Classifications

• • H—ELECTRICITY
  • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
  • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
  • H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network

Definitions

• the invention relates to a method for quantitative analysis and decision of transient stability of a power system and a control system thereof. Background technique
• the stable operation of the power system is extremely important to the national economy and social civilization.
• system planning, operation planning, online operation, and training simulation it is necessary to understand the dynamic behavior of the power system under anticipated accidents and quantitatively evaluate the distance from the stability limit.
• preventive and / or emergency control decisions need to be made quickly and correctly.
• emergency control it is necessary to make adaptive decisions on the control place, the type of measure, and the control amount according to the actual working conditions and faults.
• the EE AC (Extended Equal Area Criterion) method (Extended Equal Area Criterion) method invented by the inventor of this patent in 1986 is the only direct method that has been rigorously proven and analytically solved.
• EEAC has undergone three phases: static (SEEAC beginning in 1986), dynamic (DEEAC beginning in 1991), and integration (IEEAC Integrated Extended Equal Area Criterion beginning in 1993). More than 30 papers in the first two stages (for example, Y, Xue et al, EEAC: justifications, generalizations, applications. IEEE Trans, on Power Systems 1989 No. 1, PP. 44-52; and Y'Xue et al, DEEAC, part 1. Basic formulation. Athems Power Tech. September, 1993) has been widely cited and tracked at home and abroad. These articles, which are cited above, shall be deemed to have been incorporated in the present application.
• SEEAC and DEEAC have the following problems: (1) can not reflect the stability of multiple pendulums; (2) can not handle the phenomenon of isolated stable area ISD; (3) the contradiction between the accuracy and speed of stability analysis of the first and second pendulum is not enough Ideally, the method is not robust enough. For the processing of complex models, the above problems are more prominent, so that the numerical integration method no longer has a clear speed advantage. In this way, online monitoring and control of the stability of the grid cannot be implemented, and its practicality is limited System. Summary of the invention
• the invention improves EEAC to IEEAC, so that it can meet the requirements for practicality of various projects.
• the improvements mentioned include extending from pendulum stability analysis to multi-pendulum stability analysis, from classic models to arbitrarily complex models, from two-period to multi-period, and from preventive control to fast-off, machine-cut, load-shedding, and Emergency control methods such as electric braking.
• IEEAC has passed the assessment of engineering applications in China's power system. Its technical content has not been disclosed in conferences, magazines or talks, nor has it been disclosed to domestic users. The technology exported to foreign countries is limited to the DEEAC stage.
• the purpose of the present invention is to overcome the shortcomings of the prior art and provide a method for stable quantitative analysis of power systems.
• Another object of the present invention is to provide a decision-making method for stability prevention control of an electric power system, a decision-making method for emergency control, and a control system thereof.
• the invention can calculate the longest duration of failure that the power system can withstand (ie, the critical cut-off time CCT, or t c ), shows the stability margin ⁇ of the power system, and provides a stability domain in the injected power space; the invention can also A new emergency control framework and corresponding control system implementation method are provided, which are "on-line periodic update of decision table and matching according to fault information in real time".
• the technical solution of the present invention is to combine the analyticity of SEEAC with the strictness of the necessary and sufficient conditions for maintaining the stability of the multi-machine system of DEEAC, and use the former to provide the initial value and approximate sensitivity coefficient to the latter; Combining the fastness of the variable step integral trajectory PCOI (partial inertia center) mapping accuracy; using the mechanism of multi-machine factors and complex model factors revealed by EEAC theory to affect the time-varying of the map to resolve the instability mode with Resolve the problem of time variation; combine the deep knowledge of EEAC and the knowledge processing capabilities of expert systems to further improve computing efficiency and identify potential isolated stable region phenomena; will deal with IEAAC (CM) of the classic model (CM) and complex processing Model (DM)
