CN101995870A - Forward channel random network time-delay compensation method for network cascade control system - Google Patents

Abstract

The invention provides a forward channel random network time-delay compensation method for a network cascade control system, and belongs to the technical field of network control systems. The real network data transmission process between forward network channel nodes is adopted to replace a network time-delay compensation model therebetween, the measurement, the observation, estimation or identification of network data transmission random time delay between the forward network channel nodes is avoided, and the requirement on node clock signal synchronization is avoided. The method can reduce the influence of the random network time delay on the system stability and improve the dynamic performance quality of the system. The compensation method is applied to dynamic compensation and control of the random network time delay in a forward channel of the network cascade control system in a network with certain data packet loss and known mathematical models of master/slave controlled objects.

Description

The compensation method of network cascade control system through path randomness network delay

Technical field

The present invention relates to the compensation method of network cascade control system through path randomness network delay, belong to the network control system technical field.

Background technology

In dcs, sensor and controller, the close-loop feedback control system that constitutes by the real-time Communication for Power network between controller and the actuator is called network control system (Networked control systems, NCS). network control system is compared with the control system of traditional point-to-point structure, it is low to have cost, be easy to information sharing, being easy to expansion safeguards, advantages such as dirigibility is big, therefore be widely used in the industrial control process in recent years. still, because the load-bearing capacity of network service bandwidth is limited, the transmission of network data exists network delay inevitably. and the existence of time delay can reduce the control performance quality of system, even cause system's instability, also give the analysis of control system simultaneously, design has brought very big difficulty.

At present, research about network control system mainly is at single-circuit control system both at home and abroad, constant at network delay respectively, in time, become or at random, network delay is less than a sampling period or greater than a sampling period, single bag transmission or many bag transmission, have or not under the various conditions such as data-bag lost, it is carried out modeling and stability analysis, but rarely have paper that the network cascade control system is studied. control loop is called network cascade control system (NCCS) by the cascade control system of real-time network closure, and the typical structure block diagram that is applicable to network cascade control system of the present invention is as shown in Figure 1.

Because the network cascade control system is the network control system of closed loop more than, analysis to network delay influence is more complex more than single-circuit network control system with the research of system performance. and the inner looping network delay will have a strong impact on the rapidity and the antijamming capability of inner looping network control system, and the while also will have a negative impact with stability and the controlling performance of external loop network delay to whole network cascade control system.

Difficult point for network delay research mainly is:

(1) because network delay is relevant with factors such as network topology structure, communication protocol, offered load, the network bandwidth and packet sizes. to network delay greater than several and even dozens of sampling period, set up accurately predict, the mathematical model of estimation or identification, almost be impossible at present.

(2) occur in the previous node network delay in node transmitting network data process backward, in previous node, no matter adopt which kind of prediction or method of estimation, all can not know the exact value of the network delay that produces thereafter in advance in advance. time delay causes system performance to descend even causes system's instability, brings difficulty also for simultaneously the analysis and the design of control system.

(3) will satisfy in the network cascade control system, all node clock signals in different distributions place are unpractical fully synchronously.

Existing only in main transformer at network send (control) device node and secondary the change to send between (control) device node (outer feedforward network), and the secondary network cascade control system as shown in Figure 2 of giving (interior feedforward network) between (control) device node and the actuator node that becomes, its input R (s) and output Y ₁(s) closed loop between is transmitted letter

$\frac{Y_1\left(s\right)}{R\left(s\right)}=\frac{C_1\left(s\right){e^{-{\mathrm{\τ}}_{1}s}C}_2\left(s\right)e^{-{\mathrm{\τ}}_{2}s}{G}_2\left(s\right)G_1\left(s\right)}{1+C_1\left(s\right){e^{-{\mathrm{\τ}}_{1}s}C}_2\left(s\right)e^{-{\mathrm{\τ}}_{2}s}{G}_2\left(s\right)G_1\left(s\right)+C_2\left(s\right)e^{-{\mathrm{\τ}}_{2}s}{G}_2\left(s\right)}---\left(1\right)$

In the formula: C ₁(s) be master controller, C ₂(s) be submaster controller; G ₁(s) be main controlled device, G ₂(s) be secondary controlled device; τ ₁Expression is sent (control) device node to be transferred to secondary the change from main transformer network data and is sent the randomness network delay that is produced between (control) device node; τ ₂Expression becomes network data from pair send (control) device node to be transferred to the randomness network delay that the actuator node is produced.

