Method, device and control system for obtaining dispatching strategy of power system quantitative accommodation

Abstract

The application discloses a kind of power system quantization accommodation scheduling strategy acquisition method, device and control system, belong to power system dispatching technical field, the method includes: based on wind power day-ahead forecast output curve and random output probability distribution information is constructed random accommodation rate index, quantification is to renewable energy accommodation level;In day-ahead scheduling process, the calculation of random accommodation rate index is realized, and implicit affine strategy is used to guarantee the unexpectedness of scheduling strategy, improve the reliability and security of scheduling strategy;Through random accommodation rate evaluation model, the maximum and minimum random accommodation rate of power system is evaluated, and a random accommodation rate constraint is constructed based on this, the quantization constraint of system accommodation level is realized, to avoid the problem that the power system operation cost is too high and the renewable energy accommodation rate is too low caused by unreasonable setting of abandoned electricity penalty cost, finally, on the basis of quantization guaranteeing that accommodation level reaches given requirement, the reliability and economy of scheduling strategy are improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of power system dispatching, and more particularly relates to a method and device for obtaining a dispatching strategy for quantifying accommodation of a power system and a control system. BACKGROUND

[0002] In recent years, in response to energy crisis, environmental pollution and other problems, renewable energy, represented by wind power, has developed rapidly, and as its penetration rate in power systems is increasing, the randomness and volatility of its output bring great challenges to system operation and put forward higher requirements for day-ahead dispatching decisions.

[0003] Stochastic optimization and robust optimization are two main methods for handling the randomness of renewable energy output in existing day-ahead dispatching methods. Stochastic optimization is based on scenario construction model, which cannot take into account all possible scenarios. The more scenarios considered, the more reliable the dispatching strategy and the lower the computational efficiency. Robust optimization can guarantee reliable operation under any scenario in the given uncertainty set, but does not take into account the operation risk corresponding to the small probability scenarios outside the set. At the same time, in existing methods, renewable energy curtailment penalty is often set to promote renewable energy accommodation, but high curtailment penalty cost may increase operation cost while improving accommodation level, and low curtailment penalty cost may result in failure to achieve the required accommodation level. In existing methods, the accommodation level that can be achieved by the power system under the dispatching strategy decided is only known through scenario simulation or actual system operation, in other words, there is an urgent need for a method to quantify the accommodation level of renewable energy to quantitatively evaluate and characterize the accommodation level of renewable energy in the day-ahead dispatching process, so as to more effectively guarantee the accommodation. In addition, in existing methods, the randomness is often handled based on complete renewable energy output curve to decide the dispatching strategy, i.e. ideal situation, which violates the timing logic of random output realization and does not meet the non-anticipation of dispatching strategy, which may result in infeasibility of the dispatching strategy decided in the day-ahead dispatching process and failure to achieve the accommodation level in the day-ahead dispatching process, in other words, the day-ahead dispatching result is too optimistic, which is not conducive to guaranteeing system operation safety and effectiveness of renewable energy accommodation.

[0004] In summary, safety, economy and green are three very important goals of power system operation, and how to quantitatively guarantee accommodation and improve the economy of power system operation under the premise of ensuring operation safety is a problem to be solved. SUMMARY

[0005] In view of the above defects or improvement needs of the prior art, the present application provides a method and device for obtaining a dispatching strategy for quantifying accommodation of a power system and a control system, which aims to solve the technical problem of how to quantitatively guarantee accommodation and improve the economy of power system operation under the premise of ensuring operation safety.

[0006] To achieve the above object, according to one aspect of the present application, a method for obtaining a scheduling strategy of power system quantitative accommodation is provided, comprising:

[0007] S1: constructing a power system day-ahead robust scheduling model based on dynamic uncertainty set and implicit affine strategy; the objective function of the power system day-ahead robust scheduling model is to minimize the sum of unit operation cost and expected load shedding penalty cost in system prediction scenario;

[0008] S2: calculating a random accommodation rate index based on wind power day-ahead prediction output curve and random output probability distribution information, and linearizing the random accommodation rate index; the random accommodation rate index is a function of the upper bound of dynamic uncertainty set as the independent variable;

[0009] S3: constructing a random accommodation rate evaluation model based on the constructed power system day-ahead robust scheduling model and the linearized random accommodation rate index; the constraints of the random accommodation rate evaluation model are the same as those of the power system day-ahead robust scheduling model, and the objective function is to maximize the system random accommodation rate index;

[0010] S4: evaluating the maximum random accommodation rate of the power system by using the random accommodation rate evaluation model, calculating the minimum random accommodation rate of the power system by using the linearized random accommodation rate index and the wind power day-ahead prediction output curve, and constructing a random accommodation rate constraint by using the maximum and minimum random accommodation rates;

[0011] S5: embedding the random accommodation rate constraint into the power system day-ahead robust scheduling model to obtain a power system quantitative accommodation day-ahead robust scheduling model, and solving the power system quantitative accommodation day-ahead robust scheduling model to obtain a scheduling strategy.