• IEEAC is based on strict theory. It separates the integration space from the observation space, and integrates according to the required full model and scene in multi-machine space. The resulting critical instability The trajectory has taken into account the effects of all non-autonomous factors. Partial inertia center transformation was then used to map the multi-machine power angle trajectory to the observation space of the two machines, and the stable characteristics were strictly retained during the transformation process.
• the corresponding mapping is a critical mapping
• the corresponding instability group is a critical group.
• Such screening is achieved by CM, and in the most severe scenarios (or working conditions) that we are interested in, using variable large step lengths to perform labor. If the trajectory does not diverge, this case will not be studied again. If the trajectory diverges, find the (not necessarily the most serious) image corresponding to the maximum angular gap and use sensitivity analysis techniques to find the candidate critical conditions. According to this condition, a set of near-critical instability trajectories is obtained.
• Candidate critical maps are selected according to the largest n angular gaps at the dynamic saddle points. Usually, n is 3 and it is sufficient. The value of n is adaptively adjusted.
• t is obtained using SEEAC (CM).
• CM SEEAC
• CM DEEAC
• DM critical result given by SEEAC
• DM dynamically converted as the initial value of DEEAC
• the IEEAC method proposed by the present invention can completely solve the defects existing in the prior art in power system stability quantitative analysis and control decision-making.
• the method includes the following steps:
• the IEEAC separates the integration space from the observation space, and uses SEEAC and DEEAC to provide the initial value of Xc to the numerical integration method,
• the result of numerical integration is mapped to the observation space for quantitative stability analysis;
• the calculation process of Xc includes the following steps:
• the stability limit value Xc of X is taken as the stability measure and limit value of the original multi-machine system to replace the estimated value given in step (2); (5) The estimated value given in step (4) above The value is used as the initial value, and the SEEAC and / or DEEAC under the complex model (DM) are used to improve the estimated value; (6) Under the DM, the estimated value given in step (5) above is used as the simulation condition to obtain the multi-machine system Approximate critical trajectory; (7) PCOI map the trajectory given in step (6) above to the candidate critical observation space to obtain the critical pendulum critical instability condition or critical stability condition Xc in multi-machine space; (8) for critical Instability Xc, gradually reduce the severity of X to search for the critical value of multi-swing instability; for critically stable Xc, gradually increase the severity of X to search for the critical value of the first pendulum instability; isolate when necessary Stability domain (ISD) check.
• ISD Stability domain
• the step of using SEEAC to provide preliminary estimates of candidate t c for each observation space includes the following steps:
• each machine When the power angle of each machine is arranged in a queue from large to small; then take the largest three or less angular gaps in the queue; each angular gap divides the queue into two, corresponding to a PCOI Mapping; (3) For each candidate mapping, use SEEAC under CM to find (CM), and then use the latter as the initial value of the four-step DEEAC under CM.
• the sensitivity coefficients of each order of the stability margin to ⁇ involved in the step of using the DEEAC to correct the initial estimate of t c are obtained by numerically perturbing the ⁇ value using four-step DEEAC.
• the step of obtaining the multi-machine trajectory under the CM includes the following steps:
• the step of using the SEEAC and DEEAC under DM to improve the t c estimate includes the following steps:
• the generator model is Eq 'constant or more complex model. If the first pendulum is critically unstable under CM, the multi-machine system is condensed into an equivalent two-machine system with a complex generator model according to the critical image under CM. , Ie os) -1 computes S
• the step of obtaining the approximate critical trajectory of the multi-machine under the DM includes the following steps:
• the generator model is processed as an E Q constant model that is updated piece by piece, and the non-linear load model is processed as a constant nozzle load model that is updated piece by piece.
• CM which is updated piece by piece;