Owing to comprise network delay τ in the denominator of the closed loop transfer function, shown in the equation (1) ₁And τ ₂Exponential term

With

The existence of time delay will worsen the control performance quality of system, even cause system's loss of stability, when serious system be broken down.

Reduce time delay to the sex key of system stability, just be to realize to send (control) device node and pair to become the network delay τ that send between (control) device node main transformer ₁Exponential term

And the pair change send (control) device node to the network delay τ between the actuator node ₂Exponential term From the denominator of equation (1), remove, promptly realize not comprising in system's closed loop secular equation the exponential term of all-network time delay, yet and then realization is to the compensation of network delay., realize compensation to network delay, at first must know the size of network delay. at present, the method that adopts usually is by to network delay τ both at home and abroad ₁And τ ₂Measurement, come delay compensation τ ₁And τ ₂Influence to system stability. still, because to network delay τ ₁And τ ₂Accurate measurement need satisfy the synchronous requirement of node clock signal, if adopt hardware to realize that node clock signal is synchronous fully, then need bigger economy input; If adopt software position signal, when then between node, transmitting, may meet with the influence of network delay owing to correction signal, be difficult to realize that nodal clock is synchronous fully; If adopt size to network delay is estimated, observation, identification or forecast method obtain network delay, then at first must know the accurate probability distribution of network delay, or mathematical model accurately, but because the size of network delay is relevant with concrete factors such as procotol, offered load size and network topology structure, to estimation, observation, the identification of network delay or predict and all may have deviation.

Therefore, how to exempt the requirement synchronous to node clock signal, release is to estimation, observation, identification or the prediction of network delay between the node, can obtain simultaneously between the node time delay value accurately again, and then realize compensation and control to network cascade control system through path randomness network delay, having become in the research of network cascade control system needs one of key issue that solves.

Summary of the invention

In order to solve the problems of the technologies described above, the invention provides a kind of network that relates to and exist only in main transformer and send (control) device node and secondary the change to send between (control) device node (outer feedforward network), and the network cascade control system randomness network delay compensation method of (interior feedforward network) between (control) device node and the actuator node is sent in secondary change.

Purpose of the present invention:

In the network cascade control system, the difficult problem of network delay " indeterminacy ", the present invention proposes a kind of release requirement synchronous to node clock signal, also exempt simultaneously delay compensation method, realize the segmentation of network delay, real-time, online and dynamic compensation and control to measurement, estimation or the identification of randomness network delay between its node.

The method that the present invention adopts is:

The first step: adopt main transformer to send (control) device node to send the live network data transmission procedure between (control) device node to replace the compensation model of network delay therebetween to secondary the change, thereby the system that structurally realizes does not comprise the compensation model of network delay therebetween. no matter send (control) device node how complicatedly and uncertain to have to the secondary network path that send (control) device node that becomes from main transformer, also no matter include therebetween what routers or (with) intermediate link, the network delay that information flow experienced is exactly a real network delay in the control procedure, has just realized the compensate function to its time delay in the information stream transmission process.

Second step: adopt secondary the change to send (control) device node to replace the compensation model of network delay therebetween to the live network data transmission procedure between the actuator node, thereby the system that structurally realizes does not comprise the compensation model of network delay therebetween. how complicated and uncertain no matter send (control) device node to have to the network path the actuator node from the pair change, also no matter include therebetween what routers or (with) intermediate link, the network delay that information flow experienced is exactly a real network delay in the control procedure, has just realized the compensate function to its time delay in the information stream transmission process.

The 3rd step: at network cascade control system shown in Figure 2, the network delay collocation structure of enforcement the inventive method is as shown in Figure 3.