[0012] In one embodiment, in S1, the following is used: represents the objective function of the power system day-ahead robust scheduling model;

[0013] wherein subscript i and t are node and unit number and time period number respectively, N TU is the number of thermal power units, Δt is the time period length, and T is the number of scheduling time periods; c sd is the unit single start-stop cost; U is the thermal power unit start-stop action state 0-1 variable; is the output of the thermal power unit in the prediction scenario, C f (·) is the fuel cost function of the thermal power unit, ELS is the expected load shedding amount, c l is the corresponding penalty cost coefficient, a ls , b ls , and c ls are the coefficients of the ELS quadratic calculation formula,P w The lower bound of the output of the wind power uncertainty set.

[0014] In one of the embodiments, in the S1, the constraint of constructing the day-ahead robust dispatch model of the power system by using the dynamic uncertainty set and the implicit affine policy includes:

[0015]

[0016]

[0017] wherein, U D is the dynamic uncertainty set, the subscripts i and t are the node and unit number and time period number respectively, is the predicted output of the wind power, P w,max is the maximum output of the wind power, P w and are the lower bound and upper bound variables of the output of the dynamic uncertainty set to be decided respectively, σ is the standard deviation; u is the 0-1 variable of the start-stop state of the thermal power unit, equal to 1 indicates that the unit is in the start state, and equal to 0 indicates that the unit is in the stop state, T i on and T i off are the minimum start-up and shutdown times of the thermal power unit respectively; P TU,min and P TU,max are the minimum and maximum outputs of the thermal power unit respectively; R U and R D are the maximum up and down ramp rates of the thermal power unit respectively; P cha，max and P dis,max are the maximum charge and discharge power of the energy storage, η cha and η dis are the charge and discharge efficiencies respectively, S min and S max are the minimum and maximum state of charge respectively, E r is the rated capacity of the energy storage, and E is the energy stored by the energy storage; and are the introduced dispatchable output range variables of the thermal power unit and the energy storage respectively, is the energy storage power under the predicted scenario; N w , N BES and N d are the numbers of the wind farm, the energy storage and the load respectively, P d is the load power; s is the line power transmission distribution factor, and the subscript l is the line number, P l max is the upper limit of the line transmission capacity; the robustness in the dynamic uncertainty set is checked by using the column and constraint generation algorithm.

[0018] In one embodiment, the S2 includes calculating the random accommodation rate index based on the wind power day-ahead prediction output curve and the random output probability distribution information, comprising:

[0019] Defining the random accommodation rate index Using a polynomial Approximating the random output probability density function, and then calculating the random accommodation rate index

[0020] Wherein, E{·} represents the expectation, is the abandoned wind power, N p is the polynomial order, p n is the polynomial coefficient, a, b and c are the coefficients of the SAR quadratic calculation formula.

[0021] In one embodiment, the S2 includes linearizing the random accommodation rate index, comprising: Wherein, n is the segment number, N pw is the linearization segment number, q 1 and q 0 are the linearization SAR calculation expression coefficients, is the introduced linear relaxation variable.

[0022] In one embodiment, the S4 includes:

[0023] Using the formula SAR≥SAR req to represent the determination of the random accommodation rate constraint, wherein, Wherein, SAR req is the system random accommodation rate requirement value, SAR is the minimum random accommodation rate, is the maximum random accommodation rate.

[0024] In one embodiment, the minimum random accommodation rate is determined by the formula

[0025] In one embodiment, the maximum random accommodation rate is determined by the formula In one embodiment, the maximum random accommodation rate of the power system is evaluated by constructing a random accommodation rate evaluation model with as the objective function, and SAR is the random accommodation rate index.

[0026] According to another aspect of the present application, a power system quantitative accommodation scheduling strategy acquisition device is provided, comprising:

[0027] ​The modeling module is configured to construct a day-ahead robust dispatch model of the power system based on the dynamic uncertainty set and the implicit affine policy; and a target function of the day-ahead robust dispatch model of the power system is to minimize a sum of a unit operation cost and a penalty cost of load shedding under a system prediction scenario.