• the unfolding is performed on the admittance matrix without reduction order, in which the high-order coefficients of the Tailau series can be obtained by using the differences of the previous steps; the dynamic equations of the controller model are integrated using small step interpolation and
• the stride series of large steps expands the corners of each machine, and then calculates the bus voltage. For more complex network models that take into account dynamic loads, DC power lines, and static voltage compensators, they are stabilized with commercial transients.
• the step of checking the multi-swing instability and ISD includes the following steps:
• the four power angle areas corresponding to the critical group and the non-critical group during and after the failure are calculated according to the SEE AC and DEE AC methods; if the same area is calculated in the two modes, If the relative difference (time-varying degree ⁇ ) is not large, the dynamics are considered to have a typical two-group mode; at this time, it is not necessary to check the multi-swing instability and ISD; (2) If ⁇ is large and the system is stable, then ⁇ increases by 10% to try again until it is unstable; on the contrary, if the system is unstable, it decreases stability; PCOI mapping is used to obtain the critical critical unstable trajectory to obtain t (CM); (3) If the critical instability mode under CM is the first Instability, it is not necessary to check the ISD under DM; (4) Otherwise, the perturbed trajectories of each machine are divided into homogeneous clusters according to the closeness of each extreme point coordinate ( ⁇ , ⁇ ) sequence of each machine rotation angle in the critical stable trajectory , According to the change trend
• a method for preventing and controlling transient stability of a power system includes the following steps: For a system with insufficient stability, the generation power Pms of the critical generator group is taken as the target parameter X, and the critical generation power Pmsl is obtained by the IEAAC. ; When the fault has never occurred, adjust the actual power generation of the cluster to below Pmsl, and increase the power generation of other clusters to maintain the normal power supply of the load.
• An emergency control decision method for transient stability of a power system includes the following steps ⁇
• an on-line transient stability monitoring and prevention control system for a power system constructed by using the power system transient stability prevention control decision-making method includes:
• a data acquisition and monitoring subsystem consisting of a remote terminal unit (RTU), a channel, a front-end processor and a SCADA unit; a real-time network analysis subsystem that performs real-time topology analysis and state estimation on the information output by the data acquisition and monitoring subsystem ; A subsystem for making transient stability prevention and control decisions using the IEEAC method.
• RTU remote terminal unit
• SCADA SCADA unit
• a power system transient stability emergency control system constructed by using the power system transient stability emergency control method according to the present invention includes:
• a data collection subsystem that collects various operating condition data of the power system; according to the operating condition information and pre-defined fault scenarios collected in real time during operation, the IEEAC calculates appropriate emergency control measures online, and periodically updates the online budget of the decision table Subsystem; when an actual fault is identified, a real-time matching subsystem that obtains measures to be taken based on real-time look-up of the fault information, and an execution subsystem that executes control commands of the real-time matching subsystem.
• the following table shows the actual calculation results of the present invention on 16 power grids covering almost all of China.
• the results show that the calculation accuracy of the stability of the first pendulum or multi-pendulum is extremely high.
• the average CPU time required to calculate a three-phase short circuit with a 4MIPS computer is: Total number of generators in the system CPU (seconds / example) Zhejiang Network 40 296 6. 12 Northeast Network 36 182 15. 72 North China Network 52 372 33. 04
• Figure 1 is a schematic flow chart of IEEAC.
• Fig. 2 is a block diagram of the transient stability online monitoring and prevention control system of the power system of the present invention.
• FIG. 3 is a block diagram of a power system transient stability emergency control system according to the present invention. Best Mode of the Invention
• FIG. 1 is a schematic flow chart of the analysis and prevention control-oriented IEEAC to obtain the critical switching time t c .
• the target parameter of interest is the total load level or the stability limit of power generation, it can be analogized to the process in Figure 1.
• steps 1 to 10 analyze the stability under CM.
