LU500700B1 - Adaptive optimization control method for failure in true bipolar flexible dc transmission system - Google Patents

Abstract

Disclosed is an adaptive optimization control method for a failure in a true bipolar flexible DC transmission system, which includes: acquiring in real time a power flow status, a topology status, and grid parameters of a DC grid, and information about a new energy field; determining an expression of an objective function for maximizing the consumption of new energy; analyzing constraint conditions of the true bipolar DC transmission system; generating a contingency set; and establishing an equivalent optimization model for the whole system, and finding optimization control schemes corresponding to all contingencies in the contingency set, to form an offline optimization scheme library. When a failure actually occurs in the system, the closest optimization scheme in the offline library is immediately matched and applied, and simultaneously an accurate real-time optimization scheme is calculated and applied based on an actual system status after the failure; and then, the contingency set is updated according to an actual operating status of the system, and the offline optimization scheme library is recalculated. The method is applicable to various types and topologies of true bipolar DC transmission systems, and the proposed optimization control strategy is generic and has a wide range of applications.

Description

ADAPTIVE OPTIMIZATION CONTROL METHOD FOR FAILURE IN 7900700

TRUE BIPOLAR FLEXIBLE DC TRANSMISSION SYSTEM

BACKGROUND OF THE INVENTION Field of the Invention The present invention relates to an adaptive optimization control method for a failure in a true bipolar flexible DC transmission system, and belongs to the technical field of DC transmission.

Description of Related Art Asthe flexible DC transmission system is developing towards higher voltage level, greater transmission capacity, multi-terminal, and networking properties, a flexible and reliable true bipolar system structure will have a broad application prospect. Because of its flexibility and strong support, the true bipolar system provides an effective technical means to realize large-scale delivery of new energy, and is widely used in new energy collection scenarios such as wind power collection. However, the flexible characteristic of the true bipolar system also puts forward higher requirements on the coordinated control of converter stations. There is still an urgent need for an optimization control strategy to effectively improve the consumption capacity of new energy and to achieve rapid self-healing and restoration for a failure, which has not been thoroughly studied at present.

In particular, when a true bipolar DC system is connected to a new energy grid, in consideration of maximizing the consumption of new energy output, control quantities of the same type are given to positive and negative converters of a converter station, such that the power injected into the positive and negative transmission lines from the AC grid is consistent in magnitude and direction. However, the advantageous characteristic of independent controllability of the positive and negative converters is not fully utilized in controlling the two converters to flexibly distribute a new energy increment to the DC network. In addition, for the true bipolar DC system targeted at the consumption maximization of the new energy, when the grid is faulty, there is currently a lack of a coordination strategy based on an offline optimization scheme library containing multiple contingencies and online optimization that can be used for restoration to normal operation by rapidly matching the most appropriate optimization control scheme, so as to realize adaptation to failure situations.

SUMMARY OF THE INVENTION HUS00700 Technical Problem In view of the shortcomings in the prior art, the present invention provides an adaptive optimization control method for a failure in a true bipolar flexible DC transmission system, which can rapidly match an optimization control scheme of a converter station at each terminal for a failure on the premise of realizing consumption maximization of new energy, and fully utilize the feature of independent controllability of positive and negative converters in the true bipolar system, thus effectively shortening a restoration time after the occurrence of a failure and improving stability of an electric power system.

Technical Solution The present invention adopts the following technical solution to solve the foregoing technical problem: An adaptive optimization control method for a failure in a true bipolar flexible DC transmission system is provided, which includes the following steps: step 1: under the condition that the true bipolar flexible DC transmission system is in a stable operating status, acquiring a grid topology, grid parameters, and an operating status in real time, where the operating status includes an initial voltage and active power of each terminal in an initial stage; step 2: establishing an equivalent optimization model of the true bipolar flexible DC transmission system, where the true bipolar flexible DC transmission system has n terminals in total which are successively numbered as 1, 2, ..., m, m+1, ..., q, q+1, …, N, the ith (i=1, 2, ..., q) terminal being a controllable terminal, the ith (i=1, 2, ..., m) terminal being a controllable new energy terminal, and the ith (/=g+1, ..., n) terminal being an uncontrollable terminal, g<n, and m<q, an objective function of the equivalent optimization model is selected according to an actual transmission capacity of a converter station, specifically: 1) when the actual transmission power of the controllable new energy terminal does not reach the upper limit of the converter station capacity, the following objective function is selected: 300 Fu = MAR yy tt AR Mu (1 =1,2,…M) being a weight coefficient of an active power increment AP of the new energy at the ith terminal, and /=1,2,...,m; and 2) when the actual transmission power of the controllable new energy terminal reaches the upper limit of the converter station capacity, the following objective function is