[0028] The calculation module is configured to calculate a random accommodation rate index based on the day-ahead predicted output curve of the wind power and the random output probability distribution information, and linearize the random accommodation rate index; the random accommodation rate index is a function of an upper bound of the dynamic uncertainty set as an independent variable.

[0029] The construction module is configured to construct a random accommodation rate evaluation model based on the constructed day-ahead robust dispatch model of the power system and the linearized random accommodation rate index; a constraint of the random accommodation rate evaluation model is the same as that of the day-ahead robust dispatch model of the power system, and a target function is to maximize a system random accommodation rate index.

[0030] The evaluation module is configured to evaluate a maximum random accommodation rate of the power system by using the random accommodation rate evaluation model, calculate a minimum random accommodation rate of the power system by using the linearized random accommodation rate index and the day-ahead predicted output curve of the wind power, and construct a random accommodation rate constraint by using the maximum and minimum random accommodation rates.

[0031] The solving module is configured to embed the random accommodation rate constraint into the day-ahead robust dispatch model of the power system to obtain a day-ahead robust dispatch model of the power system for quantified accommodation, and solve the day-ahead robust dispatch model of the power system for quantified accommodation to obtain a dispatch strategy.

[0032] According to another aspect of the present application, there is provided a control system of a power system, comprising a memory and a processor, the memory storing a computer program, and the processor implementing the steps of the method when executing the computer program.

[0033] According to another aspect of the present application, there is provided a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the method when executed by a processor.

[0034] In general, the above technical solutions conceived by the present application can achieve the following beneficial effects compared with the prior art:

[0035] (1) The present invention provides a method for obtaining a dispatching strategy for quantitative absorption of a power system. Based on the wind power day-ahead output curve and random output probability distribution information, a random absorption rate index is constructed to achieve the quantification of the renewable energy absorption level. The random absorption rate index is calculated during the day-ahead dispatching process, and an implicit affine strategy is adopted to ensure the unpredictability of the dispatching strategy, thereby improving the reliability and security of the dispatching strategy. The maximum and minimum random absorption rates of the power system are evaluated through a random absorption rate evaluation model, and a random absorption rate constraint is constructed based on this model to achieve quantitative constraints on the system absorption level, thereby avoiding the problems of excessively high power system operating costs and excessively low renewable energy absorption rates caused by unreasonable settings of power abandonment penalty costs. Ultimately, the reliability and economy of the dispatching strategy are improved on the basis of quantitatively ensuring that the absorption level meets given requirements.

[0036] (2) This scheme uses represents the objective function of the day-ahead robust dispatch model of the power system; the objective function no longer needs to include power curtailment penalty costs, which can avoid the problems of excessively high power system operating costs and low renewable energy absorption rate caused by unreasonable setting of power curtailment penalty costs.

[0037] (3) This scheme adopts a dynamic uncertainty set and takes the wind power uncertainty output boundary as the decision variable, which can improve the flexibility of the scheduling scheme while realizing the calculation of the random absorption rate; the implicit affine strategy is adopted to ensure the unpredictability of the scheduling strategy, and avoid overestimation of the random absorption rate indicator while improving the reliability and safety of the scheduling strategy.

[0038] (4) This scheme uses Define the random absorption rate indicator; by constructing and deriving the calculation method of the random absorption rate indicator, the random absorption rate indicator can be calculated in the day-ahead scheduling process, thereby quantifying the renewable energy absorption level.

[0039] (5) This scheme uses the formula Linearizing the random absorption rate index can avoid nonlinear terms in the model and improve the solution efficiency of the model.

[0040] (6) In this scheme, the formula SAR≥SAR is used req It indicates that the random absorption rate constraint is determined. By constructing the random absorption rate constraint, the quantitative guarantee of the system absorption level can be achieved from the constraint level.

[0041] (7) This scheme uses the formula The minimum random absorption rate is determined. By calculating the minimum random absorption rate of the power system, the lowest absorption level of the power system can be known, and a lower limit of the range is provided for setting the random absorption rate constraint.

[0042] (8) In this scheme, the random accommodation rate evaluation model is constructed with the target function of maximal random accommodation rate of the power system as the target function, and the maximal random accommodation rate of the power system is evaluated, so that the highest accommodation level of the power system can be known, and an upper limit of a range for setting the random accommodation rate constraint is provided. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 A flowchart of a scheduling strategy acquisition method for power system quantitative accommodation provided for Embodiment 1 of the present application is provided.

