sends a flag signal 1 to the communication network module; when the event trigger condition is not met, the event trigger sends a flag signal 0 to the communication network module.

[0018] Furthermore, the actuator module acts according to the control signal transmitted by the communication network module; the actuator acts only when a new control signal is received, otherwise the current state of the actuator remains unchanged.

[0019] An event-triggered control method with high-order unmodeled dynamics is applied to the above-mentioned event-triggered control system with high-order unmodeled dynamics; the method comprises the following steps:

[0020] Step 1: Measure the output signal of the controlled object through the sensor module and send it to the estimator module;

[0021] Step 2: The estimator module calculates parameter estimation values ​​based on the data transmitted by the sensor and sends them to the controller module;

[0022] Step 3: The controller module calculates the control signal according to the parameter estimation value and sends it to the event trigger module;

[0023] Step 4: The event trigger module determines whether to trigger an event based on the current control signal and the control signal at the last event trigger, and sends the determination result to the communication network module in the form of a flag signal;

[0024] Step 5: The communication network module determines whether to open the communication network from the controller module to the actuator module according to the received event trigger flag signal;

[0025] Step 6: The actuator module takes action or maintains the current state according to the received control signal.

[0026] Furthermore, in step 4, the event trigger module saves the output value of the controller module at the last trigger, and determines whether the absolute value of the difference between the current output value of the controller module and the output value of the controller at the last trigger time recorded by the event trigger module is greater than or equal to the set trigger threshold; if so, an event trigger flag signal 1 is sent to the communication network module; if not, an event trigger flag signal 0 is sent to the communication network.

[0027] Furthermore, in step 5, when the communication network module receives an event trigger flag signal of 1, the communication network is opened to allow the controller module to send data to the actuator module; when the communication network module receives an event trigger flag signal of 0, the communication network is closed and the controller module does not send data to the actuator module.

[0028] The beneficial effects of the present invention are as follows: the present invention uses an event trigger mechanism to reduce the frequency of data transmission from the controller to the actuator, saving communication energy while avoiding excessive action of the actuator to increase the service life of the actuator; an estimator is designed to estimate the unknown parameters of the system; a robust adaptive controller is designed to ensure the stability of the bounded input and bounded output of the closed-loop system. The control method proposed in the present invention is suitable for systems with high-order unmodeled dynamics, and can still ensure the global boundedness of the system when the upper bound of the unmodeled dynamics is unknown, and has good application value. The present invention designs the controller based on the pole configuration method, relaxing the constraint that the controlled object must be a minimum phase system. In addition, by introducing an event trigger mechanism to act on the communication network, the frequency of signal transmission from the controller to the actuator is reduced, while effectively avoiding excessive action of the actuator, thereby increasing the service life of the actuator. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is an overall block diagram of the system of the present invention;

[0030] Figure 2 It is a simulation result diagram of the control signal in the present invention;

[0031] Figure 3 is a simulation result diagram of the output signal of the controlled object in the present invention;

[0032] Figure 4 It is a simulation result diagram of the event triggering interval time in the present invention. DETAILED DESCRIPTION

[0033] Further description is given below in conjunction with the accompanying drawings.

[0034] See also Figure 1 The present invention provides an event-triggered control system with high-order unmodeled dynamics, comprising a controlled object 1, a sensor module 2, an estimator module 3, a controller module 4, a communication network module 5, an event trigger module 6, and an actuator module 7; wherein:

[0035] The sensor module is used to measure the output signal of the controlled object in real time and send it to the estimator module;

[0036] An estimator module is used to estimate unknown parameters of the controlled object according to input and output signals of the controlled object, and send the parameter estimation results to the controller module;

[0037] A controller module, used for calculating a control signal according to the received estimator data and sending the control signal to the event trigger module;

[0038] An event trigger is used to determine whether to trigger an event according to the received control signal, and send the determination result to the communication network module; when the event triggering condition is met, the event trigger sends a flag signal 1 to the communication network module, and when the event triggering condition is not met, the event trigger sends a flag signal 0 to the communication network module;

[0039] The communication network module is used to determine whether to open the communication network according to the received event trigger judgment result (that is, the event trigger flag signal is 1 or 0). When the communication network is opened, the controller module sends data to the actuator module, otherwise the controller module does not send data to the actuator module;

[0040] The actuator module is used to perform actions on the controlled object according to the received control signal. The actuator takes action only when a new control signal is received, otherwise the actuator maintains the current state unchanged.