N , , © | LU500700 selected: J £5) = Floss dan + Bits iow Ft > Paci) last (i=1,2,...,m) denoting AC- side active power of the ith terminal after optimization control; and constraint conditions for the equivalent optimization model include: Po Pr we S0{=12.00m) is met when the objective function in the case 1) is selected; Pots te = Pun €0{i=12...m} is met when the objective function in the case 2) is selected; Kon 5 Fegeman (7 = L Zeon mi) {1 7 Kes Hon % Baume» (i = bacs a} mask, cal F | Pd | in } = LA mmx on {i = mrt, rams gq} max QUA, nd! Pn Bas FE Bega = mL 200g) Pri Eins = Fis p {i= gil2...., a} Pa. ‘ + Fein 8 = Piscine. i {i = gl 2, res 1} CEN SU, SUR (I=L 2,0) 0 i] SE lati # f) where Pacñ, denotes AC-side initial active power of the ith terminal, k; denotes an active power distribution coefficient of a positive converter in the converter station at the 7th terminal of the system, Pysc@max p and Prscimax n respectively denote active power upper limits of positive and negative converters in the converter station at the ith terminal, Pas p and Pdcñ) n respectively denote DC-side active power of the positive and negative converters in the converter station at the ith terminal, P/oss(y p and Pross() n respectively denote active power losses of the positive and negative converters in the converter station at the ith terminal, Us; denotes a DC voltage of the converter station at the ith terminal, ÇA" and U respectively denote upper and lower limits of the DC voltage of the converter station at the ith terminal, lu.) denotes a current transmitted by a DC line between converter stations at the ith and jth terminals, and 1°” denotes an upper limit of the current transmitted by the DC line between the pP y converter stations at the ith and jth terminals;

step 3: initializing the equivalent optimization model established in step 2 by using LU500700 the acquired grid topology, grid parameters, and operating status as initial values; step 4: generating a contingency set according to historical failure data of the true bipolar flexible DC transmission system; step 5: finding optimization control schemes corresponding to all contingencies in the contingency set in step 3 according to the equivalent optimization model established in step 2, to form an offline optimization scheme library; and step 6: when the status of the true bipolar flexible DC transmission system changes, proceeding to the operating status determination: if a current grid topology or power flow is inconsistent with an initial grid topology or power flow, there are the following three handling methods: 1) immediately matching and applying the closest corresponding optimization control scheme in the offline optimization scheme library; 2) conducting online optimization based on the current grid status, to find a corresponding optimization control scheme in real time; and 3) immediately matching and applying the closest corresponding optimization control scheme in the offline optimization scheme library; then conducting online optimization based on the current grid status and parameters, to find a corresponding optimization control scheme in real time; and replacing the matched optimization control scheme with the optimization control scheme found in real time.

Further, the power flow status in step 1 includes an initial voltage and active power of each node.

Further, the method further includes: updating the offline optimization scheme library while the grid topology and power flow have a change: if the grid topology changes, updating the equivalent optimization model; and if only the power flow changes, by using a power flow status after restoring from the failure to a steady state as an initial value of the updated equivalent optimization model, finding optimization control schemes corresponding to all contingencies in the contingency set once again, to update the offline optimization scheme library.

Further, when the optimization control scheme 1s applied in step 6, the active power of the positive and negative converters is tuned in the following manner so as to realize optimization control: in converter stations in which positive and negative converters are independently controllable, respectively assigning active power distribution coefficients 4; and 1-k5) to the positive and negative converters in a converter station adopting master-slave

P.

LU500700 control, where A re Paci) p denoting the AC-side power of the positive converter at the 7th terminal, and ka“) meeting 0< ka) <1, in which case, an optimization control quantity is the active power distribution coefficient ka) of the positive converter, and positive and negative active power injections are respectively adjusted to An Pacd) 5 and (1-kw)Pacñy by using the optimization control scheme; for a converter station adopting droop control, the optimization control quantity includes the active power distribution coefficient ks of the positive converter or respective droop coefficients Ku p and Ka) n of the positive and negative converters; and for converter stations adopting a hybrid control strategy in which in the same converter station, converters of one polarity use an active/reactive power coupling control manner, while converters of the other polarity use an active/reactive power decoupling control manner containing the master-slave control or droop control, the optimization control quantity includes the active power distribution coefficient k; of the positive converter, respective droop coefficients Ku)» and Ka) n of the positive and negative converters, inner-loop and outer-loop controller parameters, and a control target reference value.

