Two-layer optimization dispatching method for distribution network with distributed renewable energy considering flexibility constraints

Abstract

The present invention provides a two-layer optimization dispatching method for a distribution network containing distributed new energy and considering flexibility constraints. First, the flexibility of each system of the distribution network is calculated; secondly, an upper-layer optimization model without considering flexibility constraints is established, and the model is solved to obtain an optimization dispatching strategy, based on which the lower-layer optimization model is solved to obtain the most severe net load curve; secondly, intuitive fuzzy programming is used to convert the upper-layer optimization model considering flexibility constraints into an intuitive fuzzy programming model, and then a new optimization dispatching strategy is obtained; then the operating costs corresponding to the two optimization dispatching strategies are compared, and finally it is repeatedly iterated until the absolute value of the difference in operating costs is less than a threshold. The present invention introduces intuitive fuzzy programming to solve the comprehensive most satisfactory solution of the model, which can take into account the flexibility and economy of the distribution network.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network optimization and dispatching, and in particular to a double-layer optimization and dispatching method for a distribution network containing distributed new energy sources taking flexibility constraints into consideration. Background Art

[0002] Wind power and photovoltaic power stand out among many renewable energy sources with their advantages such as wide distribution, mature technology and low cost. Since wind and solar power output has strong uncertainties such as intermittency and volatility, their large-scale access not only brings new challenges to the power quality, relay protection and dispatching of the distribution network, but also is not conducive to the low-carbon and economic operation of the distribution network. In order to suppress the volatility of distributed new energy and promote the consumption of new energy, power companies will not only formulate time-of-use electricity prices to guide demand response loads to participate in the operation and dispatch of the power grid, but also use virtual energy storage systems such as electric vehicle charging stations to absorb or release energy to optimize the operation and dispatch of the distribution network. Therefore, in the case of uncertainty in the combined output of wind and solar power, how to fully dispatch the flexible resources in the distribution network to improve the low-carbon economy of the distribution network is of great research significance.

[0003] The flexibility of the power system refers to its ability to respond quickly to large fluctuations in both supply and demand in the system under boundary constraints, and its ability to respond to foreseeable or unforeseeable changes and events. At present, the research on power system flexibility is mainly focused on two aspects: flexibility level assessment and power system optimization scheduling considering flexibility. The former mainly proposes reliable and reasonable system flexibility assessment indicators and calculates the margin of flexibility indicators to provide a reference for formulating efficient optimization scheduling strategies; the latter mainly considers renewable energy, adjustable power sources and controllable loads to establish a coordinated optimization scheduling model, so as to solve the optimization scheduling strategy that meets the flexibility required for all scenarios of uncertain fluctuations in new energy output. However, the probability of the extreme scenario of uncertain fluctuations in wind and solar combined output is extremely low in actual operation, and the optimization scheduling strategy that meets the flexibility required for extreme scenarios often pays a very high economic cost. How to balance the flexibility and economy of the power system optimization scheduling strategy is an issue that must be paid attention to in the construction of new power systems.

[0004] The traditional method of solving the two-layer optimization dispatch model of the distribution network is to alternately iterate the upper and lower models to obtain the optimal solution that meets the extreme output scenario of distributed renewable energy. However, the optimization dispatch strategy obtained by this method has a high economic cost, so the optimal solution is not the most satisfactory solution. Therefore, the intuitive fuzzy programming is introduced to solve the comprehensive most satisfactory solution of the model, so as to take into account the flexibility and economy of the distribution network. Summary of the invention

[0005] The purpose of the present invention is to address the problem of high economic cost of the optimization scheduling strategy caused by extreme scenarios of uncertain fluctuations in wind and solar combined output with extremely low probability. A two-layer optimization scheduling method for distributed renewable energy distribution networks considering flexibility constraints is proposed, and intuitive fuzzy planning is introduced to solve the optimization scheduling strategy that takes into account the flexibility requirements of most scenarios and economic operation.