In Fig. 3, from the input R (s) and the output Y of system ₁(s) closed loop transfer function, between is

$\frac{Y_1\left(s\right)}{R\left(s\right)}=\frac{\frac{C_1\left(s\right)e^{-{\mathrm{\τ}}_{1}s}{C}_2\left(s\right)e^{-{\mathrm{\τ}}_{2}s}{G}_2\left(s\right)G_1\left(s\right)}{1+C_2\left(s\right)G_{2m}\left(s\right)}}{1+Q\left(s\right)+P\left(s\right)+\frac{C_1\left(s\right)C_{2m}\left(s\right)G_{2m}\left(s\right)G_{1m}\left(s\right)}{1+C_{2m}\left(s\right)G_{2m}\left(s\right)}}---\left(2\right)$

Wherein:

$Q\left(s\right)=\frac{C_1\left(s\right)e^{-{\mathrm{\τ}}_{1}s}{C}_2\left(s\right)e^{-{\mathrm{\τ}}_{2}s}{G}_2\left(s\right)[G_1\left(s\right)-G_{1m}\left(s\right)]}{1+C_2\left(s\right)G_{2m}\left(s\right)}---\left(3\right)$

$P\left(s\right)=\frac{C_2\left(s\right)e^{-{\mathrm{\τ}}_{2}s}[G_2\left(s\right)-G_{2m}\left(s\right)]}{1+C_2\left(s\right)G_{2m}\left(s\right)}---\left(4\right)$

When master's (pair) controlled device prediction model equals its true model, i.e. G _1m(s)=G ₁(s), G _2m(s)=G ₂(s), submaster controller satisfies C _2m(s)=C ₂(s) time, formula (2) but abbreviation be

$\frac{Y_1\left(s\right)}{R\left(s\right)}=\frac{C_1\left(s\right)e^{-{\mathrm{\τ}}_{1}s}{C}_2\left(s\right)e^{-{\mathrm{\τ}}_{2}s}{G}_2\left(s\right)G_1\left(s\right)}{1+C_2\left(s\right)G_2\left(s\right)+C_1\left(s\right)C_2\left(s\right)G_2\left(s\right)G_1\left(s\right)}---\left(5\right)$

In the closed loop transfer function, denominator of the network cascade control system shown in the formula (5), do not comprise network delay τ ₁And τ ₂Exponential term

With Promptly realized closed loop secular equation 1+C ₂(s) G ₂(s)+C ₁(s) C ₂(s) G ₂(s) G ₁(s)=0 do not comprise the exponential term of network delay in, thereby eliminated the influence of network delay, improved the control performance quality of system, realized compensate function the randomness network delay to system stability.

The scope of application of the present invention:

The present invention is applicable to that main (pair) controlled device mathematical model is known, network delay can be greater than 1 and even dozens of sampling period, network exists only in main transformer and send (control) device node and secondary the change to send between (control) device node (outer feedforward network), and secondary compensation and the control that becomes the network cascade control system randomness network delay that send (interior feedforward network) between (control) device node and the actuator node.

The invention is characterized in that this method may further comprise the steps:

1, when sending (control) device node, main transformer is sampled cycle h ₁During triggering, will adopt mode A to carry out work;

2, send (control) device node with control signal u when main transformer ₁(s) become when sending (control) device node to transmit to pair by outer feedforward network path, will adopt mode B to carry out work;

3, when becoming, pair send (control) device node to be sampled cycle h ₂During triggering, will adopt mode C to carry out work;

4, when becoming, pair send (control) device node controlled signal u ₁When (s) triggering, will adopt mode D to carry out work;

5, when becoming, pair send (control) device node with control signal u ₂(s) pass through interior feedforward network path when the actuator node transmits, will adopt mode E to carry out work;

6, as actuator node controlled signal u ₂When (s) triggering, will adopt mode F to carry out work.

