Multi-embedded direct current-containing power grid continuous time scheduling optimization method, storage medium and program product

Abstract

The invention belongs to the related technical field of electrical engineering, and discloses a continuous time scheduling optimization method for a power grid containing multiple embedded direct currents, a storage medium and a program product, and the method comprises the steps: building a continuous time power grid scheduling original optimization model containing the multiple embedded direct currents; representing a continuous time variable as a Bernstein polynomial, and converting a relational expression in the original optimization model into a relational expression about a Bernstein polynomial coefficient according to a conversion strategy to realize model reconstruction; and solving the reconstruction model, and substituting a coefficient obtained by solving into a corresponding Bernstein polynomial to obtain a final scheduling result of a corresponding continuous time variable. According to the method, the properties of the Bernstein polynomial are analyzed, the conversion strategy of the nonlinear relational expression of the Bernstein polynomial is obtained, the model is solved, and the continuous time scheduling curve is finally output, so that the power system scheduling is promoted to cope with the net load fluctuation brought by new energy, and the safety and economy of the power system are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field related to electrical engineering, and more specifically, relates to a method for optimizing continuous time scheduling of a power grid containing multiple embedded direct currents, a storage medium and a program product. Background Art

[0002] As the penetration rate of new energy represented by wind power and photovoltaic power in traditional power systems continues to increase, the volatility of net load of power systems within hours becomes more and more obvious, which brings huge challenges to the power and electricity balance of power systems. Specifically, the power system will experience large power fluctuations, which will affect the safe and stable operation of the power grid. At the same time, with the continuous increase in the penetration rate of power electronic devices in power systems, traditional power systems are gradually developing from traditional pure AC systems to AC / DC hybrid power grids containing multiple embedded DC. The multi-embedded DC system can not only promote the interconnection of cross-regional power systems, but also effectively promote the consumption of new energy. Compared with traditional AC transmission, the embedded DC system has the following advantages: first, the embedded DC system has low transmission loss; second, the embedded DC system has more flexible and rapid power control; third, the embedded DC system has a large transmission capacity. The large-scale grid connection of new energy has put forward higher requirements on the line transmission capacity, and the transmission capacity of the embedded DC system is higher than that of the traditional AC system, realizing the effective transfer of large-scale new energy.

[0003] However, the existing grid dispatching methods with multiple embedded DC are often based on discrete time models, which cannot fully utilize the fast response characteristics of the multiple embedded DC system to smooth the net load fluctuation within the hour. With the continuous increase in the penetration rate of new energy, the existing grid dispatching methods with multiple embedded DC will not be able to ensure the power balance of the power system, endangering the safety of the system. Summary of the invention

[0004] In response to the above defects or improvement needs of the prior art, the present invention provides a continuous-time scheduling optimization method, storage medium and program product for a power grid containing multiple embedded DCs, which can establish a continuous-time model, fully tap the flexibility of the multi-embedded DC system, and promote the power system scheduling to cope with the net load fluctuations caused by new energy, thereby improving the safety and economy of the power system.

[0005] To achieve the above object, according to one aspect of the present invention, a method for optimizing continuous time scheduling of a power grid containing multiple embedded DC is provided, which comprises:

[0006] Obtain the grid system parameters of the grid to be dispatched, take the minimization of dispatching cost within the dispatching period as the objective function, set the operation constraints, and establish the original optimization model of continuous-time grid dispatching with multiple embedded DC;

[0007] The original optimization model is subjected to a solution space transformation to transform the continuous time variables into discrete variables to obtain a reconstructed optimization model; the solution space transformation comprises: dividing the scheduling period into a plurality of time periods, expressing the continuous time variables in the original optimization model as n-order Bernstein polynomials and transforming the relational expressions in the original optimization model into relational expressions about Bernstein polynomial coefficients according to a transformation strategy; n is an integer greater than 2;

[0008] Solving the reconstruction optimization model to obtain a first solution result of the discrete variable, substituting the coefficients of the Bernstein polynomial in the first solution result into the Bernstein polynomial of the corresponding continuous time variable to obtain a final scheduling result of the corresponding continuous time variable;

[0009] The conversion strategy includes:

[0010] Convert the continuous time variable integral value, continuous time variable differential value and continuous time variable in linear relationship into variables related to Bernstein polynomial coefficients;

[0011] The form is (P1(τ)) 2 +(P2(τ)) 2 The constraint ≤P3(τ)P4(τ) is transformed into The form is (P1(τ)) 2 +(P2(τ)) 2 ≤(C) 2 The constraint is converted to

[0012] Where τ is the time variable in each time period, P1(τ), P2(τ), P3(τ) and P4(τ) are the Bernstein polynomials corresponding to the continuous time variables. They are the Bernstein polynomials P m The coefficients of the pth basis function and the qth basis function in (τ), m = 1, 2, 3, 4, p = 0, 1, ..., n, q = 0, 1, ..., n, C is a constant.

[0013] Preferably, the objective function in the original optimization model is:

[0014]

[0015] C su，i (τ′)=c su，i S u，i (τ′), C sd，i (τ′)=c sd，i S d，i (τ′), C f，i (τ′)=C f，i Pg，i (τ′)

[0016] Where H is the total duration of the dispatch cycle, τ′ is the time variable within the dispatch cycle, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the grid node number, c su 、c sd They represent the single startup cost and single shutdown cost of thermal power units, respectively, and c f represents the fuel cost coefficient, S u Indicates the variable indicating the power-on action. If it is 1, it means the power-on action is executed. Otherwise, it is 0. d Indicates the variable indicating the shutdown action. 1 indicates the shutdown action is executed, otherwise it is 0. Each period can execute the startup or shutdown action at most once at the beginning of the period. S u and S d is a discrete variable, P g Indicates the active power output of thermal power units;

[0017] The reconstructed objective function is:

[0018]

[0019] Where t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients.

