CN102842909A - Method for controlling power electronic hybrid system - Google Patents

Method for controlling power electronic hybrid system Download PDF

Info

Publication number
CN102842909A
CN102842909A CN2012103360743A CN201210336074A CN102842909A CN 102842909 A CN102842909 A CN 102842909A CN 2012103360743 A CN2012103360743 A CN 2012103360743A CN 201210336074 A CN201210336074 A CN 201210336074A CN 102842909 A CN102842909 A CN 102842909A
Authority
CN
China
Prior art keywords
power filter
power
controllers
active power
control
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN2012103360743A
Other languages
Chinese (zh)
Other versions
CN102842909B (en
Inventor
帅智康
涂春鸣
盘宏斌
姚鹏
蒋玲
戴晓宗
楚烺
肖凡
张杨
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hunan University
Original Assignee
Hunan University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hunan University filed Critical Hunan University
Priority to CN201210336074.3A priority Critical patent/CN102842909B/en
Publication of CN102842909A publication Critical patent/CN102842909A/en
Application granted granted Critical
Publication of CN102842909B publication Critical patent/CN102842909B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/20—Active power filtering [APF]

Abstract

The invention discloses a method for controlling a power electronic hybrid system. The method comprises a power electronic hybrid control system comprising an active power filter and a static var compensator, wherein the static var compensator comprises a passive power filter group and a thyristor control reactor (TCR); and the active power filter, the passive power filter group and the TCR access between a power grid and loads connected with the power grid in sequence. The passive power filter and the active power filter are used jointly to perform harmonic suppression on high-voltage busbars; and the static var compensator is used to perform reactive compensation, so that delay compensation and online control are realized, control accuracy is enhanced, tracking performance is improved, electric energy loss of the power grid is reduced, and pollution of the power grid is purified.