[0041] Figure 1 Herein, the communication network refers to a wireless communication network.

[0042] Specifically, the controlled object is an n+N order discrete uncertain system, in which the parameters of the N-order high-order dynamics are unknown; the parameters of the n-order low-order dynamics are unknown, but the set of parameters of the n-order low-order dynamics is known; at the same time, the system is subject to unknown time-varying interference, and the upper bound of the interference is unknown.

[0043] The estimator module estimates the unknown parameters of the controlled object based on an improved projection algorithm according to the received historical output signal of the controlled object, the historical input signal (historical actuator signal) and the set of parameters of the n-order low-order dynamic part of the controlled object.

[0044] The controller module calculates the control signal based on the pole configuration method according to the received estimator estimation result, actuator signal, historical output signal of the controlled object and tracking reference signal.

[0045] The event trigger module determines whether to trigger an event based on the received control signal. The event trigger condition is: determine whether the absolute value of the difference between the currently received control signal and the control signal at the last event trigger exceeds the set trigger threshold; if the absolute value of the difference between the current control signal and the control signal at the last event trigger exceeds the set trigger threshold, the trigger condition is met; otherwise, it is considered that the trigger condition is not met; when the event trigger condition is met, the event trigger sends a flag signal 1 to the communication network module, and records the current control signal as the control signal at the latest event trigger time. When the event trigger condition is not met, the event trigger sends a flag signal 0 to the communication network module, and does not update the recorded control signal at the event trigger time.

[0046] The communication network module decides whether to open the communication network according to the received event trigger flag signal. When the received event trigger flag signal is 1, the communication network is opened, and when the received event trigger flag signal is 0, the communication network is closed.

[0047] The actuator module acts according to the control signal transmitted by the communication network module. The actuator acts only when a new control signal is received, otherwise the actuator maintains the current state unchanged.

[0048] Specifically, the controlled object can be described by the following difference equation:

[0049]

[0050] in, represents discrete moments, Indicates the current output signal. represents all unknown high-order and low-order parameters, ; is a vector consisting of the historical input and output data of the controlled object. represents the output signal at the i-step time before the current time k, represents the actuator signal applied to the controlled object at the i-step time before the current time k (i.e., the historical input signal of the controlled object), represents the low-order dynamic order, represents the higher-order dynamic order, Represents unknown time-varying external disturbance.

[0051] The low-order dynamics of the controlled object are as follows:

[0052]

[0053] in, , Represents the low-order dynamic corresponding parameter vector, which is unknown, but the set it belongs to known; , , , .

[0054] The present invention provides an event-triggered control method with high-order unmodeled dynamics, which is applied to the above-mentioned event-triggered control system with high-order unmodeled dynamics; the method comprises the following steps:

[0055] Step 1: The sensor module measures the output signal of the system (controlled object) in real time and sent to the estimator module.

[0056] Step 2: The estimator module calculates the parameter estimates based on the data transmitted by the sensor and sends them to the controller module.

[0057] Specifically, the estimator module receives the historical output signal of the controlled object. , actuator output signal , the accused The set of low-order dynamic part parameters The unknown parameters of the controlled object are estimated based on the improved projection algorithm, which is different from the existing projection method using historical The present invention uses historical input and output data as variables. Variables consisting of input and output data As a variable of the estimator module to improve the robustness of the estimation algorithm.

[0058] The parameter estimates are as follows:

[0059]

[0060] in, , Representation History Variables consisting of input and output data of order, , It indicates the order of information of historical input and output data used in the improved projection algorithm compared with the existing projection algorithm. Represents the projection operator, which is used to ensure that the parameter estimates are always in the set within, that is , , parameter estimation results Sent to the controller module.