[0020] Preferably, the operating constraints in the original optimization model include the following thermal power unit constraints:

[0021] Thermal power unit start and stop variable association constraints:

[0022] S u，i (τ′)-S d，i (τ′)=I g，i (τ′ + )-I g，i (τ′ - )

[0023] S u，i (τ′)+S d，i (τ′)≤1

[0024] Minimum start and stop time constraints for thermal power units:

[0025]

[0026]

[0027] Active power output constraints of thermal power units:

[0028]

[0029] Reactive power output constraints of thermal power units:

[0030]

[0031] Thermal power unit climbing constraints:

[0032] dP g，i (τ′) / dτ′≤r u，i I g，i (τ′ - )+r su，i S u，i (τ′)

[0033] -dP g，i (τ′) / dτ′≤r d，i I g，i (τ ′+ )+r sd，i S d，i (τ′)

[0034] Where τ′ is the time variable in the dispatching period, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the number of the grid node, S u Indicates the variable indicating the power-on action. 1 indicates the power-on action, otherwise 0. S d Indicates the variable indicating the shutdown action. 1 indicates the shutdown action, otherwise it is 0. Set each time period to execute the startup or shutdown action at most once at the beginning of the time period. g Indicates the status of the thermal power unit. When it is 1, it means it is online, and when it is 0, it means it is offline. u , S d ,I g is a discrete variable, τ′ + and τ′ - Respectively represent the moment after and before the moment τ′, t on and t off are the minimum online and offline state durations, respectively, g Indicates the active power output of the thermal power unit, and P g are the maximum and minimum active output of the thermal power unit, Q g Indicates the reactive power output of thermal power units. and Q g are the maximum and minimum reactive power outputs of thermal power units, r u and r d are the upward and downward climbing speed limits of thermal power units, rsu and r sd They are respectively the ramp limits during the startup and shutdown processes of thermal power units;

[0035] The reconstruction optimization model includes the following reconstructed thermal power unit constraints:

[0036] The associated constraints of the start and stop variables of the thermal power unit after reconstruction:

[0037] S u，i，t -S d，i，t =I g，i，(t+1) -I g，i，t

[0038] S u，i，t +S d，i，t ≤1

[0039] After reconstruction, the minimum start and stop time constraints of the unit are:

[0040]

[0041] Active power output constraints of thermal power units after reconstruction:

[0042]

[0043] Reactive power output constraints of thermal power units after reconstruction:

[0044]

[0045] The reconstructed thermal power unit climbing constraints:

[0046]

[0047] Where t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients, W n P g The actual coefficient matrix of the differential terms of the corresponding Bernstein polynomial during n-order BP interpolation;

[0048] Contains n+1 elements: When t is not the last period: If n is an even number, the first The element is I g，i，t ,back The element is I g，i，(t+1) , if n is an odd number, I g，i，t and I g，i，(t+1) The number of When t is the last period, Contains n+1 I g，i，t.

[0049] Preferably, the operating constraints in the original optimization model include the following AC power grid constraints:

[0050] Active power balance constraints:

[0051]

[0052] Reactive power balance constraints:

[0053]

[0054] AC active power network loss constraints:

[0055] R lAC，j ((P rAC，j (τ′) 2 +(Q rAC，j (τ′) 2 )≤P lsAC，j (τ′)W rAC，j (τ′)

[0056] Active network loss and reactive network loss association constraints:

[0057] X l，j P lsAC，j (τ′)=R lAC，j Q lsAC，j (τ′)

[0058] AC voltage amplitude constraint:

[0059] W sAC，j (τ′)-W rAC，j (τ′)

[0060] =2R lAC，j P rAC，j (τ′)+2X l，j Q rAC，j (τ′)+R lAC，j P lsAC，j (τ′)

[0061] +X l，j Q lsAC，j (τ′)

[0062] AC voltage phase angle constraint:

[0063] θ sAC，j (τ′)-θ rAC，j (τ′) = X l，j P rAC，j (τ′)-R lAC，j Q rAC，j (τ′)

[0064] Where τ′ is the time variable in the dispatching period, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the grid node number, j is the line number, N lAC is the number of AC lines, P g Indicates the active power output of the thermal power unit, Q g Represents the reactive power output of the thermal power unit, P wd Represents the predicted active power of new energy, P ld and Q ld Represent the predicted load active power and reactive power, P cv and Q cv Represent the active power and reactive power of the converter station respectively, P rAC and Q rAC Respectively represent the active power and reactive power at the receiving end of the AC line, P lsAC and Q lsAC They represent the active power and reactive power lost in the AC line, B represents the parallel susceptance parameter of the AC line, and W AC Represents the square of the AC node voltage, W rAC and W sAC Represents the square of the voltage at the receiving end and the voltage at the sending end of the AC line, R lAC and X1 represent the AC line resistance and reactance parameters respectively, θ sAC and θ rAC Indicates the phase angle of the AC line sending end and receiving end, M PQAC and M lAC They are the correlation matrix of AC line power flow and the correlation matrix of AC line loss respectively. The specific calculation expressions are:

[0065]

[0066]

[0067] The reconstruction optimization model includes the following reconstructed AC power grid constraints:

[0068] Reconstructed active power balance constraint:

[0069]

[0070] Reconstructed reactive power balance constraint:

[0071]

[0072] The AC active power network loss constraints after reconstruction are:

[0073]

[0074] The associated constraints of active network loss and reactive network loss after reconstruction are:

[0075]

[0076] The reconstructed AC voltage amplitude constraint:

[0077]

[0078] The reconstructed AC voltage phase angle constraint:

[0079]

[0080] Where t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients, containing the elements or and They represent the coefficients of the p-th and q-th terms in the Bernstein polynomial respectively.

[0081] Preferably, the operating constraints in the original optimization model include the following DC grid constraints:

[0082] DC power balance constraints:

[0083]

[0084] DC active power loss constraint:

[0085] R lDC，j (P rDC，j (τ′) 2 ≤P lsDC，j (τ′)W rDC，j (τ′)

[0086] DC voltage amplitude constraint:

[0087] W sDC，j (τ′)-W rDC，j (τ′)=2R lDC，j P rDC，j (τ′)+R lDC，j P lsDC，j (τ′)

[0088] Where τ′ is the time variable in the dispatching period, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the grid node number, j is the line number, N lDC Indicates the number of DC lines, P DC represents the DC node injection power, P rDCRepresents the active power at the receiving end of the DC line, P lsDC Indicates the active power lost in the DC line, W rDC and W sDC Respectively represent the square of the voltage at the receiving end and the voltage at the sending end of the DC line, R lDC Indicates the DC line resistance parameter, M PQDC and M lDC They are the correlation matrix of DC line power flow and the correlation matrix of DC line loss respectively. The specific expressions are calculated as follows:

[0089]

[0090] The reconstruction optimization model includes the following reconstructed DC grid constraints:

[0091] The DC power balance constraint after reconstruction:

[0092]

[0093] Reconstructed DC active power loss constraint:

[0094]

[0095] The reconstructed DC voltage amplitude constraint:

[0096]

[0097] Where t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients, containing the elements or and They represent the coefficients of the p-th and q-th terms in the Bernstein polynomial respectively.