Description

A kind of power electronics hybrid system control method
Technical field
The present invention relates to power electronic system, particularly a kind of power electronics hybrid system control method. 
Background technology
With continuing to develop for Power Electronic Technique, network load is more and more in nonlinear user and reactive load, such as rolling mill, electric arc furnaces, electric railway.Increasing so that harmonic components are increasingly severe in power network for this kind of load, causes electric network element loss increase, and influence equipment is normally run, and causes power network particular resonance to cause major accident what is more；And the increase of load or burden without work can cause the problems such as Voltage Drop, loss increase.Therefore have pure, high-quality a power grid environment and power supply quality be both power network running department simultaneously and load user institute jointly it is desired.But current product is most of also to have many deficiencies： 
1st, Active Power Filter-APF and SVC SVC are single use, and function is single, it is impossible to while realizing the dual-use function of filtering and reactive-load compensation. 
2nd, the common control method of Active Power Filter-APF has PI controls, Hysteresis control etc.；Control accuracy is poor when being controlled using PI, is easily vibrated when compensation capacity is low and PI parameters are relatively difficult to adjust；The setting of ring width is larger to systematic influence during Hysteresis control, and the frequency requirement of ring width hour switch tube is also high, and it not is quite varied to cause to apply. 
3rd, control accuracy itself is not just high using PI controllers for Active Power Filter-APF, once occurring load changing, when harmonic wave situation changes, filter effect is especially undesirable, it is impossible to realize On-line Control. 
4th, SVC SVC uses PI controllers, once reactive load changes, compensation effect is undesirable, can not realize On-line Control. 
5th, general first-harmonic resonance injection active power filter, its first-harmonic resonance injection branch is when applied to high pressure, and first-harmonic resonance part will bear very high voltage, and capacity is larger, and cost is too high during engineer applied. 
The content of the invention
The technical problems to be solved by the invention are, in view of the shortcomings of the prior art, providing a kind of power electronics hybrid system control method, realize compensation of delay and On-line Control, improve control accuracy, and improve tracking performance, reduce the electric energy loss of power network, purify electric network pollution. 
In order to solve the above technical problems, the technical solution adopted in the present invention is：A kind of power electronics hybrid system control method, including power electronics hybrid control system, power electronics hybrid control system includes Active Power Filter-APF, SVC, the SVC includes passive power filter group and thyristor-controlled reactor, Active Power Filter-APF, passive power filter group, thyristor-controlled reactor successively and are accessed between power network and the load being connected with power network, characterized in that, the power electronics hybrid system control method includes Control Method of Active Power Filter and SVC control method： 
The Control Method of Active Power Filter comprises the following steps：
1）Detect load current、
Figure 2012103360743100002DEST_PATH_IMAGE004
、
Figure 2012103360743100002DEST_PATH_IMAGE006
, Active Power Filter-APF is actually implanted into the electric current in three phase network
Figure 2012103360743100002DEST_PATH_IMAGE008
、
Figure 2012103360743100002DEST_PATH_IMAGE010
、
Figure 2012103360743100002DEST_PATH_IMAGE012
, computational load harmonic current
Figure 2012103360743100002DEST_PATH_IMAGE014
、
Figure 2012103360743100002DEST_PATH_IMAGE016
、；
2）Utilize
Figure 2012103360743100002DEST_PATH_IMAGE020
- Smith Compensator is carried out to system
Figure 2012103360743100002DEST_PATH_IMAGE022
Compensation of delay so that
Figure 466276DEST_PATH_IMAGE008
、
Figure 836077DEST_PATH_IMAGE010
、
Figure 664356DEST_PATH_IMAGE012
With
Figure 461411DEST_PATH_IMAGE014
、、
Figure 243739DEST_PATH_IMAGE018
Between phase difference be；
3）Load harmonic current is adjusted using PI controllers
Figure 192104DEST_PATH_IMAGE014
、
Figure 97743DEST_PATH_IMAGE016
、
Figure 193875DEST_PATH_IMAGE018
The electric current in three phase network is actually implanted into Active Power Filter-APF
Figure 905479DEST_PATH_IMAGE008
、、
Figure 345742DEST_PATH_IMAGE012
Between error, with triangle wave produce Active Power Filter-APF in IGBT drive signal；
4）The electric current of signal and Active Power Filter-APF injection three phase network after PI is adjusted obtains predicted value as the input of prediction neural network
Figure 2012103360743100002DEST_PATH_IMAGE026
；
Figure 2012103360743100002DEST_PATH_IMAGE028
Represent that s-i moment wave filters are injected into the electric current in power network,
Figure 866853DEST_PATH_IMAGE026
Represent the output that neural network prediction is obtained；
5）According to load harmonic current、
Figure 101843DEST_PATH_IMAGE016
、
Figure 411601DEST_PATH_IMAGE018
The electric current being actually implanted into Active Power Filter-APF
Figure 420008DEST_PATH_IMAGE008
、
Figure 473415DEST_PATH_IMAGE010
、
Figure 313195DEST_PATH_IMAGE012
Between error
Figure 2012103360743100002DEST_PATH_IMAGE030
, PI regulation after signal
Figure 2012103360743100002DEST_PATH_IMAGE032
Variable quantity and neural network prediction output, using wide area PREDICTIVE CONTROL criterion, calculate the updated value of PI controller parameters