[0061] Step 3: The controller module calculates the control signal based on the parameter estimates as follows (pole placement method) , and sent to the event trigger module.

[0062] Step 3.1: Based on the received estimator module data Calculating controller parameters and .

[0063] The received estimator module output data , define the polynomial , ,in Represents a shift operator.

[0064] Specified by the user based on convergence performance requirements Order 1 polynomial .

[0065] Define polynomial , ,in Represents a shift operator.

[0066] According to the following formula:

[0067]

[0068] Evaluating Polynomials and Parameters and .

[0069] Step 3.2: According to the controller parameters , Tracking reference signal , calculate the control signal .

[0070] According to the controller parameters , Defining vectors .

[0071] The control signal is calculated according to the following formula :

[0072]

[0073] in , Represents the control signal at the i-step time before the current time k.

[0074] Step 4: The event trigger module determines whether to trigger an event based on the current control signal and the control signal at the last event trigger, and sends the determination result to the communication network module in the form of a flag signal.

[0075] Specifically, the event trigger module saves the output value of the controller module at the last trigger, and determines whether the absolute value of the difference between the current output value of the controller module and the output value of the controller at the last trigger time recorded by the event trigger module is greater than or equal to the set trigger threshold. If so, an event trigger flag signal 1 is sent to the communication network module; if not, an event trigger flag signal 0 is sent to the communication network module.

[0076] In order to reduce the frequency of signal transmission from the controller module to the actuator module, reduce communication energy consumption, and reduce the actuator action frequency, the present invention designs an event trigger module that follows the following event trigger mechanism to act on the communication network.

[0077]

[0078] in, Indicates The time when the event is triggered, Indicates The time when the event is triggered, represents the set of positive integers, represents the infimum of a set, It is the controller signal received by the event trigger module in real time. It is the event trigger threshold parameter set by the user. It is the data value sent by the controller module to the sensor module at the last event triggering moment and is also the signal actually received by the actuator module, recorded as:

[0079]

[0080] The event trigger module receives the real-time output value of the controller module , and determine whether the trigger condition is met If the event trigger condition is met, the event trigger flag value is set to 1 and the flag value is sent to the communication network module. The event trigger module records the output value of the current controller. for ; If the event trigger condition is not met, the event trigger flag value is set to 0 and the flag value is sent to the controller module without updating .

[0081] Step 5: The communication network module decides whether to open the communication network from the controller module to the actuator module according to the received event trigger flag signal. Specifically, when the communication network module receives an event trigger flag signal of 1, the communication network is opened, allowing the controller module to send data to the actuator module; when the communication network module receives an event trigger flag signal of 0, the communication network is closed, and the controller module does not send data to the actuator module.

[0082] Step 6: The actuator module takes action or maintains the current state according to the received control signal.

[0083] The actuator module receives the control signal Perform actions and apply them to the controlled object.

[0084] The controlled object used in the simulation follows the following differential equation:

[0085] ,

[0086] in , , , low-order dynamic order =2, high-order dynamic order .

[0087] Tracking reference signal .

[0088] The parameters of the event-triggered robust adaptive control algorithm are set as follows: , , .

[0089] The simulation results are as follows Figure 2 , Figure 3 and Figure 4 shown. Figure 2 For control signal ; Figure 3 The output signal of the controlled object , it can be seen that under the action of the event-triggered robust adaptive control algorithm proposed in this invention, the tracking error is bounded; Figure 4 The interval time for event triggering It can be seen that under the event triggering mechanism proposed in the present invention, the frequency of information transmission from the controller module to the actuator module is significantly reduced.

[0090] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

[0091] The above embodiments are only used to illustrate the design ideas and features of the present invention, and their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. The protection scope of the present invention is not limited to the above embodiments. Therefore, any equivalent changes or modifications made based on the principles and design ideas disclosed by the present invention are within the protection scope of the present invention.