[0098] Preferably, the operation constraints in the original optimization model include the following converter station constraints:

[0099] Power balance constraints of converter stations:

[0100] P cv，i (τ′)=P DC，i (τ′)-P lscv，i (τ′)

[0101] Active power loss constraints of converter stations:

[0102] P lscv，i (τ′)=β|P cv，i (τ′)|

[0103] Converter station power capacity constraints:

[0104] (P cv，i (τ′) 2 +(Q cv，i (τ′) 2 ≤(S cv，i ) 2

[0105] Where τ′ is the time variable in the dispatching period, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the number of the grid node, P cv and Q cv Represent the active power and reactive power of the converter station respectively, P DC represents the DC node injection power, β represents the converter station loss coefficient, P lscv Indicates the active power loss of the converter station, S cv Indicates the power capacity of the converter station;

[0106] The reconstruction optimization model includes the following reconstructed converter station constraints:

[0107] The power balance constraints of the converter station after reconstruction are:

[0108]

[0109] Active power loss constraints of converter stations after reconstruction:

[0110]

[0111] The power capacity constraint of the converter station after reconstruction:

[0112]

[0113] Where t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients, containing the elements Represents the coefficient of the p-th term in the Bernstein polynomial.

[0114] Preferably, the operating constraints in the original optimization model include the following upper and lower limit constraints on variables:

[0115]

[0116] |P rAC，j (τ′)+P lsAC，j (τ′)|≤F AC，j

[0117] |PrAC，j (τ′)|≤F AC，j

[0118] |Q rAC，j (τ′)+Q lsAC，j (τ′)|≤F AC，j

[0119] |Q rAC，j (τ′)|≤F AC，j

[0120]

[0121] |P rDC，j (τ′)+P lsDC，j (τ′)|≤F D C ，j

[0122] |P rDC，j (τ′)|≤F DC，j

[0123] Where τ′ is the time variable in the dispatching period, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the grid node number, j is the line number, and W AC represents the square of the AC node voltage, and W AC Respectively represent the upper and lower limits of the square of the AC node voltage, θ AC represents the AC node phase angle, and θ AC Respectively represent the upper and lower limits of the AC node phase angle, and P rAC and Q rAC Respectively represent the active power and reactive power at the receiving end of the AC line, P lsAC and Q lsAC Respectively represent the active power and reactive power lost in the AC line, F AC Represents the AC branch transmission capacity, W DC represents the square of the DC node voltage, and W DC They represent the upper and lower limits of the DC node voltage square, P rDC Represents the active power at the receiving end of the DC line, P lsDC Indicates the active power lost in the DC line, F DC Indicates the DC branch transmission capacity;

[0124] The reconstructed optimization model includes the following reconstructed upper and lower limit constraints of the following variables:

[0125]

[0126]

[0127] Where t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients.

[0128] Preferably, the continuous time variables in the original optimization model are represented as cubic Bernstein polynomials, n=3.

[0129] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0130] According to another aspect of the present invention, there is provided a computer program product, comprising a computer program or instructions, wherein the computer program or instructions implement the steps of any of the above methods when executed by a processor.

[0131] In general, compared with the prior art, the above technical solutions conceived by the present invention mainly have the following beneficial effects:

[0132] Since the constraints in the continuous-time power grid dispatch optimization model containing multiple embedded DC often include second-order cone programming relations, that is, they have quadratic nonlinear terms, it is difficult to solve the model using conventional methods. In the present invention, by using Bernstein polynomials to approximate continuous-time variables and analyzing the properties of Bernstein polynomials, a conversion strategy for nonlinear relations of Bernstein polynomials is obtained, and the first-order and second-order relations containing continuous-time variables are converted into relations about Bernstein polynomial coefficients using the conversion strategy, thereby realizing solution space transformation, converting the problem of solving continuous-time variables into the problem of solving discrete variables, and obtaining a reconstructed optimization model, thereby reducing the difficulty of solving. After solving the reconstructed optimization model, the Bernstein polynomial coefficients obtained by solving are again substituted into the Bernstein polynomial of the corresponding time variable to obtain the final dispatch result in its continuous-time form, thereby guiding the power grid dispatcher to dispatch power resources according to the dispatch result. The present invention schedules the power system containing multiple embedded DC by establishing a continuous time scheduling optimization model. Since the final output is a continuous time scheduling curve, it can fully tap the intra-hour flexibility of the multi-embedded DC system, promote the power system scheduling to cope with the net load fluctuations caused by new energy, thereby improving the safety and economy of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0133] Figure 1It is a flowchart of the steps of a method for optimizing continuous time scheduling of a power grid containing multiple embedded DC in one embodiment of the present invention. DETAILED DESCRIPTION

[0134] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0135] Example 1

[0136] like Figure 1 The figure shows a flow chart of the steps of a method for optimizing continuous time scheduling of a power grid containing multiple embedded DC in an embodiment of the present invention, which mainly includes steps S1 to S4. Each step is described below.

[0137] Step S1: Obtain the grid system parameters of the grid to be dispatched.

[0138] Specifically, it is necessary to obtain the AC line parameters, multi-embedded DC line parameters, converter station parameters, thermal power unit parameters, new energy output data, load data of the power grid to be dispatched, and economic parameters.

[0139] Specifically, the acquired parameters and data of the power grid containing multiple embedded DC to be dispatched include:

[0140] (1) Thermal power unit parameters: upper and lower limits of active output P g , reactive power upper and lower limits Q g , the maximum up and down climbing power during operation r u 、r d , maximum climbing power when starting and stopping su 、r sd , minimum startup and shutdown time t on ,t off , single start-up and shutdown cost c su 、c sd , fuel cost coefficient c f .

[0141] (2) AC system parameters: Number of AC system nodes N bAC , the number of AC system lines N lAC , predict the new energy active power P wd , predict the load active power P ld and reactive power Q ld, AC line parallel susceptance parameter B, AC line power flow correlation matrix parameter M PQAC , the correlation matrix parameter M of AC line loss lAC , AC line resistance parameter R lAC and reactance parameter X l , the upper limit of the square of the AC node voltage and lower limit W AC , the upper limit of the AC node phase angle and lower limit W AC , AC branch transmission capacity F AC .