[0006] The specific contents involved in the present invention are:

[0007] A two-layer optimization scheduling method for a distributed renewable energy distribution network considering flexibility constraints comprises the following steps:

[0008] Step 1: Calculate the flexibility of each system in the distribution network, the flexibility provided to the distribution network at the main network access point includes the flexibility provided to the distribution network at the main network access point, the flexibility provided to the distribution network by the electric vehicle charging station, and the flexibility provided to the distribution network by the demand response load;

[0009] Step 2: Establish an upper-level optimization model without considering flexibility constraints to solve the dispatching strategy that minimizes the distribution network operation cost in a scenario of wind and solar combined output;

[0010] Step 3: Establish a lower-level optimization model, and solve the corresponding most severe net load curve based on the initial optimization scheduling strategy obtained in step 2;

[0011] Step 4: Use intuitionistic fuzzy programming to convert the upper model considering flexibility constraints into an intuitionistic fuzzy programming model;

[0012] Step 5: Combine the most severe net load curve obtained in step 3 and the relevant parameters z0, z1, y0, y1 of the intuitive fuzzy programming model to solve the intuitive fuzzy programming model and obtain a new optimal scheduling strategy and operating cost;

[0013] Step 6: Compare the operating cost calculated in step 2 with the operating cost obtained in step 5. If the absolute value of the difference between the two is less than the threshold, the most satisfactory optimization scheduling strategy is obtained; otherwise, continue to repeat steps 3 to 6 until the absolute value of the difference in the operating costs is less than the threshold.

[0014] Furthermore, the flexibility of each system in the distribution network is reflected in the climbing rate of the adjustable power in the distribution network, that is, the fluctuation amplitude of the net load curve cannot exceed the maximum climbing rate of the adjustable total power in the distribution network. The net load curve refers to the difference between the corresponding points in the time series between the daily load curve and the wind-solar combined output curve, and the corresponding curve in the time series composed of the difference.

[0015] Furthermore, the detailed calculation steps of the flexibility of the distribution network operation described in step 1 are as follows:

[0016] (a) Flexibility provided to the distribution network at the main grid access point

[0017] The main determining factors are: the capacity limit of the interconnection transformer, the upper limit of the climbing rate at the access point to the main grid set by the power company according to the actual situation, and the maximum climbing rate provided by the access point to the distribution network is the minimum of the two determining factors, namely:

[0018]

[0019]

[0020] Where: S 1max , They respectively represent the rated capacity and rated power factor of the interconnection transformer at the access point to the main grid; Represents the active power injected into the distribution network at the main grid access point during time period t; R Bupmax , R Bdnmax They respectively represent the maximum rising and falling ramp rates of active power injected into the distribution network from the main grid access point set by the power company according to actual conditions;

[0021] (b) Flexibility provided by EV charging stations to the distribution network

[0022] Its maximum climbing rate is mainly determined by the remaining power in the station and the charging and discharging power in the current period t, that is:

[0023]

[0024]

[0025] Where: P cimax , P dimax They represent the maximum charging power and maximum discharging power of the i-th electric vehicle charging station respectively; They represent the charging power and discharging power of the i-th electric vehicle charging station in time period t respectively; represents the remaining power of the i-th electric vehicle charging station in time period t; Respectively The upper and lower limits of

[0026] (c) Flexibility provided by demand response loads to the distribution network

[0027] It mainly depends on the upper and lower limits of the adjustable power of the demand response load and the total power consumption limit of the whole period. In the current period t, the demand response load climbing rate cannot make its power fluctuation amplitude exceed its adjustable potential. At the same time, the demand response load climbing rate cannot exceed the total power consumption of the whole period. The maximum climbing rate provided to the distribution network is:

[0028]

[0029]

[0030] Where: represents the actual active power of the jth demand response load in time period t; They represent the upper and lower limits of the adjustable power of the demand response load respectively; E DjΣ It represents the total power consumption of the jth demand response load in the whole period.

[0031] Furthermore, the optimization of the upper model in step 2 is carried out with the goal of minimizing the operating cost of the distribution network. Its objective function includes the cost of purchasing electricity from the superior power grid, the operating cost of the electric vehicle charging station, and the dispatch compensation cost of the demand response load. Its constraints include conventional power balance and safety constraints, and the initial optimal dispatch strategy is then obtained.