Advantageous Effect Compared to the prior art, the present invention has the following technical effects by using the foregoing technical solution: 1) The present invention designs an inter-pole cooperative control strategy between positive and negative converters adopting different control manners, which fully utilizes the feature of independent controllability of the positive and negative converters in the true bipolar DC system, can coordinate specific active power distribution between the two poles according to a new energy consumption requirement and an operating condition, and can use a healthy pole and healthy line to actively bear part of the transmission power of a faulty pole and a faulty line in an abnormal operating condition, so as to avoid excess transmission power of the faulty pole, and generate a restoration control scheme according to a current operating status, which solves the current problem that it is difficult to realize adaptive restoration for a failure, and improves the flexibility and reliability of the bipolar system. 2) In addition to simulation of contingencies, the adaptive optimization control strategy for a failure proposed by the present invention also considers the applicability in uncertain scenarios such as the output prediction of new energy by using universal true bipolar DC power flow calculation and multiple optimization algorithms; and the LU500700 optimization results cover all common failure scenarios and meet the consumption requirement for the new energy, thus providing a reasonable forecasting and adjustment scheme for the optimization and scheduling on incorporating a new energy electric field into the flexible DC grid. In this way, the grid can rapidly restore to a stable operating status by use of the restoration control scheme after the occurrence of a failure.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is an overall flowchart of an adaptive optimization control strategy for a failure in a true bipolar system; FIG. 2 is a flowchart of an adaptive offline optimization control strategy for a failure in a true bipolar system; FIG. 3 is a flowchart of an adaptive online optimization control strategy for a failure in a true bipolar system; FIG. 4 is a diagram of a ring topology structure of a four-terminal true bipolar VSC- MTDC system; and FIG. 5 is a schematic structural diagram of a true bipolar single-terminal converter station.

DETAILED DESCRIPTION OF THE INVENTION The technical solution provided by the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the following detailed description is only used to illustrate the present invention and is not intended to limit the scope of the present invention.

An adaptive optimization control method for a failure in a true bipolar flexible DC transmission system is provided, which includes the following steps: step 1: under the condition that the true bipolar flexible DC transmission system is in a stable operating status, acquiring a grid topology, grid parameters, and an operating status in real time, where the operating status includes an initial voltage and active power of each terminal in an initial stage; step 2: establishing an equivalent optimization model of the true bipolar flexible DC transmission system; step 3: initializing the equivalent optimization model by using the acquired grid topology, grid parameters, and operating status as initial values; step 4: generating a contingency set according to historical failure data of the true bipolar flexible DC transmission system; LU500700 step 5: finding optimization control schemes corresponding to all contingencies in the contingency set in step 3 according to the equivalent optimization model established in step 2, to form an offline optimization scheme library; and step 6: when the status of the true bipolar flexible DC transmission system changes, proceeding to the operating status determination: if a current grid topology or power flow is inconsistent with an initial grid topology or power flow, there are the following three handling methods: 1) immediately matching and applying the closest corresponding optimization control scheme in the offline optimization scheme library; 2) conducting online optimization based on the current grid status, to find a corresponding optimization control scheme in real time; and 3) immediately matching and applying the closest corresponding optimization control scheme in the offline optimization scheme library; then conducting online optimization based on the current grid status and parameters, to find a corresponding optimization control scheme in real time; and replacing the matched optimization control scheme with the optimization control scheme found in real time.

Further, the establishing the equivalent optimization model of the true bipolar flexible DC transmission system includes the following steps:

1. The true bipolar flexible DC transmission system has # terminals in total which are successively numbered as 1, 2, ..., m, m+1, ..., q, q+1, ..., n, where the ith (i=1, 2, ..., q) terminal is a controllable terminal (q<n); the ith (i=1, 2, ..., m) terminal is a controllable new energy terminal, a converter station being connected to a new energy electric field (m<q), and the ith (i=g+1, ..., n) terminal is an uncontrollable terminal. The equivalent optimization model of the true bipolar flexible DC transmission system is established according to the foregoing numbering rule.

An objective function is selected according to an actual transmission capacity of the converter station. When the actual transmission power of the controllable new energy terminal does not reach the upper limit of the converter station capacity and there is still a transmission margin, the following objective function is selected: Th = iy By + May Foo, where #(1) denotes a weight coefficient of an active power increment AP of the new energy at the ith terminal (1<7<m). Further detailing may be considered according to an actual output forecasting curve regarding the new energy field, but the present invention is not limited thereto; or a value is proportionally assigned according to a LU500700 capacity of the new energy field in lack of forecasting information.

When the actual transmission power of the controllable new energy terminal approaches the upper limit of the converter station capacity, the following objective function is selected: Fx), FF, (itr Ÿ Posy cr FE it where Pac) las denotes AC-side active power of the ith terminal after optimization control (1<i<m).