[0044] Figure 2 A system topology diagram for verifying an example provided for Embodiment 1 of the present application is provided.

[0045] Figure 3 An electric load curve and a wind power prediction output curve provided for Embodiment 1 of the present application are provided.

[0046] Figure 4 A dynamic boundary diagram obtained by the method and a comparative method 2 provided for Embodiment 1 of the present application are provided.

[0047] Figure 5 A wind power actual output curve within the dynamic boundary provided for Embodiment 1 of the present application is provided.

[0048] Figure 6 Scheduling results of the comparative method 2 in ideal and actual situations provided for Embodiment 1 of the present application are provided.

[0049] Figure 7 A range of schedulable output of a unit decided by the method provided for Embodiment 1 of the present application is provided. DETAILED DESCRIPTION

[0050] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application. In addition, the technical features involved in each embodiment of the present application described below can be combined with each other as long as they do not conflict with each other.

[0051] Embodiment 1

[0052] As shown in Figure 1 , the present embodiment provides a scheduling strategy acquisition method for power system quantitative accommodation, which comprises: S1-S5.

[0053] ​S1: constructing a day-ahead robust dispatch model of the power system based on the dynamic uncertainty set and the implicit affine policy; the objective function of the day-ahead robust dispatch model of the power system is to minimize the sum of the expected unit operation cost and the load shedding penalty cost in the predicted scenario.

[0054] S2: calculating a stochastic accommodation rate index based on the day-ahead predicted output curve and the random output probability distribution information of the wind power, and linearizing the stochastic accommodation rate index; the stochastic accommodation rate index is a function of the upper bound of the dynamic uncertainty set.

[0055] S3: constructing a stochastic accommodation rate evaluation model based on the constructed day-ahead robust dispatch model of the power system and the linearized stochastic accommodation rate index; the constraints of the stochastic accommodation rate evaluation model are the same as those of the day-ahead robust dispatch model of the power system, and the objective function is to maximize the stochastic accommodation rate index of the system.

[0056] S4: evaluating the maximum stochastic accommodation rate of the power system by using the stochastic accommodation rate evaluation model, calculating the minimum stochastic accommodation rate of the power system by using the linearized stochastic accommodation rate index and the day-ahead predicted output curve of the wind power, and constructing a stochastic accommodation rate constraint by using the maximum and minimum stochastic accommodation rates.

[0057] S5: embedding the stochastic accommodation rate constraint into the day-ahead robust dispatch model of the power system to obtain a day-ahead robust dispatch model of the power system for quantified accommodation, and solving the day-ahead robust dispatch model of the power system for quantified accommodation to obtain a dispatch strategy; specifically, the dispatch strategy can include the day-ahead start-stop mode of the thermal power unit, the adjustable output range of the thermal power and energy storage unit, the unit output in the predicted scenario, and the dynamic boundary of the wind power uncertainty set, and the obtained dispatch strategy can improve the reliability and economy of system operation under the premise of ensuring that the wind power accommodation level meets the given requirements.

[0058] To construct the day-ahead robust dispatch model of the power system designed in the application, first, the parameters of the power system and the day-ahead predicted curve are collected, including:

[0059] (1) the topological structure, line parameters and device technical parameters of the power system;

[0060] (2) the wind power output prediction curve, the output probability density function and the load curve.

[0061] The objective function of the day-ahead robust dispatch model of the power system designed in the application is constructed as follows:

[0062]

[0063] wherein formula (1) minimizes the sum of the expected unit operation cost and the load shedding penalty cost in the predicted scenario, the subscripts i and t are the node / device number and the time period number respectively, N TUis the number of thermal power units, Δt is the time period length, and T is the number of scheduling periods; c sd is the single start-stop cost of a unit; U is a 0-1 variable of the start-stop action of a thermal power unit, and is equal to 1 to indicate a start action of the unit and is equal to 0 to indicate a stop action of the unit; formula (2) is a fuel cost calculation formula of a thermal power unit, is the output of a thermal power unit under a predicted scenario, C f is a fuel cost function of a thermal power unit, c 2 , c 1 , c 0 is a cost coefficient; ELS is a load shedding amount expectation, c l is a corresponding penalty cost coefficient.

[0064] In the existing day-ahead robust scheduling model of a power system, wind power uncertainty sets are described by the following formula:

[0065]

[0066] wherein U is a wind power output uncertainty set, P w and are wind power uncertainty output, output lower bound and output upper bound, respectively.