• step 1 uses the scene filtering method (ICMR) to filter out the calculation examples with t c greater than 1 second, and can reliably avoid the danger caused by ISD.
• step 2 After the trajectory is obviously unstable (for example, when the maximum power angle difference is greater than 300 °), the power angles of the various machines are arranged into an initial queue of the critical degree of the generator, and the system is divided into complementary two according to the maximum angular gap Group to get the corresponding initial critical map. For the latter, SEEAC and DEEAC in 4-step Tailo series expansion are used to obtain the initial t co .
• ICMR scene filtering method
• Step 4 uses SEEAC to calculate t for each candidate critical image.
• Step 5 is modified by 4-step DEEAC, where the value of ⁇ is taken as the minimum value of t and E existing in tg and other images.
• Step 6 take each candidate threshold Minimum value t E as the original image in the multi-machine system t E, and the corresponding image of that first swing critical mappings.
• Step 7 calculates four kinds of acceleration areas of the S (corresponding A) group during and after the fault in the image. If the relative difference (called the time-varying degree ⁇ ) between the value obtained by the SEEAC principle and the corresponding value obtained by the DEEAC principle is not large (for example, less than 60%), it is confirmed as a typical two-group dynamics; Here are two steps. When the ⁇ value of any kind of area exceeds a threshold value (for example, 20%), it is considered that the multi-group dynamic characteristic is significant, and the following two steps must be performed.
• a threshold value for example, 20%
• Step 8 tg E as initial value of ⁇ , the variable-length steps Taylor series search ti; E, if the first swing instability resulting multi-machine track, the other larger gaps and the critical mappings (if present ) To map the corresponding image to get the stability margin ⁇ .
• the 4-step DEEAC is then used to perform numerical perturbation on the value of ⁇ to obtain the first- and second-order sensitivity of ⁇ to ⁇ to assist the search of ⁇ until the resolution reaches the requirement. If a stable multi-machine trajectory of the first pendulum is obtained, the multi-machine rotation angle is read at the time of the equivalent angle return on the critical image, and the corresponding equivalent two-machine time-invariant power angle curve is calculated to obtain ⁇ to assist Search for ⁇ .
• Step 9 uses the deep knowledge of the expert system to determine whether there is a possibility of potential ISD. If possible, take a smaller value of ⁇ and integrate again to detect ISD.
• the above-mentioned deep knowledge refers to the division of multi-machine clusters according to the closeness of the coordinate sequence of each extreme point of each machine's rotation angle curve of the critically stable trajectory of the first pendulum. When each machine cluster is an instable group, its ⁇ varies with the frequency If there is a downward trend in ⁇ corresponding to adjacent swings in the same direction, and ⁇ is small, ISD may exist.
• step ⁇ 19 If it is required to use a model that is more complex than the classic model, but not more complicated than the E'q constant plus nonlinear load model (denoted as DM1), perform steps 13 to 19. If the model requirements are more complicated than DM1, you need to perform steps 12 to 18 first, and then perform step 20 instead of step ⁇ 19.
• Step 12 ignores more complex factors than DM1 for the time being, and obtains an approximate DM1 model. In step 20, these more complex factors are taken into account.
• Step 13 uses the inertia as a weighting factor to condense a multi-machine system with a complex generator model into a two-machine system with a complex generator model, and then tries to integrate the two-machine system with the initial value of t (CM) as ⁇ . To solve (DM1). If the load adopts a complex model, the parameters of the equivalent two-machine system must be modified in sections. Step 14 quickly find the initial value of t E (CM1). The starting point is that the influence of time-varying factors on the critical resection angle is basically irrelevant to the complexity of the model.
• Steps 15 and 16 make the pendulum instability only under the CM, and the dynamic characteristics of the two groups are not obvious. Only need to perform step 17.
• Step 17 uses large steps to update the parameters of the equivalent two-machine system under DM1, thereby improving the estimation accuracy of t E (DM1).
• Step 19 expands the DM1 step size series in a multi-machine space.
• the second derivative of the rotation angle is obtained from the acceleration equation
• the third derivative is obtained from the acceleration value obtained by the previous two expansions
• the first derivative is obtained from the higher-order derivative of the previous step. Therefore, the first two steps after each disturbance are expanded with a fixed small step size, and the third derivative of the corner is not considered.