2. Decision variables involved in the optimization and the number of the variables are determined.

For a large-scale complex true bipolar multi-terminal flexible DC transmission system which adopts an active/reactive power coupling control manner, an active/reactive decoupling control manner containing master-slave and droop control strategies, and a hybrid control manner of simultaneously using active/reactive power coupling control and active/reactive power decoupling control in a converter station at the same terminal; or other control strategies with a similar principle, the decision variables involved in the optimization includes: optimization control quantities which include an active power distribution coefficient ky of a positive converter, respective droop coefficients ka) p and kg) n of positive and negative converters, inner-loop and outer-loop controller parameters, and a control target reference value.

The uncontrollable terminal is only required to meet constraints on the power flow and restrictions on converters and converter stations, and does not participate in the optimization control.

3. Constraint conditions are determined.

The constraint conditions include: constraint conditions for the converters and converter stations, constraint conditions for the new energy electric field or AC network to which the converter stations are connected, and constraint conditions for a DC transmission line: 1) Meet the condition of ensuring an increase in the active power output of the controllable new energy terminal: Pac Poi iw SO Fn 12,07) where Paci ase denotes the AC-side active power of the ith terminal after optimization control (1<i<m), and Pi denotes the AC-side initial active power of the LU500700 ith terminal. 2) Satisfy a power flow equation for iterative calculation of a true bipolar DC power flow and meet certain iterative convergence conditions.

3) The active power at each terminal, a DC node voltage, and a current of the DC line that are obtained by the true bipolar DC power flow calculation need to meet constraints on upper and lower limit values.

The steady-state active power of the positive and negative converters at the controllable new energy terminal separately meets constraints on an active power limit value: fn Pan = Pyne oA SES mi le Eo Ha S Broan (US TS nm) where k denotes an active power distribution coefficient of the positive converter in the converter station at the 7th terminal, Paca) denotes AC-side initial active power of the 7th terminal, and Pyscijmar p and Prsciijmax n respectively denote active power upper limits of the positive and negative converters in the converter station at the ith terminal. The steady-state active power of positive and negative converters at the remaining controllable terminals separately meets constraints on an active power limit value: max{ik, LiPo ERR a a DS Pine » LL where Paci p and Pac n respectively denote DC-side active power of the positive and negative converters in the converter station at the ith terminal. The steady-state active power of positive and negative converters at the uncontrollable terminal separately meets the following constraints: 4, ; SEES where Pyossi)_p and Pioss() n respectively denote active power losses of the positive and negative converters in the converter station at the ith terminal. The DC voltage of converter stations at all terminals in the true bipolar flexible DC transmission system meets a constraint on a voltage limit value: DER gl, VUS CIS PS n) where L" and L respectively denote upper and lower limits of the DC voltage of the converter station at the ith terminal. LUS00700 The positive and negative DC line currents in the true bipolar flexible DC transmission system meet a DC current constraint for each DC line: Ol | SENG =m J) where li) denotes a current transmitted by a DC line between the 7th and jth nodes, and 7; denotes a current upper limit of the DC line between the ith and jth nodes.

Step 4: A contingency set is generated, which includes, but is not limited to, N-1 and N-2 contingency sets. In an abnormal operating condition, the use of a healthy pole and healthy line to transfer part of the power transmitted by a faulty pole or faulty line can improve the whole transmission power of the converter station on the premises that the DC voltage does not exceed the limit, the power of the converters and the converter station does not exceed the limit, and the DC line does not exceed a current upper limit. The abnormal operating condition includes, but is not limited to, DC disconnection or converter outage in a unipolar grid for N-1 failure, a combination of a line fault and a converter fault for N-2 failure, and a failure condition in which a converter station at a particular terminal is out of operation.

Step 5: Optimization solution of an inter-pole control scheme between the positive and negative converters is performed. An equivalent optimization model for a system failure is established according to steps 1, 2, 3, and 4, to find optimization control schemes corresponding to failures covered by the contingency set. The optimization method includes, but is not limited to, an artificial intelligence-based optimization solution or a classical numerical solution to a problem with a constrained nonlinear multivariate function. The contingency set is traversed until completion, and various found optimization schemes are stored, to generate an offline optimization scheme library.

Further, when the status of the true bipolar DC grid changes, the process enters an operating status determination module, where determination objects include the grid topology and the status of operating power flow. When the grid topology and the power flow change, a control strategy may be directly generated by using a corresponding optimization scheme in the generated offline library, or online optimization solution 1s directly conducted based on the current grid status. Alternatively, a combination of offline matching and online optimization is used. That is, first, an optimization scheme corresponding to a similar failure condition that exists in an offline library 1s immediately matched, and then the equivalent optimization model of the system 1s LUS00700 updated based on the current power flow status, grid topology, and parameters, to find an optimization scheme online. Afterwards, the generated online optimization scheme is used to replace the offline optimization scheme.