The step of mode A comprises:

A1: main transformer send (control) device node to work in the time type of drive, and it triggers the sampling period is h ₁

A2: after main transformer send (control) device node to be triggered, to main controlled device G ₁(s) output signal Y ₁(s) and the prediction model G of main controlled device _1m(s) output signal y _1m(s) sample;

A3: to Y ₁(s) and y _1m(s) implement additive operation, obtain model bias signal w ₁(s);

A4: with given signal R of system (s) and w ₁(s) and m ₁(s) implement additive operation, obtain external loop systematic error signal e ₁(s);

A5: to e ₁(s) implement master controller C ₁(s) computing, controlled signal u ₁(s);

A6: with u ₁(s), act on the predictive algorithm C of submaster controller as given signal _2m(s) with the prediction model G of secondary controlled device _2m(s) backfeed loop of Gou Chenging, it is output as y _Cg2m(s);

A7: with y _Cg2m(s) act on G _2m(s) and G _1m(s), it is output as m ₁(s).

The step of mode B comprises:

B1: main transformer send (control) device node with control signal u ₁(s), send the transmission of (control) device node by outer feedforward network path to the pair change.

The step of mode C comprises:

C1: the secondary change send (control) device node to work in the time type of drive, and it triggers the sampling period is h ₂

C2: after secondary change send (control) device node to be triggered, to secondary controlled device G ₂(s) output signal Y ₂(s) and the prediction model G of secondary controlled device _2m(s) output signal y _2m(s) sample;

C3: to Y ₂(s) and y _2m(s) implement additive operation, obtain model bias signal w ₂(s).

The step of mode D comprises:

D1: drive signal u ₁(s) trigger secondary the change and send (control) device node, the pair of this moment becomes send (control) device node to work in event driven manner;

D2: send in (control) device node, with u in the pair change ₁(s) deduct model bias signal w ₂(s) and prediction model G _2m(s) output signal y _G2m(s), obtain inner looping systematic error signal e ₂(s);

D3: at e ₂(s), implement submaster controller algorithm C ₂(s), the control signal of its node output is u ₂(s).

The step of mode E comprises:

E1: the secondary change send (control) device node with control signal u ₂(s), transmit to the actuator node by interior feedforward network path.

The step of mode F comprises:

F1: the actuator node works in event driven manner;

F2: actuator node controlled signal u ₂(s) trigger;

F3: with u ₂(s) as drive signal, to secondary controlled device G ₂(s) implement control; G ₂(s) output signal is Y ₂(s), to main controlled device G ₁(s) implement control;

F4: in the actuator node, finish G simultaneously _2m(s) discreet value y _2m(s) calculating.

7, delay compensation method of the present invention is characterized in that it is by the embedded master controller C of main transmitter that main transformer send (control) device node ₁(s) form, the shared same node of promptly main transmitter and master controller, main transformer send (control) device node employing time to drive triggered mode of operation (sampling period is h ₁).

8, delay compensation method of the present invention is characterized in that it is to be embedded in submaster controller C by secondary transmitter that (control) device node is sent in secondary change ₂(s) form the shared same node of promptly secondary transmitter and submaster controller. the secondary transmitter employing time drives triggered mode of operation, and (sampling period is h ₂), (trigger pip is u and submaster controller adopts the event-driven triggered mode of operation ₁(s)).

9, delay compensation method of the present invention, it is characterized in that system comprises main transformer and send (control) device node, the secondary change to send unit such as (control) device node, actuator node, main controlled device and secondary controlled device, each unit carries out work according to the working method of setting separately.

10, delay compensation method of the present invention, it is characterized in that sending the network data transmission process of external loop feedforward network path (control) device node to replace network delay compensation model therebetween, thereby the system that structurally realizes does not comprise the compensation model of network delay therebetween with sending (control) device node to become from main transformer really to pair.

11, delay compensation method of the present invention, it is characterized in that sending (control) device node to replace network delay compensation model therebetween, thereby the system that structurally realizes does not comprise the compensation model of network delay therebetween to the network data transmission process of inner looping feedforward network path the actuator node with becoming from pair really.

12, delay compensation method of the present invention is characterized in that exempting send (control) device node to send measurement, observation, estimation or the identification of network delay (control) device node to secondary change from main transformer from structure.

13, delay compensation method of the present invention is characterized in that exempting pair become from structure sending measurement, observation, estimation or the identification of (control) device node to network delay the actuator node.

14, delay compensation method of the present invention is characterized in that sending (control) device node, the secondary change to send (control) device node and the synchronous requirement of actuator node clock signal from the structure release to main transformer.