• the third-order derivative is counted from the third step, and the variable step length is estimated according to the third-order term of the Taiwan Labor Series, and the voltage change is predicted and checked, and the step size is re-corrected if necessary.
• the equivalent angle returns (determined as stable) or reaches the dynamic saddle point (determined as unstable)
• the integration can be stopped.
• ⁇ (DM1) is the minimum value.
• ⁇ and its sensitivity is similar to step 8, but the DM1 model should be used.
• t (DMl) can directly take tg E (DMl) without detailed integration of the multi-machine system. Only when the CM is multi-swing critical instability, or when the valve is reopened after fast closing, the multi-swing instability problem of a complex model needs to be considered. ISD verification under complex models is similar to that under CM.
• step 20 no longer uses the Tailau series expansion scheme, but uses any suitable commercial transient stable numerical integration software package to implement the numerical integration of the full model (DM) .
• the EEAC provides initial estimates of critical conditions, and a quantitative analysis based on PCOI mapping is performed on the integrated trajectories instead of blind search.
• the target parameter of interest is not r, but other single parameters, such as the total load level L under the specified load distribution factor, the extreme value in each step of FIG. 1 is no longer t c but:. Therefore, when using IMCR to select the initial image in step 1, the working condition used is the maximum value in the range of L values we are interested in.
• the critical value L c can be obtained by analyzing the sensitivity of the stability margin to L.
• Figure 2 is a block diagram of the transient stability on-line monitoring and prevention control system based on the above algorithm.
• Various states to be collected by the remote terminal units 33 distributed throughout the power system 31 Information (such as the position of circuit breakers and knife gates, the status of various control devices and equipment) and numerical information (such as voltage, current, active power and reactive power) are sent to the front-end machine 37 through the channel.
• the processed information is sent to the energy management system (EMS) 40.
• EMS energy management system
• the front-end computer 37 and the EMS can be a computer system with a main frame structure, or a fully distributed computer system.
• the online information is sent to the data acquisition and monitoring database (SCADA DB) 44 through the interface 38, and the SCADA software package 46 performs general monitoring report processing.
• SCADA DB data acquisition and monitoring database
• RTNET real-time network analysis
• NETMODEL network model
• PES power system application software
• TSA DB transient stability analysis database
• This function module can be started automatically periodically, or automatically after an operating event occurs in the power system or at the request of a dispatcher, to analyze all the expected faults specified in the accident table one by one. In addition, it can also analyze the research conditions provided by the dispatcher power flow to achieve operation planning or system planning.
• the remote terminal unit 33, channel, front end 37 and SCADA unit 42 constitute a data acquisition and monitoring subsystem
• the RTNET database 50, software module 54 and NETMODEL database 52 constitute a real-time network analysis subsystem 48
• PAS database 58 And PAS module 60 constitutes an online economic analysis and static safety analysis subsystem 56
• TSA database 64 and IEEAC module 66 constitute a transient stability prevention control decision-making subsystem 62.
• IEEAC can give a stability region surrounded by the transient stability limit power of each generator (group) in the injected power space.
• the latter can be intuitively represented by a bar graph. Means. Compare the actual power generation of each machine (group) with the stability domain. If no one machine (group) 's power exceeds the stability domain, no preventive control is required, and the power system can withstand various expectations based on Failure without instability. Otherwise, the dispatcher can use the man-machine interface and the data acquisition and monitoring subsystem to The generation power of the (group) is reduced to the corresponding limit value to ensure transient safety.
• IEEAC can also provide stable regions defined by other parameters.
• the functions provided by the subsystem 62 are not available in all current EMS systems.
• the hierarchical structure of SCADA and EMS systems in this system starts from the bottom: computer hardware system, real-time operating system, support software platform, SCADA system, real-time network topology and EMS system.