[0142] (3) DC system parameters: Number of DC system nodes N bDC , the number of DC system lines N lDC , DC line resistance parameter R lDC , the correlation matrix parameter M of the DC line power flow PQDC , the correlation matrix parameter M of DC line loss lDC , the upper and lower limits of the square of the DC node voltage and W DC , DC branch transmission capacity F DC .

[0143] (4) Converter station parameters: converter station loss coefficient β, converter station power capacity S cv .

[0144] Step S2: Taking the minimization of dispatching cost within the dispatching period as the objective function and setting the operating constraints, a continuous-time grid dispatching primitive optimization model with multiple embedded DC is established.

[0145] The objective function is to minimize the scheduling cost within the scheduling period. In one embodiment, the objective function can be expressed as:

[0146]

[0147] C su，i (τ′)=c su，i S u，i (τ′), C sd，i (τ′)=c sd，i S d，i (τ′), C f，i (τ′)=c f，i P g，i (τ′)

[0148] Where H is the total duration of the dispatch cycle, τ′ is the time variable within the dispatch cycle, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the grid node number, c su , C sd They represent the single startup cost and single shutdown cost of thermal power units, respectively, and c f represents the fuel cost coefficient, S u Indicates the variable indicating the power-on action. If it is 1, it means the power-on action is executed. Otherwise, it is 0. d Indicates the variable indicating the shutdown action. 1 indicates the shutdown action is executed, otherwise it is 0. Each period can execute the startup or shutdown action at most once at the beginning of the period. S u and S d is a discrete variable, P g Indicates the active power output of the thermal power unit.

[0149] Specifically, the constraints generally include thermal power unit constraints, AC power grid constraints, DC power grid constraints, converter station constraints, and upper and lower limit constraints of variables.

[0150] In one embodiment, the thermal power unit constraints include unit start-stop variable association constraints, unit minimum start-stop time constraints, unit active output constraints, unit reactive output constraints, and unit climbing constraints.

[0151] In one embodiment, the AC power grid constraints include active power balance constraints, reactive power balance constraints, AC active network loss constraints, active network loss and reactive network loss association constraints, AC voltage amplitude constraints, and AC voltage phase angle constraints.

[0152] In one embodiment, the DC grid constraints include a DC power balance constraint, a DC active power loss constraint, and a DC voltage amplitude constraint.

[0153] In one embodiment, the converter station constraints include converter station power balance constraints, converter station active power loss constraints, and converter station power capacity constraints.

[0154] The specific expression form of each constraint can be set according to the actual situation of the power grid.

[0155] Step S3: performing a solution space transformation on the original optimization model to convert continuous time variables into discrete variables to obtain a reconstructed optimization model; performing the solution space transformation includes: dividing the scheduling cycle into multiple time periods, expressing the continuous time variables in the original optimization model as n-order Bernstein polynomials and converting the relationship in the original optimization model into a relationship about the Bernstein polynomial coefficients according to the conversion strategy.

[0156] This step uses Bernstein orthogonal polynomials to reconstruct the continuous time scheduling optimization model. Specifically, the scheduling period H is divided into multiple time periods. For example, if the scheduling period is 24 hours, it can be divided into 24 time periods, and the length of each time period is 1 hour. At this time, t = 1, 2, 3, ..., 24. After using Bernstein polynomials to approximate each continuous time variable curve of each time period t, the relationship is converted into the relationship of Bernstein polynomial coefficients, so as to transform the solution space of the continuous time power grid optimization model containing multiple embedded DC, and convert the continuous time optimization problem into a discrete optimization problem of Bernstein polynomial control points, which is convenient for model solution.

[0157] Assuming that the continuous curve of a variable in a time period is P(τ), then

[0158]

[0159] In the formula, τ is the time variable in each period, B n .k(τ) is the kth basis function in the nth-order Bernstein polynomial, P B，k Corresponding to B n，k The coefficient of (τ), P B and B n (τ) are P B，k and B n，k (τ). Among them, the basis function

[0160] In one embodiment, n=3, that is, a cubic Bernstein polynomial is used to approximate each continuous time variable curve in each time period t. The cubic Bernstein polynomial can better approximate the continuous time variable curve and has a low computational complexity. At this time, B 3，k (τ) is the kth basis function, P B，k Corresponding to B 3，k The polynomial coefficients of (τ), P B and B3(τ) are P B，k and B 3，k (τ). Among them,

[0161] The conversion strategies include:

[0162] Convert the continuous time variable integral value, continuous time variable differential value and continuous time variable in the linear relationship into variables related to Bernstein polynomial coefficients, including:

[0163] 1) Replace the continuous time variables in the linear relationship of the continuous time variables with their Bernstein polynomial coefficient vectors.

[0164] Assume that there is a linear inequality constraint ∑a m P m (τ)≤c,P m (τ) is the mth continuous time variable in the inequality, a m As coefficients, the inequality can be transformed into: ∑a m P m B，k ≤c,P m B，k P m The coefficients of the kth basis function of (τ);

[0165] Assume there is an equality constraint ∑a m P m (τ)=0, then the inequality can be transformed into: ∑a m P m B，k =c.

[0166] 2) Replace the continuous time variable integral value with the sum of its Bernstein polynomial coefficients divided by (n+1).

[0167] Assume that there is an integral term in the model You can replace it with:

[0168] 3) Replace the differential value of the continuous time variable with the coefficients obtained by differentiating its Bernstein polynomial.

[0169] Assuming that there is a differential term dP(τ) / dτ in the model, it can be replaced by: dP(τ) / dτ=(W n P B ) T .

[0170] The existing patent "CN1 17498447A A continuous time source grid load storage coordination optimization method considering flexible electrolytic aluminum" discloses the design of a linear constraint scheduling model and uses the first three conversion strategies to convert the solution space. However, the present invention constructs a continuous time scheduling model for a power grid containing multiple embedded DC, which contains nonlinear second-order cone programming relationships. The second-order cone programming relationship cannot be converted based on the first three conversion strategies. Through analysis and derivation, the present invention proposes a conversion strategy for the second-order cone programming relationship and applies it to the continuous time scheduling model of a power grid containing multiple embedded DC, which can realize the transformation of the model solution space and finally realize the solution of the model.