Figure 2012103360743100002DEST_PATH_IMAGE034
、
Figure 2012103360743100002DEST_PATH_IMAGE036
, realize the on-line control of PI controllers；
The SVC control method comprises the following steps：
1）Detect load current
Figure 466057DEST_PATH_IMAGE002
、
Figure 323154DEST_PATH_IMAGE004
、
Figure 17441DEST_PATH_IMAGE006
And three-phase power grid voltage、
Figure 2012103360743100002DEST_PATH_IMAGE040
、
Figure 2012103360743100002DEST_PATH_IMAGE042
, calculate the reactive current of three phase network
Figure 2012103360743100002DEST_PATH_IMAGE044
、
Figure 2012103360743100002DEST_PATH_IMAGE046
、
Figure 2012103360743100002DEST_PATH_IMAGE048
；
2）The rated voltage that load normal work is adjusted with PI controllers is set as reference voltage, reference voltage
Figure 482051DEST_PATH_IMAGE050
With the three-phase power grid voltage detected
Figure 527368DEST_PATH_IMAGE038
、
Figure 358795DEST_PATH_IMAGE040
、
Figure 907588DEST_PATH_IMAGE042
Between error adjusted with PI controllers, after linearized link, produce the control signal of IGCT in thyristor-controlled reactor；
3）Signal after PI is adjustedWith the three-phase power grid voltage detected
Figure 262663DEST_PATH_IMAGE038
、
Figure 399247DEST_PATH_IMAGE040
、
Figure 802546DEST_PATH_IMAGE042
As the input of prediction neural network, the three-phase power grid voltage output valve predicted；
4）Signal after being adjusted according to PI
Figure 999172DEST_PATH_IMAGE032
Variable quantity, reference voltageWith the output of neural network prediction, using wide area PREDICTIVE CONTROL criterion, the updated value of PI controllers is calculated、, realize the on-line control of PI controllers.
The present invention operation principle be：The blended electric power electronics hybrid system is by injection active power filter, passive power filter group PPF, thyristor-controlled reactor TCR composition.Passive filter group assume responsibility for main harmonics and filter out, and injection active power filter carries out the suppression of remaining harmonic wave, and PPF and TCR constitute reactive-load compensation of the SVC SVC progress from perception to capacitive.The wave filter of the blended electric power electronic system passes through
Figure 381426DEST_PATH_IMAGE020
- Smith Compensator carries out compensation of delay, and by the output of neural network prediction Active Power Filter-APF, wide area PREDICTIVE CONTROL criterion calculates the updated value control parameter of PI controllers, realizes On-line Control；Its SVC calculates the updated value of PI controllers, realizes On-line Control by neural network prediction line voltage, wide area PREDICTIVE CONTROL criterion. 
Compared with prior art, the advantageous effect of present invention is that：The power electronic system of the present invention is mixed with wave filter and reactive-load compensator, realizes harmonics restraint and the dual-use function of reactive-load compensation；The Active Power Filter-APF of the blended electric power electronic system is used
Figure 357472DEST_PATH_IMAGE020
- Smith Compensator structure is filtered compensation of delay, and filter effect is preferable；The Active Power Filter-APF of the blended electric power electronic system is predicted output using neutral net, and online modification PI parameters are realized by wide area PREDICTIVE CONTROL criterion, and control accuracy is greatly promoted, and realizes accurate filtering；The SVC SVC of the blended electric power electronic system is predicted line voltage using neutral net, and online modification PI parameters are realized by wide area PREDICTIVE CONTROL criterion, and control accuracy is greatly promoted, and realizes accurate compensation. 
Brief description of the drawings
Fig. 1 is one embodiment of the invention power electronics hybrid system structural representation； 
Fig. 2 is one embodiment of the invention injection active filter system control principle drawing；
Fig. 3 is one embodiment of the invention reactive-load compensator system control principle drawing；
Fig. 4 is that one embodiment of the invention prediction neural network trains schematic diagram；
Fig. 5 is one embodiment of the invention prediction neural network structure chart；
Fig. 6 is that one embodiment of the invention PI parameters update flow chart.
Embodiment
As shown in Figure 1-Figure 3, power electronics hybrid control system of the present invention includes Active Power Filter-APF, SVC, the SVC includes passive power filter group and thyristor-controlled reactor, Active Power Filter-APF, passive power filter group, thyristor-controlled reactor successively and are accessed between power network and the load being connected with power network, and SVC SVC is made up of thyristor-controlled reactor TCR and passive power filter PPF.Passive power filter PPF is made up of several groups of single tuned filters.Injection active power filter is made up of injection branch and Active Power Filter-APF, and wherein injection branch includes two groups of single tuned filters, one group of high-pass filter and partial pressure inductance.This injection branch, which both meets compensation electric current, can smoothly inject power network, and the cost of injection branch is reduced again.Because under 35Kv high pressures, using first-harmonic resonance it is pouring-in when its injection branch fundamental wave capacity it is very big, cost is very high.Injection active power filter by can the PI controllers of online modification parameter be adjusted, IGBT control signals are formed after triangular modulation, the output of neural network prediction Active Power Filter-APF, the on-line amending of PI parameters is carried out by wide area PREDICTIVE CONTROL criterion, accurate control is realized, is passed through simultaneously