• the IEEAC unit works in real-time or research mode. Not only can it provide CCT, but it can also express the stability limit with parameters that the operator can observe and control. If the system develops in an unsafe direction, it will warn the operator of this danger and tell them how to take precautions to move the system to a stable and safe state, no longer relying on the experience of the operator.
• Figure 3 is a block diagram of the transient stability emergency control system of the power system.
• the system's decision table is constantly updated according to the actual working conditions, so it is highly adaptive.
• the invention utilizes the features of the IEEAC to quickly calculate the stability margin and perform sensitivity analysis to evaluate the effects and limit values of various emergency control measures. Because the actual working conditions can be tracked online, all the anticipated faults are refreshed and the decision table is refreshed in a short period of time. Thousands of irrelevant working conditions no longer need to appear in the decision table. This not only avoids a lot of offline calculations, but also makes full use of all quasi-real-time information. In this sense, of course, it does n’t exist. Possibility of mismatch in working conditions. In addition, the functional requirements of the expert system only improve the calculation efficiency without affecting the accuracy of the calculation and the correctness of the decision.
• various numerical and switching information of the power system 31 is sent to the EMS 40 through data acquisition equipment and channels (not shown), and the quasi real-time operating conditions obtained after the latter are sent to the emergency control through dedicated channels.
• the intelligent communication interface 84 of the system 86 should include at least the total power generation amount, total load amount, unit operation and network topology information of the power system and each zone. Any quantity that can be obtained should be kept in the calculation model as much as possible.
• the real-time value of a certain quantity is not obtained due to accident, the previous value is used, but after a certain time, the corresponding planned value is automatically adopted.
• the near real-time information is also sent to the pre-processing unit 82 of the emergency control system through a dedicated channel.
• DSPi 94, ⁇ DSP n 92 are parallel processors of the execution layer, DSP. It is the coordination layer processor.
• DSPi 94, ⁇ DSP n 92 are parallel processors of the execution layer, DSP. It is the coordination layer processor.
• a complete information set consistent with the calculation model is obtained, and then the expert system ES at the coordination layer is obtained.
• 116 is loaded into each DSP of the execution layer.
• various available emergency measures in the available measures data block 112 and corresponding priority and cost rights information are also loaded at the same time.
• the expected accident table is stored in the expected accident data block 114, ES.
• the calculation task is assigned to each machine of the execution layer in units of expected accidents.
• the latter is responsible for the pre-decision of a failure scenario at a time.
• the expert systems 96, 102 and IEEAC modules 98 and 104 are used together. System operating conditions and by ES. Releasing fault scenarios to select appropriate emergency measures (intelligent integer programming problem).
• DSP. Middle 118 is decision table 1, and 120 is decision table 2. These two tables are alternately in preparation and execution. DSP. After receiving the result of the execution layer processor, it is stored in the table in the ready state. When the latter has completed all updates, it has the conditions to switch to the execution state.
• the EMS 40, pre-processing unit 82, and intelligent communication interface 84 constitute a data acquisition subsystem 80; multiple execution-layer parallel processors DSPi 94 to DSP n 92 are formed in Line budget subsystem 92; coordination layer processor DSP.
• the real-time matching subsystem 106 is formed; the control signal output unit 88 and the channel constitute a control subsystem that executes control commands.
• the real-time matching subsystem includes: expert system 116, information processing module 108, real-time operating data block 110, available measures data block 112, expected accident data block 114, decision tables 118, 120; it may also include a human-machine interface 70 .

Landscapes

• Engineering & Computer Science (AREA)
• Power Engineering (AREA)
• Supply And Distribution Of Alternating Current (AREA)

Applications Claiming Priority (4)

Publications (1)

Family

ID=25743748

Family Applications (1)

Country Status (1)

Cited By (15)

• 1996
  • 1996-02-09 WO PCT/CN1996/000012 patent/WO1996027231A2/fr active Application Filing

Similar Documents

Legal Events