[0032] Furthermore, the objective function of the upper optimization model is:

[0033]

[0034] Where: N T represents the total number of time periods in the scheduling cycle; Δt represents the length of the time period; The unit cost of the distribution network purchasing electricity from the main grid access point during period t; Indicates the active power injected by the main network access point into the distribution network during time period t; K ci , K di They represent the cost coefficients of charging and discharging at electric vehicle charging stations respectively; They represent the charging and discharging power of the i-th electric vehicle charging station in time period t; K Dj represents the compensation cost of the jth demand response load per unit active power; represents the original power consumption of the jth demand response user before responding in time period t;

[0035] The constraints of the upper optimization model include: node power balance constraints, safety constraints, interconnection transformer capacity constraints, electric vehicle charging station charging and discharging constraints, and demand response load dispatchable potential constraints, which are as follows:

[0036] (a) Node power balance constraint:

[0037]

[0038] (b) Safety constraints:

[0039]

[0040] Where: Ui , U j , U k Represent the voltage amplitudes of nodes i, j, and k respectively; represents the active power on line i; They represent the upper and lower limits of active power on line i respectively; g jk 、b jk Respectively represent the conductance and susceptance of the i-th line. jk represents the voltage difference between nodes j and k;

[0041] (c) Connection transformer constraints:

[0042]

[0043] Where: represents the reactive power injected into the distribution network at the main grid access point during time period t;

[0044] (d) Electric vehicle charging and discharging constraints:

[0045]

[0046] Where: η ci , η di They represent the charging and discharging efficiency of the electric vehicle charging station respectively;

[0047] e) Demand response load constraints:

[0048]

[0049] Where: They respectively represent the upper and lower limits of the power consumption of the j-th demand response load in time period t.

[0050] Furthermore, the step three specifically includes:

[0051] The actual combined output of wind and solar power fluctuates within a certain error level range of the predicted combined output mean. The fluctuation range of the net load curve is obtained by the difference between the fluctuation range of the wind and solar power combined output curve at each moment and the load value at each moment on the internal load curve.

[0052] The most severe net load curve refers to the net load curve corresponding to the maximum sum of the parts whose climbing rates in each period of the entire dispatching cycle are greater than the maximum climbing rate of the total adjustable power of the distribution network within the fluctuation range of the net load curve. Its severity is measured by the index S, which represents the sum of the areas of the parts whose climbing rates in each period of the entire dispatching cycle are greater than the maximum climbing rate of the total adjustable power of the distribution network within the fluctuation range of the net load curve, that is:

[0053]

[0054] After the upper optimization model determines the optimal dispatching strategy with the minimum operating cost of the distribution network, the ramp rate of the adjustable power in each period of the entire cycle can be obtained. The lower optimization model searches for the most severe net load curve within the fluctuation range of the net load curve, and its expression is:

[0055]

[0056] Where: It represents the ramp rate of the most severe net load curve in time period t; It indicates the maximum and minimum power of the net load curve within the fluctuation range of time period t.

[0057] Furthermore, the flexibility constraint in step 4 means that within the fluctuation range of wind-solar combined output, the maximum ramp rate of the most severe net load curve in each period cannot exceed the maximum ramp rate of the total adjustable power of the distribution network, and its expression is:

[0058]

[0059] Where: Indicates the net load size of the most severe net load curve in time period t;

[0060] The fuzzy form corresponding to the flexibility constraint is:

[0061]

[0062]

[0063] Where:

[0064] The objective function in the upper optimization model considering flexibility constraints is converted into a corresponding fuzzy form using an intuitive fuzzy programming method, namely:

[0065]

[0066] Where: z0, z1, y0, y1 are the parameters of the intuitive fuzzy programming model;

[0067] Intuitive fuzzy programming considers membership function and non-membership function, that is, the difference between overall satisfaction and dissatisfaction is used as the objective function, so as to obtain an optimal scheduling strategy that satisfies both aspects at the same time;

[0068] Overall satisfaction α optimization model:

[0069]

[0070] Overall dissatisfaction β optimization model:

[0071]

[0072] The upper optimization model considering flexibility constraints can be transformed into an intuitive fuzzy programming form, namely:

[0073]

[0074] At this point, the upper-level optimization model considering flexibility constraints is transformed into an intuitive fuzzy programming model, and optimization iterations are performed based on it.

[0075] Furthermore, in step 5, z0 represents a conservative solution within an acceptable range, and z1 represents an optimal solution. The calculation model is shown in the following optimization problem:

[0076]

[0077]

[0078] Where: H1≤b represents the conventional constraint; H 2i ≤ b i and H 2i ≤ b i + p i represents a fuzzy constraint, where b i represents the threshold of the membership function corresponding to the i-th intuitionistic fuzzy constraint, p i represents the maximum acceptable deviation of the i-th fuzzy membership constraint;

[0079] The calculation model of y0 and y1 is shown in the following optimization problem:

[0080]

[0081]

[0082] Where: as well as represents a fuzzy constraint, where represents the threshold of the membership function corresponding to the i-th intuitionistic fuzzy constraint, represents the maximum acceptable deviation of the i-th fuzzy non-membership constraint.

