2
At a glance Dynamic computer simulations of elec- trical power systems play an important role in electrical network planning. The dynamic representation of the power system typically includes power plants, induction machines, load characteris- tics, high-voltage direct current (HVDC) connections and flexible AC transmis- sion systems (FACTS), such as static VAr compensators (SVC). The control- lers of these units influence the per- formance of the power system and are therefore modeled as well, based on standard or user-defined models. Modeling the dynamic representation of machines and controllers is con- ducted in Siemens PTI’s PSS ® software suite, i.e. PSS ® E, PSS ® SINCAL and PSS ® NETOMAC. Dynamic simulations are performed for: Grid interconnection studies, Control of system voltage and fre- quency, Damping of oscillations, Design, optimization and location of power system stabilizers and power oscillation damping devices Assessment of grid code compli- ance of power plants. The challenges One of the main challenges is to meet the required accuracy of the dynamic model for the planned investigation. In other cases models need to be simpli- fied and standardized to enable sophis- ticated simulations with limited availa- ble data, especially during the devel- opment phase of a project. Validation and improvement of indi- vidual dynamic models to enhance dy- namic simulations is also a challenging and ongoing task. Model validation can be done after commissioning of a unit by means of the commissioning records or with individual field tests at the units in service. Due to the fact that power systems are operated increasingly closer to their dynamic limits, accurate dynamic in- vestigations are essential to assess the dynamic limits and countermeasures. Our solution In order to perform a detailed and ac- curate dynamic simulation of an elec- trical power system, measurements, analyses, modeling and validation of control systems are conducted for par- ticular projects or entire power sys- tems. Some examples of control sys- tems are: Automatic voltage regulator (AVR), Power System Stabilizer (PSS), Governor and turbine performance, Power plant performance, FACTS, such as SVC and SVC PLUS, HVDC and HVDC PLUS, Power oscillation damping devices Development, coding and validation of standard or user-defined models for controllers are performed. Depending on the PSS ® software used for analysis, the controller structure can be coded in Siemens PTI’s graphical model build- er (GMB), in FORTRAN or using the BOSL code. Machine parameters of synchronous generators and induction machines are determined based on factory tests and measurements. Consequently, exten- sive data collection forms the basis for building dynamic models. Parameter identification is used for optimizing model parameters. Application examples Tuning of a PSS A PSS has to damp the inherent fre- quency of a generating unit, which is essential for proper operation of the power plant. Tuning of a PSS to this specific frequency is achieved and the resulting improvement is shown in the Bode diagram (Figure 1). It shows that the correct tuning of the PSS ensures a high damping in the range between 0.2 Hz and 2.0 Hz. Figure 1: Bode plot with and without a PSS Controller and machine measurements, modeling and analysis siemens.com/power-technologies

Controller and machine measurements, modeling and analysis€¦ · trical power systems play an important role in electrical network planning. The dynamic representation of the power

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Controller and machine measurements, modeling and analysis€¦ · trical power systems play an important role in electrical network planning. The dynamic representation of the power

At a glanceDynamic computer simulations of elec-trical power systems play an importantrole in electrical network planning. Thedynamic representation of the powersystem typically includes power plants,induction machines, load characteris-tics, high-voltage direct current (HVDC)connections and flexible AC transmis-sion systems (FACTS), such as staticVAr compensators (SVC). The control-lers of these units influence the per-formance of the power system and aretherefore modeled as well, based onstandard or user-defined models.

Modeling the dynamic representationof machines and controllers is con-ducted in Siemens PTI’s PSS® softwaresuite, i.e. PSS®E, PSS®SINCAL andPSS®NETOMAC.

Dynamic simulations are performedfor:

· Grid interconnection studies,

· Control of system voltage and fre-quency,

· Damping of oscillations,

· Design, optimization and locationof power system stabilizers andpower oscillation damping devices

· Assessment of grid code compli-ance of power plants.

The challengesOne of the main challenges is to meetthe required accuracy of the dynamicmodel for the planned investigation. In

other cases models need to be simpli-fied and standardized to enable sophis-ticated simulations with limited availa-ble data, especially during the devel-opment phase of a project.

Validation and improvement of indi-vidual dynamic models to enhance dy-namic simulations is also a challengingand ongoing task. Model validationcan be done after commissioning of aunit by means of the commissioningrecords or with individual field tests atthe units in service.

Due to the fact that power systems areoperated increasingly closer to theirdynamic limits, accurate dynamic in-vestigations are essential to assess thedynamic limits and countermeasures.

Our solutionIn order to perform a detailed and ac-curate dynamic simulation of an elec-trical power system, measurements,analyses, modeling and validation ofcontrol systems are conducted for par-ticular projects or entire power sys-tems. Some examples of control sys-tems are:

· Automatic voltage regulator (AVR),

· Power System Stabilizer (PSS),

· Governor and turbine performance,

· Power plant performance,

· FACTS, such as SVC and SVC PLUS,

· HVDC and HVDC PLUS,

· Power oscillation damping devices

Development, coding and validation ofstandard or user-defined models forcontrollers are performed. Dependingon the PSS® software used for analysis,the controller structure can be codedin Siemens PTI’s graphical model build-er (GMB), in FORTRAN or using theBOSL code.

Machine parameters of synchronousgenerators and induction machines aredetermined based on factory tests andmeasurements. Consequently, exten-sive data collection forms the basis forbuilding dynamic models. Parameteridentification is used for optimizingmodel parameters.

Application examplesTuning of a PSSA PSS has to damp the inherent fre-quency of a generating unit, which isessential for proper operation of thepower plant. Tuning of a PSS to thisspecific frequency is achieved and theresulting improvement is shown in theBode diagram (Figure 1). It shows thatthe correct tuning of the PSS ensures ahigh damping in the range between0.2 Hz and 2.0 Hz.

Figure 1: Bode plot with and without a PSS

Controller and machinemeasurements, modelingand analysis

siemens.com/power-technologies

Page 2: Controller and machine measurements, modeling and analysis€¦ · trical power systems play an important role in electrical network planning. The dynamic representation of the power

Parameter identification of aninduction machineIn order to determine the parametersof an induction machine, the electricaltorque vs. speed characteristic given bythe manufacturer is used. As an exam-ple, the results of a parameter identifi-cation process with PSS®NETOMAC isdepicted in Figure 2, where the origi-nal characteristics are shown in blackand the characteristics determined bythe identified parameters are plotted inred.

Figure 2: Results of a parameter identificationof an induction machine

Modeling and validation of anAutomatic Voltage Controller (AVR)After commissioning of a gas turbine,the dynamic simulation models are op-timized and validated with field testrecords of the commissioning.

Figure 3 shows a measured 5% voltagestep change of an AVR (red chart) andthe simulation in PSS®E with the opti-mized parameters of an AVR model(blue chart).

Figure 3: Graphical comparison of +5.0% volt-age step change at AVR (simulation and meas-urement)

Published bySiemens AG 2017

Energy Management DivisionFreyeslebenstrasse 191058 Erlangen, Germany

For more information, please contact:[email protected]

AL=N, ECCN=EAR99Subject to changes and errors. The infor-mation given in this document only con-tains general descriptions and/or perfor-mance features which may not alwaysspecifically reflect those described, orwhich may undergo modification in thecourse of further development of theproducts. The requested performance fea-tures are binding only when they are ex-pressly agreed upon in the concluded con-tract.