Further, when the process enters an online optimization stage, initial value conditions of the optimization model are determined according to the current power flow status, grid topology, and parameters. After the optimization model 1s established according to steps 1 to 3, optimization solution is performed for the current failure type including, but not limited to, an artificial intelligence-based optimization solution or a classical numerical optimization solution, to generate an optimization scheme. Further, the offline optimization scheme library is updated while the grid status fluctuates or the grid is faulty. Changes in the topology and power flow are determined according to the current grid status. If the topology changes, it is required to re-establish the equivalent optimization model; or if only the power flow changes, it is only required to re-initialize the equivalent optimization model. By using a power flow status after restoring from the failure to a steady state as an initial value for optimization after update, and by means of artificial intelligence-based optimization solution or classical numerical optimization solution, the contingency set is traversed to generate an optimal operating scheme corresponding to each status, and the offline optimization scheme library is updated. An inter-pole active power distribution strategy based on a true bipolar flexible DC transmission system is provided, and the following strategy 1s used: In converter stations in which positive and negative converters are independently controllable, active power distribution coefficients ks; and 1-k; are assigned respectively to the positive and negative converters in a converter station adopting P… master-slave control, where A; “po Paci p denoting the AC-side power of the positive converter at the current terminal, and ka“) meeting 0<k»<1. In this case, the optimization control quantity is the active power distribution coefficient k; of the positive converter, and positive and negative active power injections are respectively adjusted to kw Pacwy and (1-K5)Pacwy by using the optimization control scheme. For a converter station adopting droop control, the optimization control quantity includes the active power distribution coefficient ka) of the positive converter or respective droop coefficients Ky p and Ka n of the positive and negative converters.

For converter LU500700 stations adopting a hybrid control strategy in which in the same converter station, converters of one polarity use an active/reactive power coupling control manner, while converters of the other polarity use an active/reactive power decoupling control manner containing the master-slave control or droop control, the optimization control quantity includes the active power distribution coefficient k; of the positive converter, respective droop coefficients Kj; p and Ky n of the positive and negative converters, inner-loop and outer-loop controller parameters, and a control target reference value.

The converters are independently controllable under the condition that the converter power does not exceed the limit, and part of the active power of a faulty pole may be transferred by a healthy pole under the condition that a single-pole converter is faulty or out of operation, thus ensuring output of all active power of the faulty terminals as much as possible.

The present invention establishes a unified optimization model applicable to different operating conditions, different topologies, and different control manners of the system.

In a manner of combining offline and online optimization, when the system operates in different operating conditions including a failure and an abnormal condition, the model can rapidly match a corresponding optimal system operating scheme, so that the flexibility of independent controllability of the two-pole networks in a true bipolar system is utilized to a great extent, thus solving the problem that it is difficult to implement real-time optimization control when the true bipolar DC system is faulty or abnormally operates.

The method provided by the present invention can be used in different types of flexible DC transmission systems and is applicable to system structures using pseudo- bipolar and true-bipolar wiring manners and a single-pole wiring manner; and is further applicable to various control schemes, including in the true bipolar system, an active/reactive power coupling control manner or active/reactive power decoupling control manner for both positive and negative converters in the same converter station, and a hybrid control manner in which in the same converter station, converters of one polarity use the active/reactive power coupling control manner, while converters of the other polarity use the active/reactive power decoupling control manner.

By use of the proposed adaptive optimization control method, a control effect of maximizing the consumption of the new energy can be achieved in any case.

FIG. 1 shows an adaptive optimization control strategy, which can realize coordination between offline optimization and online optimization, to match an LU500700 accurate optimization scheme for the post-failure control. FIG. 2 shows an offline optimization control strategy, which is applicable to a scenario where the power flow does not change much and is relatively stable, and may result in an inaccurate optimization scheme due to the lack of the online optimization stage. FIG. 3 shows an online optimization control strategy, which is applicable to a DC system with a small scale, a non-complex topology structure and a small amount of calculation.

The adaptive optimization control strategy proposed by the present invention is specifically described by using a four-terminal true bipolar flexible DC transmission system shown in FIG. 4 as an example. The positive and negative converters in the same converter station are independently controlled, as shown in FIG. 5.