15, delay compensation method of the present invention is characterized in that realizing from structure the enforcement and concrete control strategy C of network delay compensation method ₁(s) and C ₂(s) selection is irrelevant.

16, delay compensation method of the present invention is characterized in that realizing that from structure the enforcement of network delay compensation method is irrelevant with the selection of concrete network communication protocol.

17, delay compensation method of the present invention is characterized in that as major-minor controlled device G ₁(s) and G ₂(s) with its prediction model G _1m(s) and G _2m(s) equate, and the submaster controller prediction model satisfies C _2m(s) equal C ₂(s) time, can realize full remuneration, improve the control performance quality of system network cascade control system through path randomness network delay.

18, delay compensation method of the present invention, what it is characterized in that adopting is the compensation method that " soft " changes the control system structure, need not increases any hardware device again, and the software resource that utilizes existing network cascade control system intelligent node to carry just is enough to realize its compensate function.

19, delay compensation method of the present invention, the mode A of it is characterized in that are applicable to that main transformer send (control) device node periodic sampling and signal is handled.

20, delay compensation method of the present invention, the mode B of it is characterized in that are applicable to that main transformer send (control) device node transmitting network data.

21, delay compensation method of the present invention, the mode C of it is characterized in that are applicable to that secondary the change send (control) device node periodic sampling and signal is handled.

22, delay compensation method of the present invention, the mode D of it is characterized in that are applicable to that secondary the change send (control) device node enforcement control algolithm.

23, delay compensation method of the present invention, the mode E of it is characterized in that are applicable to that secondary the change send (control) device node transmitting network data.

24, delay compensation method of the present invention, the mode F of it is characterized in that are applicable to that the enforcement of actuator node drives and control function.

The present invention has following advantage:

1, send (control) device node to send (control) device node (external loop through path) owing to having exempted to main transformer to secondary the change from structure, and secondary the change sent the measurement of (control) device node to (inner looping through path) randomness network delay between the actuator node, observation, estimate or identification, also exempted the synchronous requirement of node clock signal simultaneously, and then avoided the inaccurate evaluated error that causes of time delay estimation model, avoided the required waste that expends the node storage resources of time delay identification, also avoided simultaneously because the compensating error that " the empty sampling " that time delay causes or " many samplings " bring.

2, owing to realized with the selection of concrete network communication protocol irrelevantly from structure, thereby both be applicable to the network cascade control system that adopts wired network protocol, also be applicable to the network cascade control system of wireless network protocol; Both be applicable to the deterministic network agreement, also be applicable to the procotol of uncertainty.

3, owing to realized with the selection of concrete network communication protocol irrelevant from structure, thereby both be applicable to heterogeneous network cascade control system based on wired network protocol, also be applicable to heterogeneous network cascade control system, also be applicable to the delay compensation of the network cascade control system of heterogeneous (as wired and wireless mixing) simultaneously based on wireless network protocol.

4, owing to realized with the selection of concrete master (or secondary) controller control strategy irrelevant from structure, thereby both can be used for adopting the network cascade control system of conventional control, also can be used for adopting Based Intelligent Control or adopt the network cascade control system of complicated control strategy.

5, because the present invention adopts is the compensation method that " soft " changes the control system structure, thereby in its implementation procedure, need not to increase again any hardware device, the software resource that utilizes existing network cascade control system intelligent node to carry, just be enough to realize its compensate function, thereby can save hardware investment, be easy to be extended and applied.

Description of drawings

Fig. 1 exists only in network cascade control system block scheme in the through path for network.

Fig. 2 exists only in network cascade control system structural drawing in the through path for network.

Fig. 3 is a network cascade control system through path randomness network delay compensation method structural drawing of the present invention

In the block diagram of Fig. 1, system is by input signal (R), and main transformer send (control) device (S₁/C ₁) node, outer feedforward network, the secondary change sent (control) device (S₂/C ₂) node, interior feedforward network, actuator (A) node, secondary controlled device (G₂) and output (Y₂), and main controlled device (G₁) and output (Y₁) form.

Built-in main controller in the main transmitter, namely main transmitter and master controller share same node (S₁/C ₁). node adopts the time type of drive to carry out work, and the triggering cycle is h₁, to the sampling of main controlled device implementation cycle, and to deviation signal enforcement C₁Control.