Figure 898175DEST_PATH_IMAGE020
- Smith prediction carries out compensation of delay.SVC SVC is adjusted, neural network prediction by voltage close loop control using PI controllers, the parameter of wide area PREDICTIVE CONTROL criterion online modification PI controllers. 
Power electronics hybrid system control method, including the control to Active Power Filter-APF and the control to SVC： 
Wherein to the control of Active Power Filter-APF, including specifically include following steps：
1）Detect load current
Figure 977864DEST_PATH_IMAGE002
、、
Figure 979636DEST_PATH_IMAGE006
, injection active power filter is actually implanted into the electric current in power network
Figure 58450DEST_PATH_IMAGE008
、
Figure 556427DEST_PATH_IMAGE010
、
Figure 178DEST_PATH_IMAGE012
, computational load harmonic current
Figure 216396DEST_PATH_IMAGE014
、
Figure 98901DEST_PATH_IMAGE016
、
Figure 451385DEST_PATH_IMAGE018
。
2）Control targe is to realize
Figure 331617DEST_PATH_IMAGE008
、
Figure 769551DEST_PATH_IMAGE010
、=-
Figure 98956DEST_PATH_IMAGE014
、-
Figure 946826DEST_PATH_IMAGE016
、-
Figure 872057DEST_PATH_IMAGE018
, i.e., the load harmonic current that the compensation electric current of injection active power filter output is obtained with detecting is equal in magnitude and in opposite direction, and system has delay in itself
Figure 2012103360743100002DEST_PATH_IMAGE052
, then pass through
Figure 34048DEST_PATH_IMAGE020
- Smith prediction is carried out to system
Figure 298807DEST_PATH_IMAGE022
Compensation of delay so that
Figure 583158DEST_PATH_IMAGE008
、
Figure 933368DEST_PATH_IMAGE010
、
Figure 695788DEST_PATH_IMAGE012
With、
Figure 332622DEST_PATH_IMAGE016
、Between phase difference be
Figure 228915DEST_PATH_IMAGE024
, it is not necessary to the harmonic wave that detection is obtained is carried out reversely, so as to reach the function of compensation of delay in instruction current. 
3）Error between the electric current that load-side harmonic wave and Active Power Filter-APF are actually implanted into is adjusted using PI, and the drive signal of IGBT in wave filter is produced with triangle wave. 
4）Neutral net carries out Model Distinguish, and its identification process is the training for carrying out weights modification as foundation using the error that neutral net is exported and sampling is obtained between wave filter injection power network current, and identification process is completed untill error is in the range of setting.The input of neutral net has unit time delay process, and the electric current of signal and wave filter injection power network after PI regulations obtains predicted value as the input of prediction neural network
Figure 265004DEST_PATH_IMAGE026
。 
5）Load harmonic
Figure 2012103360743100002DEST_PATH_IMAGE054
The electric current being actually implanted into wave filter
Figure 2012103360743100002DEST_PATH_IMAGE056
Between error
Figure 828840DEST_PATH_IMAGE030
, PI regulation after signal
Figure 215959DEST_PATH_IMAGE032
Variable quantity and neural network prediction outputAs the input of wide area PREDICTIVE CONTROL criterion, the updated value of the parameter of PI controllers is calculated
Figure 414040DEST_PATH_IMAGE034
、, realize the on-line control of PI controllers. 
SVC SVC control method key steps include： 
1）Detect load current、
Figure 931106DEST_PATH_IMAGE004
、
Figure 941787DEST_PATH_IMAGE006
And line voltage
Figure 909743DEST_PATH_IMAGE038
、
Figure 442093DEST_PATH_IMAGE040
、
Figure 888118DEST_PATH_IMAGE042
, calculate reactive current in power network
Figure 956568DEST_PATH_IMAGE044
、
Figure 95425DEST_PATH_IMAGE046
、。
2）The rated voltage of user load normal work is set as reference voltage
Figure 928569DEST_PATH_IMAGE050
, reference voltage and the error detected between line voltage are adjusted using PI, after linearized link, form the control signal of IGCT. 
3）Signal after PI regulations
Figure 851526DEST_PATH_IMAGE032
With input of the line voltage detected as prediction neural network, the line voltage output valve predicted. 
4）Signal after being adjusted according to PI
Figure 161285DEST_PATH_IMAGE032
Variable quantity, reference voltage
Figure 232009DEST_PATH_IMAGE050
With the output of neural network prediction, using wide area PREDICTIVE CONTROL criterion, the updated value of PI controllers is calculated
Figure 223099DEST_PATH_IMAGE034
、
Figure 62879DEST_PATH_IMAGE036
, realize the on-line control of PI controllers. 
Referring to Fig. 3, the realization of compensation of delay function：The principle of filtering is exactly to detect that harmonic wave present in load-side makes the null process of power network current by injection and its electric current in opposite direction equal in magnitude.
Figure 2012103360743100002DEST_PATH_IMAGE058
, i.e.,
Figure 2012103360743100002DEST_PATH_IMAGE060
,Represent power network current,
Figure 2012103360743100002DEST_PATH_IMAGE064
Load-side harmonic current is represented,
Figure 2012103360743100002DEST_PATH_IMAGE066
Represent that Active Power Filter-APF is injected into the electric current in power network.But due to being the presence of delay in system, if its total time delay process is
Figure 2012103360743100002DEST_PATH_IMAGE068
If, one time link of compensated in advance
Figure 2012103360743100002DEST_PATH_IMAGE070
, then always delay will become, i.e.,
Figure 2012103360743100002DEST_PATH_IMAGE074
, realize
Figure 855123DEST_PATH_IMAGE066
With
Figure 147564DEST_PATH_IMAGE064
Between it is reverse.Implement step as follows： 
1）First assume
Figure 2012103360743100002DEST_PATH_IMAGE076
、
Figure 2012103360743100002DEST_PATH_IMAGE078