[0083] The traditional alternating iterative method for solving the two-layer model can obtain an optimized dispatching strategy that meets the flexibility requirements of extreme scenarios, but the probability of extreme scenarios is extremely low. If the optimized dispatching strategy is obtained in this way, the corresponding dispatching cost will be relatively high. The present invention converts the upper-layer optimization model into an intuitive fuzzy programming form, and obtains the most satisfactory comprehensive solution that takes into account both economy and flexibility in most scenarios. By introducing intuitive fuzzy programming to solve the most satisfactory comprehensive solution of the model, the flexibility and economy of the distribution network can be taken into account. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 It is a flow chart of a two-layer optimization dispatching method for a distributed new energy distribution network taking flexibility constraints into consideration according to the present invention. DETAILED DESCRIPTION

[0085] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0086] The specific implementation of the method of the present invention includes:

[0087] Step 1: Calculate the flexibility of distribution network operation. The flexibility provided to the distribution network at the main grid access point includes the flexibility provided to the distribution network at the main grid access point, the flexibility provided to the distribution network by the electric vehicle charging station, and the flexibility provided to the distribution network by the demand response load.

[0088] The net load curve refers to the difference between the daily load curve and the wind-solar combined output curve at corresponding points in the time series, and the corresponding curve in the time series composed of the difference. The greater the fluctuation amplitude of the wind-solar combined output, the more violent the net load curve fluctuation will be. At this time, it is necessary to optimize the dispatching of flexible resources in the distribution network to smooth and absorb the violent fluctuations of the net load curve. The more flexible the distribution network is, the stronger its ability to respond quickly to system changes. The flexibility of each system in the distribution network is mainly reflected in the climbing rate of the adjustable power in the distribution network, that is, the fluctuation amplitude of the net load curve cannot exceed the maximum climbing rate of the adjustable total power in the distribution network, otherwise the distribution network will abandon wind, abandon light or even cut load due to its insufficient maximum climbing rate.

[0089] The total adjustable power in the distribution network is the sum of the adjustable power at the main network access point, the adjustable power of the electric vehicle charging station, and the adjustable power of the demand response load. The maximum increase and decrease climbing rates of the adjustable power in the distribution network at time t are:

[0090]

[0091]

[0092] Where: They represent the maximum rising and falling ramp rates of the total adjustable power of the distribution network in time period t respectively; It represents the maximum rising and falling ramp rates provided to the distribution network by the main grid access point in time period t; They represent the maximum rising and falling ramp rates that the ith electric vehicle charging station can provide to the distribution network in time period t respectively; They represent the maximum ramp rate and ramp rate that the j-th demand response load can provide to the distribution network during time period t; N S 、N D They represent the total number of electric vehicle charging stations and the total number of demand response loads in the distribution network, respectively.

[0093] The detailed calculation steps of the flexibility of the distribution network operation are as follows:

[0094] (a) Flexibility provided to the distribution network at the main grid access point

[0095] The main determining factors are: the capacity limit of the interconnection transformer and the upper limit of the ramp rate at the access point to the main grid set by the power company based on actual conditions. The maximum ramp rate provided to the distribution network at the access point to the main grid is the minimum of the two determining factors, namely:

[0096]

[0097]

[0098] Where: S 1max , They respectively represent the rated capacity and rated power factor of the interconnection transformer at the access point to the main grid; Represents the active power injected into the distribution network at the main grid access point during time period t; R Bupmax , R Bdnmax They respectively represent the maximum rising and falling ramp rates of active power injected into the distribution network at the main grid access point, which are set by the power company according to actual conditions.

[0099] (b) Flexibility provided by EV charging stations to the distribution network

[0100] Its maximum climbing rate is mainly determined by the remaining power in the station and the charging and discharging power in the current period t, that is:

[0101]

[0102]

[0103] Where: P cimax , P dimax They represent the maximum charging power and maximum discharging power of the i-th electric vehicle charging station respectively; They represent the charging power and discharging power of the i-th electric vehicle charging station in time period t respectively; E i t represents the remaining power of the i-th electric vehicle charging station in time period t; Respectively The upper and lower limits of .