[0171] The transformation strategies of the second-order cone programming relation include:

[0172] 4) Assume there is an inequality constraint (P1(τ)) 2 +(P2(τ)) 2 ≤P3(τ)P4(τ), it can be replaced by:

[0173] 5) Assume there is an inequality constraint (P1(τ)) 2 +(P2(τ)) 2 ≤(C) 2 , which can be expressed as: C is a constant.

[0174] The proof process of the fourth strategy is as follows:

[0175] According to the definition of Bernstein polynomial, we get the following relationship:

[0176]

[0177] According to the properties of Bernstein's first point, we can further obtain the following relationship:

[0178]

[0179] According to the Cauchy-Schwarz inequality, we can get:

[0180]

[0181] Combining (1) and (2), we can obtain:

[0182]

[0183] The proof process of the fifth strategy is as follows:

[0184]

[0185] According to the properties of Bernstein's first point, we get:

[0186]

[0187] At the same time, we can also get:

[0188]

[0189] Combining (3) and (4), we get:

[0190]

[0191] Step S4: Solve the reconstruction optimization model to obtain a first solution result of the discrete variable, substitute the coefficients of the Bernstein polynomial in the first solution result into the Bernstein polynomial of the corresponding continuous time variable, and obtain the final scheduling result of the corresponding continuous time variable.

[0192] In one embodiment, the final dispatch result includes the output plan of the thermal power units, the AC system power flow results, the DC system power flow results and the converter station output plan.

[0193] The following is an example of the specific design and reconstruction results in the model.

[0194] In one embodiment, the objective function in the original optimization model is:

[0195]

[0196] C su，i (τ′)=c su，i S u，i (τ′), C sd，i (τ′)=c sd，i S d，i (τ′), C f，i (τ′)=c f，i P g，i (τ′)

[0197] Where H is the total duration of the dispatch cycle, τ′ is the time variable within the dispatch cycle, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the grid node number, c su 、c sd They represent the single startup cost and single shutdown cost of thermal power units, respectively, and c f represents the fuel cost coefficient, S u Indicates the variable indicating the power-on action. If it is 1, it means the power-on action is executed. Otherwise, it is 0. d Indicates the variable indicating the shutdown action. 1 indicates the shutdown action is executed, otherwise it is 0. Each period can execute the startup or shutdown action at most once at the beginning of the period. S u and S d is a discrete variable, P g Indicates the active power output of the thermal power unit.

[0198] In one embodiment, the operating constraints in the original optimization model include the following thermal power unit constraints:

[0199] Thermal power unit start and stop variable association constraints:

[0200] S u，i (τ′)-S d，i (τ′)=I g，i (τ′+ )-I g，i (τ′ - )

[0201] S u，i (τ′)+S d，i (τ′)≤1

[0202] Minimum start and stop time constraints for thermal power units:

[0203]

[0204] Active power output constraints of thermal power units:

[0205]

[0206] Reactive power output constraints of thermal power units:

[0207]

[0208] Thermal power unit climbing constraints:

[0209] dP g，i (τ′) / dτ′≤r u，i I g，i (τ′ - )+r su，i S u，i (τ′)

[0210] -dP g，i (τ′) / dτ′≤r d，i I g，i (τ′ + )+r sd，i S d，i (τ′)

[0211] In the formula, I g Indicates the status of the unit. When it is 1, it means it is online, and when it is 0, it means it is offline. τ′ + and τ′ - Respectively represent the moment after and before the moment τ′, t on and t off are the minimum online and offline state durations, respectively. and P g are the maximum and minimum active output of the unit, Q g Indicates the reactive power of the unit output. and Q g are the maximum and minimum reactive output of the unit, r u and rd are the upward and downward climbing speed limits of the unit respectively, r su and r sdThey are the climbing limits during the unit startup and shutdown processes respectively.

[0212] In one embodiment, the operating constraints in the original optimization model include the following AC grid constraints:

[0213] Active power balance constraints:

[0214]

[0215] Reactive power balance constraints:

[0216]

[0217] AC active power network loss constraints:

[0218] R lAC，j ((P rAC，j (τ′) 2 +(Q rAC，j (τ′) 2 )≤P lsAC，j (τ′)W rAC，j (τ′)

[0219] Active network loss and reactive network loss association constraints:

[0220] X l，j P lsAC，j (τ′)=R lAC，j Q lsAC，j (τ′)

[0221] AC voltage amplitude constraint:

[0222] W sAC，j (τ′)-W rAC，j (τ′)

[0223] =2R lAC，j P rAC，j (τ′)+2X l，j Q rAC，j (τ′)+R 1AC，j P lsAC，j (τ′)

[0224] +X l，j Q lsAC，j (τ′)

[0225] AC voltage phase angle constraint:

[0226] θ sAC，j (τ′)-θ rAC，j (τ′) = X l，j P rAC，j (τ′)-R lAC，j Q rAC，j (τ′)

[0227] Where j is the line number, N lAC is the number of AC lines, P wd Represents the predicted active power of new energy, P ld and Q ld Represent the predicted load active power and reactive power, P cv and Q cv Respectively represent the active power and reactive power of the converter station, P rAC and Q rAC Respectively represent the active power and reactive power at the receiving end of the AC line, P lsAC and Q lsAC They represent the active power and reactive power lost in the AC line, B represents the parallel susceptance parameter of the AC line, and W AC Represents the square of the AC node voltage, W rAC and W sAC Represents the square of the voltage at the receiving end and the voltage at the sending end of the AC line, R lAC and X1 represent the AC line resistance and reactance parameters respectively, θ sAC and θ rAC Indicates the phase angle of the AC line sending end and receiving end, M PQAC and M lAC They are the correlation matrix of AC line power flow and the correlation matrix of AC line loss respectively. The specific calculation expressions are:

[0228]

[0229] In one embodiment, the operating constraints in the original optimization model include the following DC grid constraints:

[0230] DC power balance constraints:

[0231]

[0232] DC active power loss constraint:

[0233] R lD C ，j (P rDC，j (τ′) 2 ≤P lsDC，j (τ′)W rDC，j (τ′)

[0234] DC voltage amplitude constraint:

[0235] W sDC，j (τ′)-W rDC，j (τ′)=2R lDC，j P rDC，j (τ′)+R lDC，j P lsDC，j (τ′)