All it is that delay is not present, controller is designed with this situation.ThenClosed loop transfer function, it is as follows
Figure DEST_PATH_IMAGE082
Figure 72849DEST_PATH_IMAGE080
Represent the transmission function of PI controllers；
Figure 724411DEST_PATH_IMAGE076
Represent the transmission function of inverter；
Figure 769727DEST_PATH_IMAGE078
Represent the transmission function that electric current is output to from inverter between injection power network；
2）Our target is that design obtains controller, makeContainWithTime delay process
Figure 677137DEST_PATH_IMAGE068
, it is such to closed loop transfer function, be
Figure DEST_PATH_IMAGE088
I.e.
Solve
Figure DEST_PATH_IMAGE092
Wherein
Figure DEST_PATH_IMAGE094
3）It can be seen from analysis above,
Figure 881854DEST_PATH_IMAGE066
With
Figure 257252DEST_PATH_IMAGE064
Between it is in opposite direction, i.e., phase difference is therebetween
Figure 926131DEST_PATH_IMAGE020
, this means that
Figure DEST_PATH_IMAGE096
Figure 185074DEST_PATH_IMAGE068
=
Figure 611507DEST_PATH_IMAGE096
Figure DEST_PATH_IMAGE098
4）According to the circuit analysis to model, the transmission function of closed-loop system is
Figure DEST_PATH_IMAGE102
It can be seen that active power filter system eliminates the influence of delay by the closed loop transfer function, and realize control targe
Figure 551781DEST_PATH_IMAGE060
, this process is also
Figure 75167DEST_PATH_IMAGE020
- Smith predicts process.
The neural network prediction model of Active Power Filter-APF and SVC, wide area PREDICTIVE CONTROL realizes that the process of On-line Control is identical, therefore concentrates explanation herein. 
Referring to Fig. 4, the Model Distinguish process of neutral net is the training of neutral net, until output error stops within the scope of setting.For wave filter be by PI adjust after signal
Figure DEST_PATH_IMAGE104
With the reality output of wave filter
Figure DEST_PATH_IMAGE106
As input, for SVC be by PI adjust after signal
Figure 177115DEST_PATH_IMAGE104
The foundation all changed with actual electric network voltage as input, two models by the error between predicted value and actual value as weights, weights modification is carried out using Levenberg-Marquardt gradient descent algorithms.This process is constantly repeated, terminates training after the scope of setting until error is reached, as forecast model. 
Referring to Fig. 5, the forecast model of neutral net uses the three layer feedforward neural networks with time delay.Wherein the activation primitive of hidden layer is hyperbolic tangent function, and the activation primitive of output layer is linear activation primitive. 
Figure DEST_PATH_IMAGE108
The output activation primitive of hidden layer is represented,
Figure DEST_PATH_IMAGE114
、
Figure DEST_PATH_IMAGE116
Hidden layer is represented respectively to the weights and deviation of output layer,
Figure DEST_PATH_IMAGE118
、
Figure DEST_PATH_IMAGE120
Input layer is represented to the weights of hidden layer,
Figure DEST_PATH_IMAGE122
Input layer is represented to the deviation of hidden layer,Represent the summation of the product of the corresponding weights of all inputs. 
Referring to Fig. 2, Fig. 3, the on-line control realization to PI is reached by wide area PREDICTIVE CONTROL criterion.Traditional its parameters of PI controllers are constant when being, but are all nonlinear systems in Practical Project, and being usually required PI parameters is time-varying and can realize online modification in whole control process.To realize this application of function wide-area control criterion GPC criterion.It is as follows that it shifts process onto： 
1）Traditional PI controllers can be written as：
Figure DEST_PATH_IMAGE126
Wherein
Figure DEST_PATH_IMAGE128
,
Figure DEST_PATH_IMAGE130
Scaling multiple and integration time constant are represented respectively.
The Discrete PI control of incremental form can be expressed as： 
Wherein
Figure DEST_PATH_IMAGE134
,
Figure DEST_PATH_IMAGE136
,
Figure DEST_PATH_IMAGE138
2）The object function for the wide-area control criterion that this control system is applied is as follows：
Figure DEST_PATH_IMAGE140
Wherein,
Figure DEST_PATH_IMAGE142
：Minimum prediction time domain length
Figure DEST_PATH_IMAGE144
：Maximum predicted time domain length
Figure DEST_PATH_IMAGE146
：Control time domain length
：(s+k) expected response at moment
Figure DEST_PATH_IMAGE150
：(s+k) output of moment neural network prediction
Figure DEST_PATH_IMAGE152
：Minimize object function J control variable
Figure DEST_PATH_IMAGE154
：Weight
3）From analysis above;
Figure 979592DEST_PATH_IMAGE134
, it is contemplated that
Figure DEST_PATH_IMAGE158
For vector, order
Figure DEST_PATH_IMAGE160
, just obtain
Figure DEST_PATH_IMAGE162
, that is, obtain
Figure DEST_PATH_IMAGE164
,
Figure 894197DEST_PATH_IMAGE130
Updated value.Which achieves the On-line Control ability of PI controllers.
Referring to Fig. 6, whole calculating process can be described below： 
The first step：The parameter of PI controllers is determined using Ziegler-Nichols setting methods
Figure 272088DEST_PATH_IMAGE164
,。
Second step：If s=1, the output of neutral net is determined.S represents current time. 
3rd step：The error between desired value and predicted value is calculated, if error is zero to be maintained for adjuster
Figure 336176DEST_PATH_IMAGE164
,Parameter constant；If error is not zero, the 4th step is carried out.Desired value is the harmonic signal that is compensated the need for calculating and without function signal.Predicted value is the output valve of neural network prediction. 
4th step：Utilize, it is determined that
Figure 116230DEST_PATH_IMAGE158
。
5th step：WillIt is converted into the parameter of PI controllers. 
6th step：S=s+1 is made, the 3rd step is jumped to. 