[0067] For formula (3), P w and are given, and operation risks outside the uncertainty boundary cannot be considered, although the probability of a random wind power scenario appearing outside the set is small, but the consequences can be very serious; on the other hand, different day-ahead start-stop strategies of units correspond to different regulation capabilities, and correspondingly, the system's accommodation level and operation economy are different, for this, the method proposed in the present application constructs a dynamic uncertainty set and proposes an implicit affine strategy considering the dynamic boundary of random output of renewable energy, which can guarantee the unexpectedness of the scheduling strategy while considering the above two factors, and guarantee safe and reliable operation, and the details are as follows:

[0068]

[0069]

[0070] wherein U D is a constructed dynamic uncertainty set, the subscripts i and t are node and unit number and time period number, respectively, is a predicted wind power output, P w,max is a maximum wind power output, P w and are output lower bound and upper bound variables of the dynamic uncertainty set to be decided, respectively, σ is the standard deviation; u is the 0-1 variable of the start-stop state of the thermal power unit, equal to 1 indicating that the unit is in the start state, and equal to 0 indicating that the unit is in the stop state, T i on and T i off are the minimum start and stop times of the thermal power unit, respectively; P TU,min and P TU,max are the minimum and maximum outputs of the thermal power unit, respectively; R U and R D are the maximum up and down ramp rates of the thermal power unit, respectively; P cha，max and P dis,max are the maximum charge and discharge powers of the energy storage, η cha and η dis are the charge and discharge efficiencies, S min and S max are the minimum and maximum state of charge, E r is the rated capacity of the energy storage, and E is the energy stored by the energy storage; and are the introduced dispatchable output range variables of the thermal power unit and the energy storage, is the energy storage power under the prediction scenario; N w , N BES and N d are the numbers of wind farms, energy storages and loads, respectively, P d is the load power; s is the line power transmission distribution factor, and subscript l is the line number, P l max is the upper limit of the line transmission capacity; the robustness in the dynamic uncertainty set is checked by using the column and constraint generation algorithm.

[0071] As a preferred embodiment, in S1: the polynomial represents the objective function of the day-ahead robust dispatching model of the power system; wherein, subscripts i and t are the node and unit numbers and the time period numbers, respectively, N TU is the number of thermal power units, Δt is the time period length, and T is the number of dispatching time periods; c sd is the unit start-stop cost per time; u is the 0-1 variable of the start-stop action state of the thermal power unit; is the output of the thermal power unit under the prediction scenario, C f (·) is the fuel cost function of the thermal power unit; ELS is the expected load shedding amount, c l is the corresponding penalty cost coefficient, the polynomial is used to approximate the random output probability density function f i,t , and then the quadratic calculation formula of the expected load shedding amount is calculated as a ls , bls and c ls are coefficients of the ELS quadratic calculation formula, P w is the lower bound of the wind power output of the wind power uncertainty set.

[0072] As a preferred embodiment, the random accommodation rate index is calculated based on the wind power day-ahead prediction output curve and the random output probability distribution information in S2, including: defining the random accommodation rate index using a polynomial approximating the random output probability density function f i,t , and further calculating the random accommodation rate index wherein E{·} represents expectation, is the curtailed wind power, N p is the polynomial order, p n is the polynomial coefficient, and a, b and c are coefficients of the SAR quadratic calculation formula.

[0073] As a preferred embodiment, the random accommodation rate index is linearized in S2, including: wherein n is the segment number, N pw is the linearization segment number, q 1 and q 0 are coefficients of the linearized SAR calculation expression, is the introduced linear relaxation variable; the expected curtailed load ELS can also be linearized in the same way.

[0074] As a preferred embodiment, S4 includes: using the formula SAR≥SAR req to represent the determination of the random accommodation rate constraint, wherein, wherein SAR req is the system random accommodation rate requirement value, SAR is the minimum random accommodation rate, is the maximum random accommodation rate.

[0075] As a preferred embodiment, the minimum random accommodation rate is determined using the formula .

[0076] As a preferred embodiment, the random accommodation rate evaluation model is constructed with as the objective function to evaluate the maximum random accommodation rate of the power system, and SAR is the random accommodation rate index.

[0077] The simulation data described herein is only used to explain the present application, and is not used to limit the present application.