[0104] (c) Flexibility provided by demand response loads to the distribution network

[0105] It mainly depends on the adjustable potential of the demand response load (i.e., the upper and lower limits of the adjustable power) and the total power consumption limit of the whole period. In the current period t, the demand response load climbing rate cannot make its power fluctuation amplitude exceed its adjustable potential, and the demand response load climbing rate cannot exceed the total power consumption of the whole period. The maximum climbing rate provided to the distribution network is:

[0106]

[0107]

[0108] Where: represents the actual active power of the jth demand response load in time period t; They represent the upper and lower limits of the adjustable power of the demand response load respectively; E DjΣ It represents the total power consumption of the jth demand response load in the whole period.

[0109] Step 2: Establish an upper-level optimization model without considering flexibility constraints to solve the dispatching strategy that minimizes the distribution network operation cost in a certain scenario of wind and solar combined output.

[0110] The optimization of the upper model is to minimize the operating cost of the distribution network. Its objective function includes the cost of purchasing electricity from the superior power grid, the operating cost of the electric vehicle charging station, and the dispatch compensation cost of the demand response load. Its constraints include conventional power balance and safety constraints, and the initial optimal dispatch strategy is obtained.

[0111] The objective function of the upper optimization model is:

[0112]

[0113] Where: N T represents the total number of time periods in the scheduling cycle; Δt represents the length of the time period; The unit cost of the distribution network purchasing electricity from the main grid access point during period t; Indicates the active power injected by the main network access point into the distribution network during time period t; K ci , K di They represent the cost coefficients of charging and discharging at electric vehicle charging stations respectively; They represent the charging and discharging power of the i-th electric vehicle charging station in time period t; K Dj represents the compensation cost of the jth demand response load per unit active power; Represents the original power consumption before the j-th demand response user responds in time period t.

[0114] The constraints include: node power balance constraints, safety constraints, interconnection transformer capacity constraints, electric vehicle charging station charging and discharging constraints, and demand response load dispatchable potential constraints, which are shown below respectively.

[0115] (a) Node power balance constraint:

[0116]

[0117] (b) Safety constraints:

[0118]

[0119] Where: U i , U j , U k Represent the voltage amplitudes of nodes i, j, and k respectively; represents the active power on line i; They represent the upper and lower limits of active power on line i respectively; g jk 、b jk Respectively represent the conductance and susceptance of the i-th line. jk Represents the voltage difference between nodes j and k.

[0120] (c) Connection transformer constraints:

[0121]

[0122] Where: It represents the reactive power injected into the distribution network at the main grid access point during time period t.

[0123] (d) Electric vehicle charging and discharging constraints:

[0124]

[0125] Where: η ci , η di They represent the charging and discharging efficiency of electric vehicle charging stations respectively.

[0126] e) Demand response load constraints:

[0127]

[0128] Where: They respectively represent the upper and lower limits of the power consumption of the j-th demand response load in time period t.

[0129] Based on this model, the initial optimization scheduling strategy can be solved.

[0130] Step 3: Establish a lower-level optimization model, based on the initial optimization scheduling strategy obtained in step 2 (if this step is looped, the optimization scheduling strategy here should be the latest iteration result), and solve the corresponding most severe net load curve.

[0131] Assuming that the actual wind-solar combined output fluctuates within a certain error level range of the predicted combined output mean, the net load curve fluctuation range is obtained by the difference between the fluctuation range of the wind-solar combined output curve at each moment and the load value at each moment on the internal load curve.

[0132] The most severe net load curve refers to the net load curve corresponding to the maximum sum of the parts whose climbing rates in each period of the entire dispatch cycle are greater than the maximum climbing rate of the total adjustable power of the distribution network within the fluctuation range of the net load curve. Its severity is measured by the index S, which represents the sum of the areas of the parts whose climbing rates in each period of the entire dispatch cycle are greater than the maximum climbing rate of the total adjustable power of the distribution network within the fluctuation range of the net load curve. That is:

[0133]

[0134] After the upper optimization model determines the optimal dispatching strategy with the minimum operating cost of the distribution network, it can obtain the ramp rate of the adjustable power in each period of the entire cycle. The lower optimization model searches for the most severe net load curve within the fluctuation range of the net load curve. Its expression is:

[0135]

[0136] Where: It represents the ramp rate of the most severe net load curve in time period t; It indicates the maximum and minimum power of the net load curve within the fluctuation range of time period t.