A converter station 1 and a converter station 2 respectively have rated capacities of 1500 MW and 3000 MW, are separately connected to a new energy wind farm, and both adopt a constant-active-power control manner. A converter station 4 has a rated capacity of 1500 MW, is connected to a pumped-storage power station, and uses a constant DC voltage to control and stabilize the whole grid voltage as a balanced station. A converter station 3 has a rated capacity of 3000 MW, is connected to an AC grid, and operates in a constant-active-power control manner. The rated voltage at the DC side is +500 kV. The converter station at each terminal has two converters which are respectively connected to positive and negative operating layers.

In the example, a capacity base value Sp of the converter station is set to 2000 MW and a DC-side voltage base value Us is set to 500 kV.

Under the condition that the DC system is in a stable operating status, a grid topology, grid parameters, and an operating status, including an initial voltage and active power of each terminal in an initial stage, are acquired.

Step 1: An objective function is determined for improving a consumption capacity of the new energy. According to that the capacities of the new energy wind farms connected to the converter stations 1 and 2 are respectively 1500 MW and 3000 MW with a capacity ratio of 1:2, proportion coefficients #1 and # of a new energy increment in each wind farm are respectively set to 1/3 and 2/3. Considering that there is still a certain margin for a power production of the new energy field, the objective function is determined as follows: FO = Ty SF, 5, A where calculation manners of AP; and APp are respectively LUS00700 AR =F ym Pat and ARS Pay ner 7 Factors Pact1) tas and Pace) tas denoting the active power actually injected to the DC grid after the power production at the new energy terminal increases, and Pa) and Paco, denoting the active power initially injected to the new energy terminal.

Step 2: Decision variables and the number of the variables are determined. For a true bipolar multi-terminal DC transmission system adopting a master-slave control strategy, the decision variables involved in optimization are as follows: X= er Fr Fos Key Ra] where Pac(1) last and Pac(2) 1ast denote the initial AC power of terminals of an actual flexible DC system after the power production at the new energy terminal increases; and kan, ke), and ke) denote power distribution coefficients of positive and negative converters in a converter station at each terminal in the flexible DC system.

Step 3: Constraint conditions are determined. The constraint conditions include: constraint conditions for the converters and converter stations, constraint conditions for the new energy electric field or AC network to which the converter stations are connected, and constraint conditions for DC transmission lines, where linear inequality constraints are as follows: Patty tas Pry SO LP er Pan SO where Pac(1) and Pact) denote the initial active power of the converter stations 1 and

2.

Non-linear constraints include constraint conditions for a power flow equation for iterative calculation of a true bipolar DC power flow, which specifically includes active power and voltage constraints on the converter stations and a current constraint on the DC lines.

The steady-state active power of the positive and negative converters at the new energy terminal (the terminals 1 and 2) separately meets the following constraints: (hPa S Racine =1,2) LUKE Bocuse = 1,2) The steady-state active power of the positive and negative converters at the terminal 3 and a balanced station terminal (the terminal 4) separately meets the following constraints: LU500700 | mast, Pot ono 18000 0S Bacon» 04 {maxi Pari) Pag a Pa 0S Pei « A) The converter stations at four terminals all meet a DC voltage constraint: DSL, SUN 34 A DC line current in the positive and negative networks meets the following constraint: OS pen IS IC 234) Step 4: A common contingency set, including but not limited to, N-1 and N-2 contingency sets, is generated. In an abnormal operating condition, the use of a healthy pole and healthy line to transfer part of the power of a faulty pole or faulty line can improve the whole transmission power of the converter station on the premises that the DC voltage does not exceed the limit, the power of the converters and the converter station does not exceed the limit, and the DC line does not exceed a current upper limit of a DC circuit breaker. The abnormal operating condition includes: DC disconnection or converter outage in a unipolar grid for N-1 failure, a combination of a line fault and a converter fault for N-2 failure, and a failure condition in which a converter station at a particular terminal is out of operation.

Step 5: Optimization solution of an inter-pole control scheme between the positive and negative converters 1s performed. Optimization schemes corresponding to failures covered by the contingency set are found. The optimization method includes, but is not limited to, an artificial intelligence-based optimization solution or a classical numerical solution to a problem with a constrained nonlinear multivariate function. Optimization results of respective decision variables and a value of the objective function, namely, a corresponding optimization control scheme, are obtained. The contingency set 1s traversed until completion, and various found optimization schemes are stored, to generate an offline optimization scheme library.