[0078] Consider Figure 2The verification system in the table 1, the parameters of three thermal power units (TU1, TU2, TU3) are shown in Table 1, the energy storage (BES) capacity is 25MW / 100MWh, the charging and discharging efficiency is 95%, the initial, minimum and maximum SOC is 60%, 20% and 100% respectively, the wind power installed capacity is 150MW, the load curve and the wind power predicted output curve are shown in Figure 3 The wind power output is subject to Gaussian distribution, the standard deviation is 20% of the predicted output, and the load penalty cost is 1000$ / MWh. In order to illustrate the effectiveness of the method, the method is compared with the existing method, and the method and the comparison method are introduced as follows.

[0079] The method: the scheduling method based on implicit affine strategy and considering random accommodation rate constraint. Comparison method 1: scheduling method based on implicit affine strategy and through wind power penalty cost to ensure accommodation. Comparison method 2: scheduling method based on traditional two-stage robust optimization and considering random accommodation rate constraint.

[0080] Table 1: Thermal power unit parameters

[0081]

[0082] The results of the comparison between the method and method 1 are used to verify the effectiveness of the accommodation rate index and the random accommodation rate constraint. The minimum and maximum random accommodation rates of the system evaluated by the method are 92.02% and 99.88% respectively, and five different SAR req The system day-ahead scheduling cost and the operation cost under 1000 intra-day random scenarios are shown in Table 2, it can be seen that the average accommodation rate in each scenario in the intra-day operation simulation is higher than the SAR index value corresponding to the day-ahead scheduling strategy, therefore, through the method, the system accommodation level can be effectively quantified and guaranteed in the day-ahead scheduling process. In addition, with the increase of the required accommodation level, the system start-stop cost and the total operation cost also increase, when SAR req The unit start-stop cost increases by 1500$ and 6000$ respectively, that is, the same degree of accommodation level improvement brings different degrees of operation cost increase, and too high pursuit of accommodation rate improvement will lead to operation economic loss.

[0083] For method 1, the wind curtailment penalty cost is set to 100, 200, 300 and 400 $ / MWh respectively, and the system day-ahead scheduling cost and operation cost under 1000 day-ahead random scenarios are shown in Table 3. It can be seen that the wind curtailment penalty cost has a great influence on the system consumption level and operation cost, and with the increase of the wind curtailment penalty cost, the system consumption level is improved, and the total operation cost is also increased. The system consumption level cannot be quantified and effectively guaranteed in the day-ahead scheduling process under this method, and the setting of too large / too small wind curtailment penalty cost will lead to too high operation cost / too low consumption level.

[0084] Table 2 solution results of the method

[0085]

[0086] Table 3 solution results of method 1

[0087]

[0088]

[0089] The results of the method and method 2 are compared to verify the effectiveness of the implicit affine strategy of the renewable energy random output dynamic boundary. The system maximum random consumption rate evaluated by method 2 is 99.94%, which is higher than the result of the method. req Set to 98%, respectively, using the method and method 2, the dynamic boundary obtained is as shown in Figure 4 , and a wind power output curve between the dynamic boundaries obtained by method 2 is as shown in Figure 5 , and the scheduling scheme obtained according to the complete output curve (ideal case) is as shown in Figure 6 , at this time there is no wind curtailment and load shedding, while in the actual scheduling process, only the current known wind power output and future prediction value can be used for decision-making, and the corresponding scheduling scheme is as shown in Figure 6 , it can be seen that there is wind curtailment at 4-5h and 23-24h, so it can be known that the result obtained by method 2 is too optimistic, which is not conducive to guarantee the safe and reliable operation of the power system. And through the method, the schedulable output range of the unit at each time can be obtained, as shown in Figure 7 , for any wind power output within the dynamic boundary, a feasible scheduling scheme can be found within the schedulable output range and will not affect the feasibility of the subsequent period, so as to guarantee the reliability of the scheduling scheme.

[0090] Example 2

[0091] The embodiment provides a device for obtaining a scheduling strategy of power system quantitative accommodation, comprising a modeling module, a calculation module, a construction module, an evaluation module and a solution module.

[0092] The modeling module is used for constructing a power system day-ahead robust scheduling model based on the dynamic uncertainty set and the implicit affine strategy; and an objective function of the power system day-ahead robust scheduling model is to minimize the sum of unit operation cost and expected load shedding penalty cost in a system predicted scenario.

[0093] The calculation module is used for calculating a random accommodation rate index based on a wind power day-ahead predicted output curve and random output probability distribution information, and linearizing the random accommodation rate index; and the random accommodation rate index is a function of an upper bound of the dynamic uncertainty set.

[0094] The construction module is used for constructing a random accommodation rate evaluation model based on the constructed power system day-ahead robust scheduling model and the linearized random accommodation rate index; constraints of the random accommodation rate evaluation model are the same as those of the power system day-ahead robust scheduling model, and an objective function is to maximize the system random accommodation rate index.