[0236] Where N lDC Indicates the number of DC lines, P DC represents the DC node injection power, P rDC Represents the active power at the receiving end of the DC line, P lsDC Indicates the active power lost in the DC line, W rDC and W sDC Respectively represent the square of the voltage at the receiving end and the voltage at the sending end of the DC line, R lDC Indicates the DC line resistance parameter, M PQDC and M lDC They are the correlation matrix of DC line power flow and the correlation matrix of DC line loss respectively. The specific expressions are calculated as follows:

[0237]

[0238] In one embodiment, the operation constraints in the original optimization model include the following converter station constraints:

[0239] Power balance constraints of converter stations:

[0240] P cv，i (τ′)=P DC，i (τ′)-P lscv，i (τ′)

[0241] Active power loss constraints of converter stations:

[0242] P lscv，i (τ′)=β|P cv，i (τ′)|

[0243] Converter station power capacity constraints:

[0244] (P cv，i (τ′) 2 +(Q cv，i (τ′) 2 ≤(S cv，i ) 2

[0245] Where β represents the converter station loss coefficient, P lscv Indicates the active power loss of the converter station, S cv Indicates the power capacity of the converter station.

[0246] In one embodiment, the operating constraints in the original optimization model include the following upper and lower limit constraints on variables:

[0247]

[0248] |P rAC，j (τ′)+P lsAC，j (τ′)|≤FAC，j

[0249] |P rAC，j (τ′)|≤F AC，j

[0250] |Q rAC，j (τ′)+Q lsAC，j (τ′)|≤F AC，j

[0251] |Q rAC，j (τ′)|≤F AC，j

[0252]

[0253] |P rDC，j (τ′)+P lsDC，j (τ′)|≤F DC，j

[0254] |P rDC，j (τ′)|≤F DC，j

[0255] In the formula, and W AC They represent the upper and lower limits of the square of the AC node voltage, and θ AC Respectively represent the upper and lower limits of the AC node phase angle, F AC represents the AC branch transmission capacity, and W DC They represent the upper and lower limits of the DC node voltage square, F DC Indicates the DC branch transmission capacity.

[0256] Correspondingly, the reconstructed objective function is:

[0257]

[0258] In one embodiment, the reconstruction optimization model includes the following thermal power unit constraints.

[0259] The associated constraints of the start and stop variables of the thermal power unit after reconstruction:

[0260] S u，i，t -S d，i，t =I g，i，(t+1) -I g，i，t

[0261] S u，i，t +S d，i，t ≤1

[0262] After reconstruction, the minimum start and stop time constraints of the unit are:

[0263]

[0264] Active power output constraints of thermal power units after reconstruction:

[0265]

[0266] Reactive power output constraints of thermal power units after reconstruction:

[0267]

[0268] The reconstructed thermal power unit climbing constraints:

[0269]

[0270]

[0271] In one embodiment, the reconstruction optimization model includes the following AC power grid constraints: Active power balance constraints after reconstruction:

[0272]

[0273] Reconstructed reactive power balance constraint:

[0274]

[0275] The AC active power network loss constraints after reconstruction are:

[0276]

[0277] The associated constraints of active network loss and reactive network loss after reconstruction are:

[0278]

[0279] The reconstructed AC voltage amplitude constraint:

[0280]

[0281] The reconstructed AC voltage phase angle constraint:

[0282]

[0283] In one embodiment, the reconstruction optimization model includes the following DC grid constraints. The DC power balance constraints after reconstruction:

[0284]

[0285] Reconstructed DC active power loss constraint:

[0286]

[0287] The reconstructed DC voltage amplitude constraint:

[0288]

[0289] In one embodiment, the reconstruction optimization model includes the following converter station constraints:

[0290] The power balance constraints of the converter station after reconstruction are:

[0291]

[0292] Active power loss constraints of converter stations after reconstruction:

[0293]

[0294] The power capacity constraint of the converter station after reconstruction:

[0295]

[0296] In one embodiment, the reconstruction optimization model includes the following upper and lower limit constraints on variables:

[0297]

[0298] In the formula, t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients, W n P g The actual coefficient matrix of the differential term of the corresponding Bernstein polynomial during n-order BP interpolation, taking n=3 as an example:

[0299]

[0300] Contains n+1 elements: When t is not the last period: If n is an even number, the first The element is I g，i，t ,back The element is I g，i，(t+1) , if n is an odd number, I g，i，t and I g，i，(t+1) The number of When t is the last period, Contains n+1 I g，i，t .

[0301] Specifically, the unit output plan includes: active output of thermal power units and reactive power output

[0302] The AC system power flow results include: the active power P at the receiving end of the AC line rAC，j (τ′)= Reactive power at the receiving end of the AC line Active power lost in AC line Reactive power lost in AC lines The square of the AC node voltage Phase angle of AC node

[0303] The DC system power flow results include: Active power at the receiving end of the DC line Active power lost in DC line and the square of the DC node voltage

[0304] The converter station output plan includes: the active power generated by the converter station Reactive power generated by the converter station and converter station active power loss

[0305] In summary, the present invention collects AC line parameters, multi-embedded DC line parameters, converter station parameters, thermal power unit parameters, new energy output data, load data and economic parameters of the power grid to be dispatched, and based on the collected parameters and data, takes minimizing the dispatching cost as the objective function, considers the operating constraints of the lines and equipment, and establishes a continuous-time power grid dispatching optimization model containing multi-embedded DC. Two new properties of Bernstein polynomials are proposed and proved to deal with the nonlinear terms brought by the embedded DC model, thereby enhancing the solution space transformation of the established optimization model, and mapping the optimization model from the continuous time function space to the algebraic space of the Bernstein polynomial interpolation coefficients. Finally, the continuous time optimization model in the function space is converted into a mixed integer second-order cone programming problem in the algebraic space and solved to obtain the first solution result, and the first solution result is substituted into the Bernstein polynomial to obtain the final dispatching result in its continuous time form, thereby guiding the power grid dispatcher to dispatch power resources according to the dispatching result. This method dispatches the power system containing multiple embedded DC based on the continuous time optimization technology of enhanced solution space transformation, fully exploits the intra-hour flexibility of the multi-embedded DC system, promotes the power system dispatch to cope with the net load fluctuations caused by new energy, and thus improves the safety and economy of the power system.

[0306] Example 2

[0307] The present invention also relates to a computer-readable storage medium having a computer program stored thereon, and the computer program implements the steps of the above method when executed by a processor.