[0137] At this point, the optimal dispatching strategy for the distribution network that meets the flexibility constraint in the extreme scenario of wind-solar combined output can be obtained through the alternating iteration of the two-layer model. Although this optimal dispatching strategy meets the flexibility in the extreme scenario of wind-solar combined output, its probability of occurrence is not high, so it has to pay a higher economic cost.

[0138] Step 4: Use intuitionistic fuzzy programming to convert the upper model considering flexibility constraints into an intuitionistic fuzzy programming model.

[0139] In addition to conventional power balance and safety constraints, flexibility constraints are also considered in the constraints.

[0140] The flexibility constraint means that within the fluctuation range of wind-solar combined output, the maximum ramp rate of the most severe net load curve in each period cannot exceed the maximum ramp rate of the total adjustable power of the distribution network. Its expression is:

[0141]

[0142] Where: It represents the net load size of the most severe net load curve in time period t.

[0143] The fuzzy form corresponding to the flexibility constraint is:

[0144]

[0145]

[0146] Where:

[0147] The objective function in the upper optimization model considering flexibility constraints is converted into a corresponding fuzzy form using an intuitive fuzzy programming method, namely:

[0148]

[0149] Where: z0, z1, y0, y1 are the parameters of the intuitionistic fuzzy programming model.

[0150] In addition, ordinary fuzzy programming often only considers the membership function, while intuitive fuzzy programming starts from both positive and negative aspects, considering both the membership function and the non-membership function at the same time, that is, taking the difference between overall satisfaction and dissatisfaction as the objective function, so as to obtain an optimal scheduling strategy that satisfies both aspects at the same time.

[0151] Overall satisfaction α optimization model:

[0152]

[0153] Overall dissatisfaction β optimization model:

[0154]

[0155] The upper optimization model considering flexibility constraints can be transformed into an intuitive fuzzy programming form, namely:

[0156]

[0157] At this point, the upper-level optimization model considering flexibility constraints is transformed into an intuitive fuzzy programming model, and optimization iterations are performed based on it.

[0158] Step 5: Combine the most severe net load curve obtained in step 3 with the relevant parameters z0, z1, y0, y1 of the intuitive fuzzy programming model to solve the intuitive fuzzy programming model and obtain a new optimal scheduling strategy and operating cost.

[0159] z0 represents a conservative solution (worst solution) within the acceptable range, and z1 represents the optimal solution. Its calculation model is shown in the following optimization problem:

[0160]

[0161]

[0162] Where: H1≤b represents the conventional constraint; H 2i ≤ b i and H 2i ≤ b i + p i represents a fuzzy constraint, where b i represents the threshold of the membership function corresponding to the i-th intuitionistic fuzzy constraint, p i represents the maximum acceptable deviation of the i-th fuzzy membership constraint.

[0163] The calculation model of y0 and y1 is shown in the following optimization problem:

[0164]

[0165]

[0166] Where: as well as represents a fuzzy constraint, where represents the threshold of the membership function corresponding to the i-th intuitionistic fuzzy constraint, represents the maximum acceptable deviation of the i-th fuzzy non-membership constraint.

[0167] Step 6: Compare the operating cost calculated in step 2 with the operating cost obtained in step 5 (if this step is repeated, the operating cost corresponding to the latest optimization scheduling strategy should be compared with the operating cost corresponding to the previous optimization scheduling strategy). If the absolute value of the difference between the two is less than the threshold, the most satisfactory optimization scheduling strategy is obtained; otherwise, continue to repeat steps 3 to 6 until the absolute value of the difference in the operating costs is less than the threshold.