[0095] The evaluation module is used for evaluating the maximum random accommodation rate of the power system by using the random accommodation rate evaluation model, calculating the minimum random accommodation rate of the power system by using the linearized random accommodation rate index and the wind power day-ahead predicted output curve, and constructing a random accommodation rate constraint by using the maximum and minimum random accommodation rates.

[0096] The solution module is used for embedding the random accommodation rate constraint into the power system day-ahead robust scheduling model to obtain a power system quantitative accommodation day-ahead robust scheduling model, and solving the power system quantitative accommodation day-ahead robust scheduling model to obtain the scheduling strategy.

[0097] Embodiment 3

[0098] The embodiment provides a control system of a power system, comprising a memory and a processor, the memory stores a computer program, and the processor implements steps of a method when executing the computer program.

[0099] Embodiment 4

[0100] The embodiment provides a computer readable storage medium, which stores a computer program, and steps of a method are implemented when the computer program is executed by a processor.

[0101] Those skilled in the art can easily understand that the above description is only the preferred embodiment of the present application, and is not used to limit the present application, and any modification, equivalent replacement and improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for obtaining a dispatching strategy for quantitative consumption of a power system, characterized in that: include: S1: Constructing a robust dispatch model for power systems based on dynamic uncertainty sets and implicit affine strategy; The objective function of the power system day-ahead robust dispatch model is to minimize the expected sum of unit operating costs and load shedding penalty costs under the system forecast scenario; S2: Calculate a random absorption rate index based on the day-ahead wind power output curve and random output probability distribution information, and linearize the random absorption rate index; the random absorption rate index is a function with the upper bound of the dynamic uncertainty set as an independent variable; S3: constructing a stochastic absorption rate evaluation model based on the constructed power system day-ahead robust dispatch model and the linearized stochastic absorption rate index; the constraints of the stochastic absorption rate evaluation model are the same as those of the power system day-ahead robust dispatch model, and the objective function is to maximize the system stochastic absorption rate index; S4: using the random absorption rate evaluation model to evaluate the maximum random absorption rate of the power system, using the linearized random absorption rate index and the day-ahead wind power output curve to calculate the minimum random absorption rate of the power system, and using the maximum and minimum random absorption rates to construct a random absorption rate constraint; S5: embedding the random absorption rate constraint into the day-ahead robust dispatch model of the power system to obtain a day-ahead robust dispatch model for quantitative absorption of the power system, and solving the day-ahead robust dispatch model for quantitative absorption of the power system to obtain a dispatch strategy.

2. The method for obtaining a dispatching strategy for quantitative power system consumption according to claim 1, wherein: In S1: using represents the objective function of the power system day-ahead robust dispatch model; Among them, the subscript is the node and unit number, subscript is the time period number, is the number of thermal power units, is the time period length, T is the number of scheduling periods; The cost of starting and stopping the unit once; It is a 0-1 variable indicating the start and stop status of the thermal power unit; is the output of the thermal power unit in the predicted scenario, is the fuel cost function of thermal power units, To reduce the load, is the corresponding penalty cost coefficient, , 、 and for The coefficients of the quadratic formula, is the lower bound of wind power output uncertainty set, is the number of wind farms.

3. The method for obtaining a dispatching strategy for quantitative power consumption according to claim 1, wherein: In S1, the constraints of constructing the power system day-ahead robust dispatch model using the dynamic uncertainty set and implicit affine strategy include: ； ； ； ； ； ； ； ； ； ； ； ； ； ； ； ； ； ； ； in, is the dynamic uncertainty set, subscript is the node and unit number, subscript is the time period number, t+1 is the next moment after t, t-1 is the previous moment after t, Contribute to the uncertainty of wind power, For wind power forecasting, The maximum output of wind power, and are the lower and upper bound variables of the dynamic uncertainty set to be decided, , represents a normal distribution, is the standard deviation; It is a 0-1 variable for the start and stop status of the thermal power unit. When it is equal to 1, it means the unit is in the start state, and when it is equal to 0, it means the unit is in the shutdown state. and are the minimum startup and shutdown time of thermal power units, It is the 0-1 variable of the start and stop action status of the thermal power unit. is the 0-1 variable of the start and stop action state of the thermal power unit in the first period, T is the number of scheduling periods, is the output of the thermal power unit in the predicted scenario, is the number of thermal power units, is the time period length; and are the minimum and maximum outputs of thermal power units respectively; and are the maximum up and down ramp rates of thermal power units respectively; and are the maximum charging and discharging power of energy storage, and are the charge and discharge efficiency, and are the minimum and maximum charge states, respectively. is the rated capacity of energy storage, Energy stored for energy storage, The energy stored in the energy storage at the initial time of dispatch; 、 、 and For the introduced thermal power units and energy storage dispatchable output range variables, To predict the energy storage power in the scenario; , and are the number of wind farms, energy storage and loads respectively, is the load power; Line power transmission distribution factor, including: thermal power generation unit, wind farm, energy storage and load line power transmission distribution factor 、 、 、 , subscript is the line number, is the upper limit of line transmission capacity; the robustness within the dynamic uncertainty set is verified using a column and constraint generation algorithm.