Further, when the status of the true bipolar DC grid changes, the process enters an operating status determination module, where determination objects include the grid topology and the status of operating power flow. Assuming that a DC disconnection fault between converter stations 3 and 4 of a negative network occurs in the DC grid, the network topology and the power flow both change accordingly: an output of the new energy electric field connected to the converter station 1 raises from 0.5 p.u. to 0.6 LU500700 p.u. and a power flow variation exceeds an allowable fluctuation range. In this case, first, a corresponding optimization scheme or an optimization scheme for a similar failure in the generated offline library is directly matched, and also, online optimization solution is directly conducted based on the current grid status, to generate an online optimization scheme. Then, the generated online optimization scheme is immediately used to replace the original scheme. Results of the offline and online optimization schemes are shown in table 1. Table 1 Results of the offline and online optimization schemes optimization optimization | scheme 0.900p.u. | -1.000p.u. 1.242p.u. | 0.529 10m = (Pan tas 7 Pu) Paz me Pr) 0.324 pa Jol) = Boy se = Pad (Poy ras 7 Pa FO. A22 pas, Herein, fi(X)max and f(<)max denote maximum values of the objective function obtained by using the offline and online optimization schemes respectively. The optimization results indicate that, when the adaptive optimization strategy for a failure is used in a DC transmission system connected to the new energy wind farm, the scheme can flexibly deal with output fluctuation of the new energy, and can rapidly restore the grid to a stable operating status by using the feature of independent controllability of the positive and negative converters in the case of a failure. The results in table 1 show that, a maximum consumption capacity of an increment at the new energy terminal that is obtained by using the offline scheme is 0.324 x 2000 = 648 MW. That is, under the condition that none of the converter voltage, power, and DC line current exceed the limit in a steady state, an active power increment of 648 MW of the connected wind farm can be consumed to a great extent. Herein, the positive active power distribution coefficients of the converter stations 1, 2, and 3 are respectively

0.520, 0.621, and 0.652. At a time point when the failure actually occurs, the active power injected at the AC side of the converter station 2 fluctuates and decreases from

0.6 p.u. to 0.5 p.u. Based on online optimization for the power flow at this time point, the following result is obtained: A maximum consumption capacity of the increment at the new energy terminal is 0.422 x 2000=844 MW, where the positive active power distribution coefficients of the converter stations 1, 2, and 3 are respectively 0.529, LU500700

0.587, and 0.620.

The optimization results show that, the inter-pole cooperative control strategy designed between the positive and negative converters fully utilizes the feature of independent controllability of the positive and negative converters in the true bipolar DC system, and can coordinate specific active power distribution between the two poles according to a new energy consumption requirement and an operating condition of the system, and can use a healthy pole and healthy line to actively bear part of the transmission power of a faulty pole and faulty line in an abnormal operating condition, so as to avoid excess transmission power of the faulty pole, thus enhancing the flexibility and reliability of the bipolar system.

Herein, in the case of a failure, the offline optimization library enables the faulty grid to restore to a suboptimal operating status by rapidly matching a closest optimization scheme, and the online optimization scheme enables the faulty grid to restore to an optimal operating status by matching an accurate optimization scheme.

The technical means disclosed in the solution of the present invention are not limited to the technical means disclosed in the foregoing detailed description, but also include technical solutions composed of any combination of the foregoing technical features. It should be noted that, several improvements and modifications can be made by those of ordinary skill in the art without departing from the principle of the present invention, and these improvements and modifications should also be construed as falling within the protection scope of the present invention.

Claims (4)

CLAIMS LU500700

1. An adaptive optimization control method for a failure in a true bipolar flexible DC transmission system, comprising the following steps: step 1: under the condition that the true bipolar flexible DC transmission system is in a stable operating status, acquiring a grid topology, grid parameters, and an operating status in real time, wherein the operating status comprises an initial voltage and active power of each terminal in an initial stage; step 2: establishing an equivalent optimization model of the true bipolar flexible DC transmission system, wherein the true bipolar flexible DC transmission system has n terminals in total which are successively numbered as 1, 2, ..., m, m+1, ..., q, g+1, ..., n, the ith (7=1, 2, ..., q) terminal being a controllable terminal, the ith (i=1, 2, ..., m) terminal being a controllable new energy terminal, and the ith (i=¢+1, ..., n) terminal being an uncontrollable terminal, g<n, and m<q; an objective function of the equivalent optimization model is selected according to an actual transmission capacity of a converter station, specifically: 1) when the actual transmission power of the controllable new energy terminal does not reach the upper limit of the converter station capacity, the following objective function is selected: Ada = 1 AR + i AR Pen, M, (1=1,2,.,m) being a weight coefficient of an active power increment AP of the new energy at the ith terminal, and /=1,2,...,m; and 2) when the actual transmission power of the controllable new energy terminal reaches the upper limit of the converter station capacity, the following objective function is selected: rn = Floss ton +P an FE a , Paci) last (i=1,2,...,m) denoting AC- side active power of the ith terminal after optimization control; and constraint conditions for the equivalent optimization model comprise: Pas Lew $0(i=120m) is met when the objective function in the case 1) is selected; Pan nes Pas S8{i=1.2,.,m} is met when the objective function in the case 2) is selected;