Claims

1. An event-triggered control system with high-order unmodeled dynamics, characterized in that include: The sensor module is used to measure the output signal of the controlled object and send it to the estimator module; An estimator module is used to estimate unknown parameters of the controlled object according to the input and output signals of the controlled object, and send the parameter estimation results to the controller module; A controller module, used for calculating a control signal according to the received estimator data and sending the control signal to the event trigger module; An event trigger module, used to determine whether to trigger an event according to the received control signal, and when the event trigger condition is met, the event trigger sends a flag signal 1 to the communication network module, and when the event trigger condition is not met, the event trigger sends a flag signal 0 to the communication network module; The communication network module is used to determine whether to open the communication network according to the received event trigger flag signal 1 or 0; when the communication network is opened, the controller module sends data to the actuator module, otherwise the controller module does not send data to the actuator module; An actuator module is used to perform actions on a controlled object according to a received control signal; The controlled object is an n+N order discrete uncertainty system, in which the parameters of the N order high order dynamics are unknown; the parameters of the n order low order dynamics are unknown, but the set of the parameters of the n order low order dynamics is known; at the same time, the system is subject to unknown time-varying interference, and the upper bound of the interference is unknown; The estimator module estimates unknown parameters of the controlled object based on an improved projection algorithm according to the received historical output signal of the controlled object, the historical input signal, and the set of n-order low-order dynamic parameters of the controlled object; In the improved projection algorithm, the historical The variables composed of the input and output data of the order are used as the variables of the estimator module, where Indicates that the improved projection algorithm uses more information about the historical input and output data than the projection algorithm. , n represents the low-order dynamic order, and N represents the high-order dynamic order; The controller module calculates the control signal based on the pole configuration method according to the received estimation result of the estimator, the historical output signal of the controlled object, and the tracking reference signal.

2. The event-triggered control system with high-order unmodeled dynamics according to claim 1, characterized in that: The event trigger condition of the event trigger module is: judging whether the absolute value of the difference between the currently received control signal and the control signal at the last event triggering exceeds the set trigger threshold; if the absolute value of the difference between the current control signal and the control signal at the last event triggering exceeds the set trigger threshold, the trigger condition is met; Otherwise the trigger condition is not met; when the event trigger condition is met, the event trigger sends a flag signal 1 to the communication network module; when the event trigger condition is not met, the event trigger sends a flag signal 0 to the communication network module.

3. The event-triggered control system with high-order unmodeled dynamics according to claim 1, characterized in that: The actuator module acts according to the control signal transmitted by the communication network module; the actuator acts only when a new control signal is received, otherwise the current state of the actuator remains unchanged.

4. An event-triggered control method with high-order unmodeled dynamics, characterized in that Applicable to the event-triggered control system with high-order unmodeled dynamics as claimed in claim 1; the method comprising the following steps: Step 1: Measure the output signal of the controlled object through the sensor module and send it to the estimator module; Step 2: The estimator module calculates parameter estimation values ​​based on the data transmitted by the sensor and sends them to the controller module; Step 3: The controller module calculates the control signal according to the parameter estimation value and sends it to the event trigger module; Step 4: The event trigger module determines whether to trigger an event based on the current control signal and the control signal at the last event trigger, and sends the determination result to the communication network module in the form of a flag signal; Step 5: The communication network module determines whether to open the communication network from the controller module to the actuator module according to the received event trigger flag signal; Step 6: The actuator module takes action or maintains the current state according to the received control signal.

5. The event-triggered control method with high-order unmodeled dynamics according to claim 4, characterized in that: In step 4, the event trigger module saves the output value of the controller module at the last trigger, and determines whether the absolute value of the difference between the current output value of the controller module and the output value of the controller at the last trigger time recorded by the event trigger module is greater than or equal to the set trigger threshold; if so, an event trigger flag signal 1 is sent to the communication network module; if not, an event trigger flag signal 0 is sent to the communication network.

6. The event-triggered control method with high-order unmodeled dynamics according to claim 4, characterized in that: In step 5, when the communication network module receives an event trigger flag signal of 1, the communication network is opened to allow the controller module to send data to the actuator module; When the communication network module receives an event trigger flag signal of 0, the communication network is closed and the controller module does not send data to the actuator module.

Images