Built-in submaster controller in the secondary transmitter, namely secondary transmitter and submaster controller share same node (S₂/C ₂). wherein: secondary transmitter adopts the time type of drive to carry out work, and the triggering cycle is h₂, the secondary controlled device implementation cycle is sampled; And submaster controller adopts event driven manner to carry out work, send (control) device (S by main transformer₁/C ₁) output signal of node triggers by outer feedforward network.

Actuator (A) is an isolated node, adopts event driven manner to carry out work, is become by pair and send (control) device (S₂/C ₂) output signal of node triggers by interior feedforward network, and drive executing agency, thus secondary controlled device (G changed₂) state, and then change main controlled device (G₁) state.

The main transformer of system send (control) device (S among Fig. 1₁/C ₁) node, the secondary change sent (control) device (S₂/C ₂) node and actuator (A) node all be intelligent node, not only possess storage computing and communication function, but also possess software configuration and control function, these nodes comprise now the hardware such as intelligent node common in the industrial field bus control system (FCS) of extensive use and the Distributed Control System (DCS) or smart machine.

In the system of Fig. 2, the randomness network transfer delay in the transfer of data has significant impact for Systems balanth and control performance quality. and the transfer of data of network cascade control system is experiencing from main transformer send (control) device node to the secondary network data transmission time delay τ that send (external loop through path) (control) device node that becomes₁, and secondary the change sent the network data transmission time delay τ of (control) device node to (inner looping through path) between the actuator node₂Impact. time delay is relevant with the concrete factors such as procotol, offered load size and network topology structure, yet for the measurement of network delay or estimate or observation or identification have become the crucial precondition that realizes its compensation., the distributivity of each node by network connection is so that each node in the network cascade control system is difficult to satisfy the requirement of clock signal synchronization. simultaneously, because the randomness of network delay and sudden will accomplish that each step can both Accurate Prediction be impossible.

In the system of Fig. 3, do not comprise from main transformer and send (control) device node to the secondary network delay prediction model that send (external loop through path) (control) device node that becomes, do not comprise from the pair change yet and send the network delay prediction model of (control) device node to (inner looping through path) the actuator node. exempted randomness network delay τ₁And τ₂Measurement, estimation, observation or identification, also exempted simultaneously the requirement to node (main transformer send (control) device, secondary change to send (control) device, actuator) clock signal synchronization. when the major-minor controlled device equates with its prediction model, can realize the output signal Y from the input signal R (s) of system to system₁(s) in the closed loop transfer function,, with network delay τ₁And τ₂Exponential termWithFrom denominator, eliminate, namely realize not comprising in the closed loop characteristic equation network delay τ₁And τ₂Exponential term, thereby reduced the impact of time delay to the stability of a system, improved the control performance quality of system, realize compensation and control to the randomness network delay.

Embodiment

To make clearer above-mentioned and other feature and advantage of the present invention of those of ordinary skill in the art by describing exemplary embodiment of the present invention in detail below with reference to accompanying drawing 3.

Concrete implementation step is as described below:

The first step: the main transformer that works in the time type of drive send (control) device node to main controlled device G ₁(s) output signal Y ₁(s) and its prediction model G _1m(s) output signal y _1m(s) (sampling period is h to carry out periodic sampling ₁), and to Y ₁(s) and y _1m(s) implement additive operation, obtain model bias signal w ₁(s);

Second step: the pair that works in the time type of drive becomes send (control) device node to secondary controlled device G ₂(s) output signal Y ₂(s) and its prediction model G _2m(s) output signal y _2m(s) (sampling period is h to carry out periodic sampling ₂), and to Y ₂(s) and y _2m(s) implement additive operation, obtain model bias signal w ₂(s);

The 3rd step: main transformer send (control) device node according to the given signal R of system (s), to w ₁(s) with intranodal compensating unit output signal m ₁(s) subtract each other, obtain error signal e ₁(s), to e ₁(s) implement C ₁(s) control, its output signal is u ₁(s): on the one hand, with u ₁(s) as the prediction model C of this intranodal by submaster controller _2m(s) and the prediction model G of secondary controlled device _2m(s) the given input signal of the negative feedback loop of Gou Chenging; On the other hand, with u ₁(s) (control) device node middle controller C is sent in change as pair ₂(s) given;