Claims (5)

1. a kind of power electronics hybrid system control method, including power electronics hybrid control system, power electronics hybrid control system includes Active Power Filter-APF, SVC, the SVC includes passive power filter group and thyristor-controlled reactor, Active Power Filter-APF, passive power filter group, thyristor-controlled reactor successively and is accessed between power network and the load being connected with power network, it is characterized in that, the power electronics hybrid system control method includes Control Method of Active Power Filter and SVC control method：
The Control Method of Active Power Filter comprises the following steps：
Detect load current
Figure 2012103360743100001DEST_PATH_IMAGE002
、
Figure 2012103360743100001DEST_PATH_IMAGE004
、
Figure 2012103360743100001DEST_PATH_IMAGE006
, Active Power Filter-APF is actually implanted into the electric current in three phase network
Figure 2012103360743100001DEST_PATH_IMAGE008
、
Figure 2012103360743100001DEST_PATH_IMAGE010
、
Figure 2012103360743100001DEST_PATH_IMAGE012
, computational load harmonic current
Figure 2012103360743100001DEST_PATH_IMAGE014
、、
Figure 2012103360743100001DEST_PATH_IMAGE018
；
Utilize
Figure 2012103360743100001DEST_PATH_IMAGE020
- Smith Compensator is carried out to system
Figure DEST_PATH_IMAGE022
Compensation of delay so that
Figure 926861DEST_PATH_IMAGE008
、
Figure 781685DEST_PATH_IMAGE010
、
Figure 561422DEST_PATH_IMAGE012
With
Figure 956631DEST_PATH_IMAGE014
、
Figure 708686DEST_PATH_IMAGE016
、
Figure 531149DEST_PATH_IMAGE018
Between phase difference be
Figure DEST_PATH_IMAGE024
；
Load harmonic current is adjusted using PI controllers
Figure 234400DEST_PATH_IMAGE014
、、
Figure 102179DEST_PATH_IMAGE018
The electric current in three phase network is actually implanted into Active Power Filter-APF
Figure 33226DEST_PATH_IMAGE008
、
Figure 787556DEST_PATH_IMAGE010
、
Figure 524567DEST_PATH_IMAGE012
Between error, with triangle wave produce Active Power Filter-APF in IGBT drive signal；
The electric current of signal and Active Power Filter-APF injection three phase network after PI is adjusted obtains predicted value as the input of prediction neural network； 
Figure DEST_PATH_IMAGE028
Represent that s-i moment wave filters are injected into the electric current in power network,
Figure 985636DEST_PATH_IMAGE026
Represent the output that neural network prediction is obtained；
According to load harmonic current
Figure 353163DEST_PATH_IMAGE014
、
Figure 594789DEST_PATH_IMAGE016
、
Figure 869912DEST_PATH_IMAGE018
The electric current being actually implanted into Active Power Filter-APF
Figure 513383DEST_PATH_IMAGE008
、
Figure 284768DEST_PATH_IMAGE010
、Between error, PI regulation after signal
Figure DEST_PATH_IMAGE032
Variable quantity and neural network prediction output
Figure 967870DEST_PATH_IMAGE026
, using wide area PREDICTIVE CONTROL criterion, calculate the updated value of PI controller parameters
Figure DEST_PATH_IMAGE034
、
Figure DEST_PATH_IMAGE036
, realize the on-line control of PI controllers；
The SVC control method comprises the following steps：
Detect load current
Figure 403531DEST_PATH_IMAGE002
、
Figure 175177DEST_PATH_IMAGE004
、
Figure 125816DEST_PATH_IMAGE006
And three-phase power grid voltage
Figure DEST_PATH_IMAGE038
、
Figure DEST_PATH_IMAGE040
、
Figure DEST_PATH_IMAGE042
, calculate the reactive current of three phase network
Figure DEST_PATH_IMAGE044
、
Figure DEST_PATH_IMAGE046
、
Figure DEST_PATH_IMAGE048
；
The rated voltage of load normal work is set as reference voltage, reference voltage
Figure 319906DEST_PATH_IMAGE050
With the three-phase power grid voltage detected
Figure 610073DEST_PATH_IMAGE038
、、Between error adjusted with PI controllers, after linearized link, produce the control signal of IGCT in thyristor-controlled reactor；
Signal after PI is adjusted
Figure 411173DEST_PATH_IMAGE032
With the three-phase power grid voltage detected
Figure 325820DEST_PATH_IMAGE038
、
Figure 439269DEST_PATH_IMAGE040
、
Figure 302183DEST_PATH_IMAGE042
As the input of prediction neural network, the three-phase power grid voltage output valve predicted；
Signal after being adjusted according to PI
Figure 526491DEST_PATH_IMAGE032
Variable quantity, reference voltage
Figure 587988DEST_PATH_IMAGE050
With the output of neural network prediction, using wide area PREDICTIVE CONTROL criterion, the updated value of PI controllers is calculated
Figure 872339DEST_PATH_IMAGE034
、
Figure 222549DEST_PATH_IMAGE036
, realize the on-line control of PI controllers.
2. power electronics hybrid system control method according to claim 1, it is characterised in that the Active Power Filter-APF is injection active power filter.
3. power electronics hybrid system control method according to claim 1, it is characterised in that the object function of the wide area PREDICTIVE CONTROL criterion is：
,
Wherein,
Figure DEST_PATH_IMAGE054
Time domain length is predicted for minimum,
Figure DEST_PATH_IMAGE056
For maximum predicted time domain length,
Figure DEST_PATH_IMAGE058
To control time domain length,
Figure DEST_PATH_IMAGE060
For (s+k) moment
Figure DEST_PATH_IMAGE062
Expected response,
Figure DEST_PATH_IMAGE064
For the output of (s+k) moment neural network prediction,
Figure DEST_PATH_IMAGE066
For J control variable,
Figure DEST_PATH_IMAGE068
For weight.
4. power electronics hybrid system control method according to claim 1, it is characterised in that the passive power filter group is made up of some groups of single tuned filter connections.
5. power electronics hybrid system control method according to claim 1, it is characterised in that the updated value of the calculating PI controllers
Figure 562132DEST_PATH_IMAGE034
、
Figure 743715DEST_PATH_IMAGE036
Concretely comprise the following steps：
1）：The parameter of PI controllers is determined using Ziegler-Nichols setting methods
Figure DEST_PATH_IMAGE070
,
Figure DEST_PATH_IMAGE072
；
2）：If s=1, the output of neutral net is determined；Wherein s represents current time；
3）：The error between desired value and predicted value is calculated, if error is zero to be maintained for PI controllers
Figure 871071DEST_PATH_IMAGE070
,
Figure 708577DEST_PATH_IMAGE072
Parameter constant；If error is not zero, into 4）；Its expected value is the harmonic signal and idle letter compensated the need for calculating；Predicted value is the output valve of neural network prediction；
4）：Utilize, it is determined that；Wherein
Figure DEST_PATH_IMAGE078
,
Figure DEST_PATH_IMAGE080
；J is the object function of wide-area control criterion；
5）：Will
Figure 710906DEST_PATH_IMAGE076
It is converted into the parameter of PI controllers
Figure 950257DEST_PATH_IMAGE034
、
Figure 310831DEST_PATH_IMAGE036
；
6）：S=s+1 is made, 3 are jumped to）.
CN201210336074.3A 2012-09-12 2012-09-12 Method for controlling power electronic hybrid system Expired - Fee Related CN102842909B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210336074.3A CN102842909B (en) 2012-09-12 2012-09-12 Method for controlling power electronic hybrid system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210336074.3A CN102842909B (en) 2012-09-12 2012-09-12 Method for controlling power electronic hybrid system