4. The method for obtaining a dispatching strategy for quantitative power consumption according to claim 1, wherein: The random absorption rate index is calculated in S2 based on the wind power day-ahead predicted output curve and random output probability distribution information, including: Defining the stochastic absorption rate indicator , using the polynomial Approximate the random output probability density function and then calculate the random consumption rate index ; Among them, the subscript is the node and unit number, subscript is the time period number, T is the number of scheduling periods, is the number of wind farms; Contribute to the uncertainty of wind power, is the time period length, Express expectations, is the wind power curtailment, is the polynomial order, are the coefficients of the n-order polynomial, where the superscript n is the order. 、 and for The coefficients of the quadratic formula, It is the upper bound variable of the dynamic uncertainty set to be decided.

5. The method for obtaining a dispatching strategy for quantitative power consumption according to claim 1, wherein: The random absorption rate indicator is linearized in S2, including: ； in, is the random absorption rate index after linearization, subscript is the node and unit number, subscript is the time period number, T is the number of scheduling periods, is the number of wind farms; Segment number, is the number of linearization segments, and For linearization Calculate the coefficients of the expression, is the introduced linear slack variable, It is the upper bound variable of the dynamic uncertainty set to be decided.

6. The method for obtaining a dispatching strategy for quantitative power consumption in a power system according to claim 5, characterized in that: The S4 includes: Using the formula represents the constraint on the random absorption rate, where ; in, is the random absorption rate index after linearization, is the required value of the system random absorption rate, is the minimum random absorption rate, is the maximum random absorption rate.

7. The method for obtaining a dispatching strategy for quantitative power consumption according to claim 6, wherein: Using the formula Determine the minimum random absorption rate, subscript is the node and unit number, subscript is the time period number, For wind power forecasting, It is the upper bound variable of the dynamic uncertainty set to be decided.

8. The method for obtaining a dispatching strategy for quantitative power consumption according to claim 6, wherein: by As the objective function, a random absorption rate evaluation model is constructed to evaluate the maximum random absorption rate of the power system. Conduct an assessment, is the random absorption rate indicator after linearization.

9. A dispatching strategy acquisition device for quantitative consumption of power system, characterized in that: A method for obtaining a dispatching strategy for implementing the power system quantitative consumption as described in any one of claims 1 to 8, comprising: A modeling module is used to construct a day-ahead robust dispatch model for the power system based on a dynamic uncertainty set and an implicit affine strategy; the objective function of the day-ahead robust dispatch model for the power system is to minimize the sum of the expected unit operating costs and the load shedding penalty costs under the system prediction scenario; a calculation module, configured to calculate a random absorption rate index based on a day-ahead wind power output curve and random output probability distribution information, and linearize the random absorption rate index; the random absorption rate index is a function with an upper bound of a dynamic uncertainty set as an independent variable; A construction module is used to construct a stochastic absorption rate evaluation model based on the constructed power system day-ahead robust dispatch model and the linearized stochastic absorption rate index; the constraints of the stochastic absorption rate evaluation model are the same as the constraints of the power system day-ahead robust dispatch model, and the objective function is to maximize the system stochastic absorption rate index; An evaluation module is configured to evaluate the maximum random absorption rate of the power system using the random absorption rate evaluation model, calculate the minimum random absorption rate of the power system using the linearized random absorption rate index and the wind power day-ahead output curve, and construct a random absorption rate constraint using the maximum and minimum random absorption rates; A solution module is used to embed the random absorption rate constraint into the power system day-ahead robust dispatch model to obtain the power system day-ahead robust dispatch model for quantitative absorption, and solve the power system day-ahead robust dispatch model for quantitative absorption to obtain a dispatch strategy.

10. A control system for an electric power system, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images