Kine = Pisctsmex p (i = 1,2... 77) (1 TT ki, VE = Dean { i = LZ. m) max(k Pat! ; | Prior i + Pie } = Pistes pe {i = ml Qu. q) max Men)? | Fotis » : Pet [rd } = Piscines, it {i = pre, 2,4) Pan pt Poin = Fise timp {i = qi, 2,0, n) Pe _ | + Flan ” = Bicone (i = gH 2, ns n) Us = Ug 2 = Sh {i = I, 2, sha A) 0 Li] SY al (WIRE wherein Pac) denotes AC-side initial active power of the ith terminal, k; denotes an active power distribution coefficient of a positive converter in the converter station at the 7th terminal of the system, Pysc@max p and Prscijmac n respectively denote active power upper limits of positive and negative converters in the converter station at the ith terminal, Pas p and Pdcñ) n respectively denote DC-side active power of the positive and negative converters in the converter station at the ith terminal, P/oss(y p and Pross() n respectively denote active power losses of the positive and negative converters in the converter station at the ith terminal, Us; denotes a DC voltage of the converter station at the ith terminal, LE and US respectively denote upper and lower limits of the DC voltage of the converter station at the ith terminal, lu.) denotes a current transmitted by a DC line between converter stations at the ith and jth terminals, and 1} denotes an upper limit of the current transmitted by the DC line between the converter stations at the ith and jth terminals; step 3: initializing the equivalent optimization model established in step 2 by using the acquired grid topology, grid parameters, and operating status as initial values; step 4: generating a contingency set according to historical failure data of the true bipolar flexible DC transmission system; step 5: finding optimization control schemes corresponding to all contingencies in the contingency set in step 3 according to the equivalent optimization model established in HUS00700 step 2, to form an offline optimization scheme library; and step 6: when the status of the true bipolar flexible DC transmission system changes, proceeding to the operating status determination: if a current grid topology or power flow is inconsistent with an initial grid topology or power flow, there are the following three handling methods: 1) immediately matching and applying the closest corresponding optimization control scheme in the offline optimization scheme library; 2) conducting online optimization based on the current grid status, to find a corresponding optimization control scheme in real time; and 3) immediately matching and applying the closest corresponding optimization control scheme in the offline optimization scheme library; then conducting online optimization based on the current grid status and parameters, to find a corresponding optimization control scheme in real time; and replacing the matched optimization control scheme with the optimization control scheme found in real time.

2 The adaptive optimization control method for a failure in a true bipolar flexible DC transmission system of claim 1, wherein the power flow status in step 1 comprises an initial voltage and active power of each node.

3. The adaptive optimization control method for a failure in a true bipolar flexible DC transmission system of claim 1, wherein the method further comprises: updating the offline optimization scheme library while the grid topology and power flow have a change: if the grid topology changes, updating the equivalent optimization model; and if only the power flow changes, by using a power flow status after restoring from the failure to a steady state as an initial value of the updated equivalent optimization model, finding optimization control schemes corresponding to all contingencies in the contingency set once again, to update the offline optimization scheme library.

4. The adaptive optimization control method for a failure in a true bipolar flexible DC transmission system of claim 1, wherein when the optimization control scheme is applied in step 6, the active power of the positive and negative converters is tuned in the following manner so as to realize optimization control: in converter stations in which positive and negative converters are independently controllable, respectively assigning active power distribution coefficients ka) and 1-4 HUS00700 to the positive and negative converters in a converter station adopting master-slave Petro control, where A, Th Paci p denoting the AC-side power of the positive ot converter at the 7th terminal, and Æ meeting 0< ka) <1, in which case, an optimization control quantity is the active power distribution coefficient ki) of the positive converter, and positive and negative active power injections are respectively adjusted to An Pacd) and (1-Æ#)Pacwy by using the optimization control scheme; for a converter station adopting droop control, the optimization control quantity comprises the active power distribution coefficient ks of the positive converter or respective droop coefficients 10g, p and Kj n of the positive and negative converters; and for converter stations adopting a hybrid control strategy in which in the same converter station, converters of one polarity use an active/reactive power coupling control manner, while converters of the other polarity use an active/reactive power decoupling control manner containing the master-slave control or droop control, the optimization control quantity comprises the active power distribution coefficient ka) of the positive converter, respective droop coefficients Ka), and Ku) n of the positive and negative converters, inner-loop and outer-loop controller parameters, and a control target reference value.