Publications (2)

Publication Number Publication Date
CN102842909A true CN102842909A (en) 2012-12-26
CN102842909B CN102842909B (en) 2015-07-08

Family

ID=47370044

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210336074.3A Expired - Fee Related CN102842909B (en) 2012-09-12 2012-09-12 Method for controlling power electronic hybrid system

Country Status (1)

Country Link
CN (1) CN102842909B (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103219733A (en) * 2013-03-18 2013-07-24 辽宁省电力有限公司电力科学研究院 Scattered power generation reactive compensation device and method with harmonic suppression function
CN103715693A (en) * 2013-11-11 2014-04-09 国家电网公司 Voltage compensation and fault current limiting hybrid system and control method thereof
CN104102752A (en) * 2013-04-01 2014-10-15 浙江省电力公司电力科学研究院 Simulation modeling method for static var compensator
CN104333002A (en) * 2014-10-29 2015-02-04 福州大学 Mixed active power filter based on ip-iq detection method and hysteresis control
CN105226676A (en) * 2015-11-16 2016-01-06 山西晋煤集团技术研究院有限责任公司 A kind of STATCOM controls the compensation method of time delay
CN105633979A (en) * 2016-03-21 2016-06-01 中国电力科学研究院 Reactive power compensation method and system
WO2018176213A1 (en) * 2017-03-28 2018-10-04 深圳中兴力维技术有限公司 High temperature early warning method and device
CN111371098A (en) * 2020-04-09 2020-07-03 武汉理工大学 Anti-noise power grid harmonic treatment system
CN111492550A (en) * 2017-12-19 2020-08-04 维斯塔斯风力系统集团公司 Method for suppressing harmonics