[0308] Specifically, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0309] Example 3

[0310] The embodiment of the present invention provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps of the method of the above embodiment of the present invention.

[0311] The technical features of the above-mentioned embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. It should be noted that "in one embodiment", "for example", "for example", etc. of the present invention are intended to illustrate the present invention, rather than to limit the present invention.

[0312] The above-mentioned embodiments only express several implementation methods of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent application. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A method for optimizing continuous time scheduling of a power grid containing multiple embedded DC, characterized in that: include: Obtain the grid system parameters of the grid to be dispatched, take the minimization of dispatching cost within the dispatching period as the objective function, set the operating constraints, and establish the original optimization model of continuous-time grid dispatching with multiple embedded DC, in which the quantities to be decided include the continuous-time variables of grid resource dispatching; The original optimization model is subjected to a solution space transformation to transform the continuous time variables into discrete variables to obtain a reconstructed optimization model; the solution space transformation comprises: dividing the scheduling period into a plurality of time periods, expressing the continuous time variables in the original optimization model as n-order Bernstein polynomials and transforming the relational expressions in the original optimization model into relational expressions about Bernstein polynomial coefficients according to a transformation strategy; n is an integer greater than 2; Solving the reconstruction optimization model to obtain a first solution result of the discrete variable, substituting the coefficients of the Bernstein polynomial in the first solution result into the Bernstein polynomial of the corresponding continuous time variable to obtain a final scheduling result of the corresponding continuous time variable; The conversion strategy includes: Convert the continuous time variable integral value, continuous time variable differential value and continuous time variable in linear relationship into variables related to Bernstein polynomial coefficients; The form is (P1(τ)) 2 +(P2(τ)) 2 The constraint ≤P3(τ)P4(τ) is transformed into The form (P1(τ)) 2 +(P2(τ)) 2 ≤(C) 2 The constraint is converted to Where τ is the time variable in each time period, P1(τ), P2(τ), P3(τ) and P4(τ) are the Bernstein polynomials corresponding to the continuous time variables. They are the Bernstein polynomials P m The coefficients of the pth basis function and the qth basis function in (τ), m=1,2,3,4, p=0,1,……,n, q=0,1,……,n, C is a constant.

2. The method for optimizing continuous time scheduling of a power grid containing multiple embedded DC as claimed in claim 1, characterized in that: The objective function in the original optimization model is: C su,i (τ′)=c su,i S u,i (t′), C sd,i (τ′)=c sd,i S d,i (t′), C f,i (τ′)=c f,i P g,i (t′) Where H is the total duration of the dispatch cycle, τ′ is the time variable within the dispatch cycle, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the number of the grid node, c su 、c sd They represent the single startup cost and single shutdown cost of thermal power units, respectively, and c f represents the fuel cost coefficient, S u Indicates the variable indicating the power-on action. If it is 1, it means the power-on action is executed. Otherwise, it is 0. d Indicates the variable indicating the shutdown action. 1 indicates the shutdown action is executed, otherwise it is 0. Each period can execute the startup or shutdown action at most once at the beginning of the period. S u and S d is a discrete variable, P g Indicates the active power output of thermal power units; The reconstructed objective function is: Where t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients.

3. The method for optimizing continuous time scheduling of a power grid containing multiple embedded DC as claimed in claim 1, characterized in that: The operating constraints in the original optimization model include the following thermal power unit constraints: Thermal power unit start and stop variable association constraints: S u,i (t′)-S d,i (τ′)=I g,i (t′ + )-I g,i (t′ - ) S u,i (t′)+S d,i (τ′)≤1 Minimum start and stop time constraints for thermal power units: Active power output constraints of thermal power units: Reactive power output constraints of thermal power units: Thermal power unit climbing constraints: dP g,i (τ′) / dτ′≤r u,i I g,i (t′ - )+r su,i S u,i (t′) -dP g,i (τ′) / dτ′≤r d,i I g,i (t′ + )+r sd,i S d,i (t′) Where τ′ is the time variable in the dispatch period, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the grid node number, S u Indicates the variable indicating the power-on action. 1 indicates the power-on action, otherwise 0. S d Indicates the variable indicating the shutdown action. 1 indicates the shutdown action, otherwise it is 0. Set each time period to execute the startup or shutdown action at most once at the beginning of the time period. g Indicates the status of the thermal power unit. When it is 1, it means it is online, and when it is 0, it means it is offline. u , S d ,I g is a discrete variable, τ′ + and τ′ - Respectively represent the moment after and before the moment τ′, t on and t off are the minimum online and offline state durations, respectively, g Indicates the active power output of the thermal power unit, and P g are the maximum and minimum active output of the thermal power unit, Q g Indicates the reactive power output of thermal power units. and Q g are the maximum and minimum reactive power outputs of thermal power units, r u and r d are the upward and downward climbing speed limits of thermal power units, r su and r sd They are respectively the ramp limits during the startup and shutdown processes of thermal power units; The reconstruction optimization model includes the following reconstructed thermal power unit constraints: The associated constraints of the start and stop variables of the thermal power unit after reconstruction: S u,i,t -S d,i,t =I g,i,(t+1) -I g,i,t S u,i,t +S d,i,t ≤1 After reconstruction, the minimum start and stop time constraints of the unit are: Active power output constraints of thermal power units after reconstruction: Reactive power output constraints of thermal power units after reconstruction: The reconstructed thermal power unit climbing constraints: Where t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients, W n P g The actual coefficient matrix of the differential terms of the corresponding Bernstein polynomial during n-order BP interpolation; Contains n+1 elements: When t is not the last period: If n is an even number, the first The element is I g,i,t ,back The element is I g,i,(t+1) , if n is an odd number, I g,i,t and I g,i,(t+1) The number of When t is the last period, Contains n+1 I g,i,t .