The 4th the step: by external loop feedforward network path with u ₁(s) be transferred to secondary the change and send (control) device node, and it is in running order to trigger its node;

The 5th step: the secondary controller C that send in (control) device node that becomes ₂(s) by u ₁(s) after the triggering, according to the given signal u of system ₁(s), to w ₂(s) with the secondary controlled device prediction model G of intranodal _2m(s) output signal y _G2m(s) subtract each other, obtain error signal e ₂(s), to e ₂(s) implement C ₂(s) control, its control output signal is u ₂(s): on the one hand, with u ₂(s) as G _2m(s) input signal; On the other hand, with u ₂(s) as secondary controlled device G ₂(s) control signal;

The 6th the step: by interior feedforward network path with u ₂(s) transmit to the actuator node;

The 7th step: the on-the-spot actuator node that works in event driven manner is by u ₂(s) signal triggering, and drive topworks, thus change secondary controlled device G ₂(s) state, and then change main controlled device G ₁(s) state.

The 8th step: return the first step.

The above only is preferred embodiment of the present invention, and is in order to restriction the present invention, within the spirit and principles in the present invention not all, any modification of being done, is equal to replacement, improvement etc., all should be included within protection scope of the present invention.

The content that is not described in detail in this instructions belongs to this area professional and technical personnel's known prior art.

Claims (16)

1. network cascade control system through path randomness network delay compensation method is characterized in that this method may further comprise the steps:

(1). when sending (control) device node, main transformer is sampled cycle h ₁During triggering, will adopt mode A to carry out work;

(2). when main transformer send (control) device node with control signal u ₁(s) become when sending (control) device node to transmit to pair by outer feedforward network path, will adopt mode B to carry out work;

(3). when becoming, pair send (control) device node to be sampled cycle h ₂During triggering, will adopt mode C to carry out work;

(4). when becoming, pair send (control) device node controlled signal u ₁When (s) triggering, will adopt mode D to carry out work;

(5). when becoming, pair send (control) device node with control signal u ₂(s) pass through interior feedforward network path when the actuator node transmits, will adopt mode E to carry out work;

(6). as actuator node controlled signal u ₂When (s) triggering, will adopt mode F to carry out work.

2. compensation method according to claim 1 is characterized in that the step of described mode A comprises:

A1: main transformer send (control) device node to work in the time type of drive, and it triggers the sampling period is h ₁

A2: after main transformer send (control) device node to be triggered, to main controlled device G ₁(s) output signal Y ₁(s) and the prediction model G of main controlled device _1m(s) output signal y _1m(s) sample;

A3: to Y ₁(s) and y _1m(s) implement additive operation, obtain model bias signal w ₁(s);

A4: with given signal R of system (s) and w ₁(s) and m ₁(s) implement additive operation, obtain external loop systematic error signal e ₁(s);

A5: to e ₁(s) implement master controller C ₁(s) computing, controlled signal u ₁(s);

A6: with u ₁(s), act on the predictive algorithm C of submaster controller as given signal _2m(s) with the prediction model G of secondary controlled device _2m(s) backfeed loop of Gou Chenging, it is output as y _Cg2m(s);

A7: with y _Cg2m(s) act on G _2m(s) and G _1m(s), it is output as m ₁(s).

3. compensation method according to claim 1 is characterized in that the step of described mode B comprises:

B1: main transformer send (control) device node with control signal u ₁(s), send the transmission of (control) device node by outer feedforward network path to the pair change.

4. compensation method according to claim 1 is characterized in that the step of described mode C comprises:

C1: the secondary change send (control) device node to work in the time type of drive, and it triggers the sampling period is h ₂

C2: after secondary change send (control) device node to be triggered, to secondary controlled device G ₂(s) output signal Y ₂(s) and the prediction model G of secondary controlled device _2m(s) output signal y _2m(s) sample;

C3: to Y ₂(s) and y _2m(s) implement additive operation, obtain model bias signal w ₂(s).