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
冷欣: "船用增压锅炉汽包水位预测控制方法研究", 《中国博士学位论文全文数据库（电子期刊）》 *
卜劲松: "《电网专业技术监督从书 电能质量专业》", 31 October 2011, 中国电力出版社 *
李俊堂: "有源电力滤波器谐波电流预测与控制方法研究", 《中国优秀硕士学位论文全文数据库（电子期刊）》 *
李翔等: "基于复合神经网络的非线性自校正控制器及其应用", 《工业仪表与自动化装置》 *
盘宏斌: "综合型电能质量调节装置的理论与技术研究", 《中国博士学位论文全文数据库（电子期刊）》 *
陈增强等: "具有比例积分结构的广义预测自校正控制器", 《控制与决策》 *

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103219733A (en) * 2013-03-18 2013-07-24 辽宁省电力有限公司电力科学研究院 Scattered power generation reactive compensation device and method with harmonic suppression function
CN104102752A (en) * 2013-04-01 2014-10-15 浙江省电力公司电力科学研究院 Simulation modeling method for static var compensator
CN104102752B (en) * 2013-04-01 2017-02-08 浙江省电力公司电力科学研究院 Simulation modeling method for static var compensator
CN103715693A (en) * 2013-11-11 2014-04-09 国家电网公司 Voltage compensation and fault current limiting hybrid system and control method thereof
CN103715693B (en) * 2013-11-11 2016-08-31 国家电网公司 A kind of voltage compensation and failure current limit hybrid system and control method thereof
CN104333002A (en) * 2014-10-29 2015-02-04 福州大学 Mixed active power filter based on ip-iq detection method and hysteresis control
CN105226676A (en) * 2015-11-16 2016-01-06 山西晋煤集团技术研究院有限责任公司 A kind of STATCOM controls the compensation method of time delay
CN105633979A (en) * 2016-03-21 2016-06-01 中国电力科学研究院 Reactive power compensation method and system
CN105633979B (en) * 2016-03-21 2018-06-26 中国电力科学研究院有限公司 A kind of Non Power Compensation Process and system
WO2018176213A1 (en) * 2017-03-28 2018-10-04 深圳中兴力维技术有限公司 High temperature early warning method and device
CN111492550A (en) * 2017-12-19 2020-08-04 维斯塔斯风力系统集团公司 Method for suppressing harmonics
CN111371098A (en) * 2020-04-09 2020-07-03 武汉理工大学 Anti-noise power grid harmonic treatment system

Also Published As

Publication number Publication date
CN102842909B (en) 2015-07-08

Similar Documents

Publication Publication Date Title
CN102842909A (en) Method for controlling power electronic hybrid system
CN100550568C (en) Harmonic wave dynamic managing and reactive-load dynamic compensation compound control method and implement device thereof
CN104734545B (en) The control method of the PWM rectifier controlled based on model prediction and voltage squared
CN108092308A (en) A kind of distributed virtual synchronous generator low voltage traversing control method
CN103956734B (en) A kind of multiple target comprehensive power quality controls and optimizes device
Farahani Intelligent control of SVC using wavelet neural network to enhance transient stability
CN108767869B (en) Static reactive power compensator voltage adjusting method based on artificial neural network
CN108448607A (en) A kind of micro-capacitance sensor battery energy storage system and off-network switching method and apparatus
CN104882886B (en) LLCL filtering-based active power filter compound control method
CN106532702A (en) Improved broadband adaptive repetitive control method of active power filter
CN103560516A (en) Parallel connection hybrid type active power filter and control method thereof
CN107611971A (en) For the net side inverter resonance full-order sliding mode control method of Voltage Harmonic distortion operating mode
CN107332282A (en) A kind of microgrid presynchronization grid-connection control system and its control method
CN105406741B (en) PWM rectifier Fuzzy Sliding Mode Control Approach during a kind of three-phase power grid voltage imbalance
CN107154650A (en) The control method for coordinating of many transverters of alternating current-direct current section in a kind of mixing microgrid
CN106300386B (en) Inhibit the Frequency servo method of power grid sub-synchronous oscillation based on SVG dynamic
CN101017980A (en) Multi-variant control method of the distribution static reactive power generator
CN106099943A (en) A kind of instantaneous power factor compensation control method and system
CN105629730A (en) UPFC control method based on nerve network sliding mode control
CN102820653B (en) Fuzzy-neural network double closed-loop control method of electric energy quality comprehensive controller
Chen et al. Control induced explicit time-scale separation to attain dc voltage stability for a vsc-hvdc terminal
CN104716856B (en) Modular multi-level converter model predictive control method
CN102957156A (en) Dynamic reactive power compensation control method
Wu et al. Research and Simulation of DC Microgrid Three-Phase AC-DC Converter Control Strategy Based on Double Loop
CN108233757A (en) Inverter control method based on NEW TYPE OF COMPOSITE control algolithm

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20150708

Termination date: 20210912