4. The method for optimizing continuous time scheduling of a power grid containing multiple embedded DC as claimed in claim 1, characterized in that: The operating constraints in the original optimization model include the following AC power grid constraints: Active power balance constraints: Reactive power balance constraints: AC active power network loss constraints: R lAC,j ((P rAC,j (t′)) 2 +(Q rAC,j (t′)) 2 )≤P lsAC,j (t′)W rAC,i (t′) Active network loss and reactive network loss association constraints: X l,j P lsAC,j (τ′)=R lAC,j Q lsAC,j (t′) AC voltage amplitude constraint: W sAC,j (t′)-W rAC,j (t′) =2R lAC,j P rAC,j (τ′)+2X l,j Q rAC,j (τ′)+R lAC,j P lsAC,j (τ′)+X l,j Q lsAC,j (τ′) AC voltage phase angle constraint: i sAC,j (τ′)-θ rAC,j (τ′)=X l,j P rAC,j (t′)-R lAC,j Q rAC,j (t′) Where τ′ is the time variable in the dispatching period, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the grid node number, j is the line number, and N lAC is the number of AC lines, P g Indicates the active power output of the thermal power unit, Q g Represents the reactive power output of the thermal power unit, P wd Represents the predicted active power of new energy, P ld and Q ld Represent the predicted load active power and reactive power, P cv and Q cv Represent the active power and reactive power of the converter station respectively, P rAC and Q rAC Respectively represent the active power and reactive power at the receiving end of the AC line, P lsAC and Q lsAC They represent the active power and reactive power lost in the AC line, B represents the parallel susceptance parameter of the AC line, and W AC Represents the square of the AC node voltage, W rAC and W sAC Represents the square of the voltage at the receiving end and the voltage at the sending end of the AC line, R lAC and X l Respectively represent the AC line resistance and reactance parameters, θ sAC and θ rAC Indicates the phase angle of the AC line sending end and receiving end, M PQAC and M lAC They are the correlation matrix of AC line power flow and the correlation matrix of AC line loss respectively. The specific calculation expressions are: The reconstruction optimization model includes the following reconstructed AC power grid constraints: Reconstructed active power balance constraint: Reconstructed reactive power balance constraint: The AC active power network loss constraints after reconstruction are: The associated constraints of active network loss and reactive network loss after reconstruction are: The reconstructed AC voltage amplitude constraint: The reconstructed AC voltage phase angle constraint: Where t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients, containing the elements or and They represent the coefficients of the p-th and q-th terms in the Bernstein polynomial respectively.

5. The method for optimizing continuous time scheduling of a power grid containing multiple embedded DC as claimed in claim 1, characterized in that: The operating constraints in the original optimization model include the following DC grid constraints: DC power balance constraints: DC active power loss constraint: R lDC,j (P rDC,j (t′)) 2 ≤P lsDC,j (t′)W r2C,j (t′) DC voltage amplitude constraint: W sDC,j (t′)-W rDC,j (τ′)=2R lDC,j P rDC,j (t′)+R lDC,j P lsDC,j (t′) Where τ′ is the time variable in the dispatching period, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the grid node number, j is the line number, and N lDC Indicates the number of DC lines, P DC represents the DC node injection power, P rDC Represents the active power at the receiving end of the DC line, P lsDC Indicates the active power lost in the DC line, W rDC and W sDC Respectively represent the square of the voltage at the receiving end and the voltage at the sending end of the DC line, R lDC Indicates the DC line resistance parameter, M PQDC and M lDC They are the correlation matrix of DC line power flow and the correlation matrix of DC line loss respectively. The specific expressions are calculated as follows: The reconstruction optimization model includes the following reconstructed DC grid constraints: The DC power balance constraint after reconstruction: Reconstructed DC active power loss constraint: The reconstructed DC voltage amplitude constraint: Where t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients, containing the elements or and They represent the coefficients of the p-th and q-th terms in the Bernstein polynomial respectively.

6. The method for optimizing continuous time scheduling of a power grid containing multiple embedded DC as claimed in claim 1, characterized in that: The operating constraints in the original optimization model include the following converter station constraints: Power balance constraints of converter stations: P cv,i (τ′)=P DC,i (t′)-P lscv,i (t′) Active power loss constraints of converter stations: P lscv,i (τ′)=β|P cv,i (t′)| Converter station power capacity constraints: (P cv,i (t′)) 2 +(Q cv,i (t′)) 2 ≤(S cv,i ) 2 Where τ′ is the time variable within the dispatch period, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the grid node number, P cv and Q cv Represent the active power and reactive power of the converter station respectively, P DC represents the DC node injection power, β represents the converter station loss coefficient, P lscv Indicates the active power loss of the converter station, S cv Indicates the power capacity of the converter station; The reconstruction optimization model includes the following reconstructed converter station constraints: The power balance constraints of the converter station after reconstruction are: Active power loss constraints of converter stations after reconstruction: The power capacity constraint of the converter station after reconstruction: Where t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients, containing the elements Represents the coefficient of the p-th term in the Bernstein polynomial.

7. The method for optimizing continuous time scheduling of a power grid containing multiple embedded DC as claimed in claim 1, characterized in that: The operating constraints in the original optimization model include the following upper and lower limit constraints on variables: |P rAC,j (t′)+P lsAC,j (τ′)|≤F AC,j |P rAC,j (τ′)|≤F AC,j |Q rAC,j (t′)+Q lsAC,j (τ′)|≤F AC,j |Q rAC,j (τ′)|≤F AC,j |P rDC,j (t′)+P lsDC,j (τ′)|≤F DC,j |P rDC,j (τ′)|≤F DC,j Where τ′ is the time variable in the dispatching period, τ′∈interval[0,H], (·)(τ′) represents the corresponding variable at time τ′, i is the grid node number, j is the line number, and W AC represents the square of the AC node voltage, and W AC Respectively represent the upper and lower limits of the square of the AC node voltage, θ AC represents the AC node phase angle, and θ AC Respectively represent the upper and lower limits of the AC node phase angle, and P rAC and Q rAC Respectively represent the active power and reactive power at the receiving end of the AC line, P lsAC and Q lsAC Respectively represent the active power and reactive power lost in the AC line, F AC Represents the AC branch transmission capacity, W DC represents the square of the DC node voltage, and W Dc They represent the upper and lower limits of the DC node voltage square, P rDC Represents the active power at the receiving end of the DC line, P lsDC Indicates the active power lost in the DC line, F DC Indicates the DC branch transmission capacity; The reconstructed optimization model includes the following reconstructed upper and lower limit constraints of the following variables: Where t is the time period number, (·) t represents the variable for period t, yes(·) t The vector of Bernstein polynomial coefficients.

8. The method for optimizing continuous time scheduling of a power grid containing multiple embedded DC as claimed in claim 1, characterized in that: The continuous time variables in the original optimization model are represented as cubic Bernstein polynomials, n=3.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

10. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images