5. compensation method according to claim 1 is characterized in that the step of described mode D comprises:

D1: drive signal u ₁(s) trigger secondary the change and send (control) device node, the pair of this moment becomes send (control) device node to work in event driven manner;

D2: send in (control) device node, with u in the pair change ₁(s) deduct model bias signal w ₂(s) and prediction model G _2m(s) output signal y _G2m(s), obtain inner looping systematic error signal e ₂(s);

D3: at e ₂(s), implement submaster controller algorithm C ₂(s), the control signal of its node output is u ₂(s).

6. compensation method according to claim 1 is characterized in that the step of described mode E comprises:

E1: the secondary change send (control) device node with control signal u ₂(s), transmit to the actuator node by interior feedforward network path.

7. compensation method according to claim 1 is characterized in that the step of described mode F comprises:

F1: the actuator node works in event driven manner;

F2: actuator node controlled signal u ₂(s) trigger;

F3: with u ₂(s) as drive signal, to secondary controlled device G ₂(s) implement control; G ₂(s) output signal is Y ₂(s), to main controlled device G ₁(s) implement control;

F4: in the actuator node, finish G simultaneously _2m(s) discreet value y _2m(s) calculating.

8. method according to claim 1 is characterized in that it is by the embedded master controller C of main transmitter that main transformer send (control) device node ₁(s) form the shared same node of promptly main transmitter and master controller; Main transformer send (control) device node employing time to drive triggered mode of operation, and (sampling period is h ₁).

9. method according to claim 1 is characterized in that it is to be embedded in submaster controller C by secondary transmitter that (control) device node is sent in secondary change ₂(s) form the shared same node of promptly secondary transmitter and submaster controller; The secondary transmitter employing time drives triggered mode of operation, and (sampling period is h ₂), (trigger pip is u and submaster controller adopts the event-driven triggered mode of operation ₁(s)).

10. method according to claim 1, it is characterized in that system comprises main transformer and send (control) device node, the secondary change to send unit such as (control) device node, actuator node, main controlled device and secondary controlled device, each unit carries out work according to the working method of setting separately.

11. method according to claim 1, it is characterized in that with sending (control) device node to send the network data transmission process of external loop feedforward network path (control) device node to replace network delay compensation model therebetween from main transformer really to secondary the change, and send (control) device node to replace network delay compensation model therebetween to the network data transmission process of inner looping feedforward network path the actuator node with becoming from pair really, thereby the system that structurally realizes does not comprise the compensation model of network delay.

12. method according to claim 1, it is characterized in that sending (control) device node to secondary the change send (control) device node from main transformer, and send measurement, observation, estimation or the identification of (control) device node to network delay the actuator node to becoming from pair from the structure release; Release send (control) device node, the secondary change to send (control) device node and the synchronous requirement of actuator node clock signal to main transformer.

13. method according to claim 1 is characterized in that realizing from structure the enforcement and concrete control strategy C of network delay compensation method ₁(s) and C ₂(s) selection is irrelevant, and is irrelevant with the selection of concrete network communication protocol.

14. method according to claim 1 is characterized in that as major-minor controlled device G ₁(s) and G ₂(s) with its prediction model G _1m(s) and G _2m(s) equate, and the submaster controller prediction model satisfies C _2m(s) equal C ₂(s) time, can realize full remuneration, improve the control performance quality of system network cascade control system through path randomness network delay.

15. method according to claim 1, what it is characterized in that adopting is the compensation method that " soft " changes the control system structure, need not increases any hardware device again, and the software resource that utilizes existing network cascade control system intelligent node to carry just is enough to realize its compensate function.

16. method according to claim 1, the mode A of it is characterized in that is applicable to that main transformer send (control) device node periodic sampling and signal is handled; Mode B is applicable to that main transformer send (control) device node transmitting network data; Mode C is applicable to that secondary the change send (control) device node periodic sampling and signal is handled; Mode D is applicable to that secondary the change send (control) device node enforcement control algolithm; Mode E is applicable to that secondary the change send (control) device node transmitting network data; Mode F is applicable to that the enforcement of actuator node drives and control function.

Images