[0168] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A two-layer optimization scheduling method for a distributed renewable energy distribution network considering flexibility constraints, characterized in that: The following steps are involved: Step 1: Calculate the flexibility of each system in the distribution network, including the flexibility provided to the distribution network by the main grid access point, the flexibility provided to the distribution network by the electric vehicle charging station, and the flexibility provided to the distribution network by the demand response load; Step 2: Establish an upper-level optimization model without considering flexibility constraints to solve the dispatching strategy that minimizes the distribution network operation cost in a certain scenario of wind and solar combined output; Step 3: Establish a lower-level optimization model, and based on the initial optimization scheduling strategy obtained in step 2, solve the corresponding most severe net load curve; Step 4: Use intuitionistic fuzzy programming to convert the upper model considering flexibility constraints into an intuitionistic fuzzy programming model; Step 5: Combine the most severe net load curve obtained in step 3 and the relevant parameters z0, z1, y0, y1 of the intuitive fuzzy programming model to solve the intuitive fuzzy programming model and obtain a new optimal scheduling strategy and operating cost; Step 6: Compare the operating cost calculated in step 2 with the operating cost obtained in step 5. If the absolute value of the difference between the two is less than the threshold, the most satisfactory optimization scheduling strategy is obtained; otherwise, continue to repeat steps 3 to 6 until the absolute value of the difference in the operating costs is less than the threshold; The detailed calculation steps of the flexibility of distribution network operation described in step 1 are as follows: (a) Flexibility provided to the distribution network at the main grid access point The main determining factors are: the capacity limit of the interconnection transformer, the upper limit of the climbing rate at the access point to the main grid set by the power company according to the actual situation, and the maximum climbing rate provided by the access point to the distribution network is the minimum of the two determining factors, namely: Where: S 1max , They respectively represent the rated capacity and rated power factor of the interconnection transformer at the access point to the main grid; Represents the active power injected into the distribution network at the main grid access point during time period t; R Bupmax , R Bdnmax They respectively represent the maximum rising and falling ramp rates of active power injected into the distribution network from the main grid access point set by the power company according to actual conditions; (b) Flexibility provided by EV charging stations to the distribution network Its maximum climbing rate is mainly determined by the remaining power in the station and the charging and discharging power in the current period t, that is: Where: P cimax , P dimax They represent the maximum charging power and maximum discharging power of the i-th electric vehicle charging station respectively; They represent the charging power and discharging power of the i-th electric vehicle charging station in time period t respectively; represents the remaining power of the i-th electric vehicle charging station in time period t; Respectively The upper and lower limits of (c) Flexibility provided by demand response loads to the distribution network It mainly depends on the upper and lower limits of the adjustable power of the demand response load and the total power consumption limit of the whole period. In the current period t, the demand response load climbing rate cannot make its power fluctuation amplitude exceed its adjustable potential. At the same time, the demand response load climbing rate cannot exceed the total power consumption of the whole period. The maximum climbing rate provided to the distribution network is: Where: represents the actual active power of the jth demand response load in time period t; They represent the upper and lower limits of the adjustable power of the demand response load respectively; E DjΣ represents the total power consumption of the jth demand response load during the entire period; The optimization of the upper model in step 2 is to minimize the operating cost of the distribution network. Its objective function includes the cost of purchasing electricity from the upper power grid, the operating cost of the electric vehicle charging station, and the dispatch compensation cost of the demand response load. Its constraints include conventional power balance and safety constraints, and then the initial optimization dispatch strategy is obtained. The objective function of the upper optimization model is: Where: N T represents the total number of time periods in the scheduling cycle; Δt represents the length of the time period; The unit cost of the distribution network purchasing electricity from the main grid access point during period t; Indicates the active power injected by the main network access point into the distribution network during time period t; K ci , K di They represent the cost coefficients of charging and discharging at electric vehicle charging stations respectively; They represent the charging and discharging power of the i-th electric vehicle charging station in time period t; K Dj represents the compensation cost of the jth demand response load per unit active power; represents the original power consumption of the jth demand response user before responding in time period t; The constraints of the upper optimization model include: node power balance constraints, safety constraints, interconnection transformer capacity constraints, electric vehicle charging station charging and discharging constraints, and demand response load dispatchable potential constraints, which are as follows: (a) Node power balance constraint: (b) Safety constraints: Where: U i , U j , U k Represent the voltage amplitudes of nodes i, j, and k respectively; represents the active power on line i; They represent the upper and lower limits of active power on line i respectively; g jk , b jk Respectively represent the conductance and susceptance of the ith line; θ jk represents the voltage difference between nodes j and k; (c) Connection transformer constraints: Where: represents the reactive power injected into the distribution network at the main grid access point during time period t; (d) Electric vehicle charging and discharging constraints: Where: η ci , η di They represent the charging and discharging efficiency of the electric vehicle charging station respectively; e) Demand response load constraints: Where: They represent the upper and lower limits of the power consumption of the j-th demand response load in time period t respectively; The step three specifically includes: The actual combined output of wind and solar power fluctuates within a certain error level range of the predicted combined output mean. The fluctuation range of the net load curve is obtained by the difference between the fluctuation range of the wind and solar power combined output curve at each moment and the load value at each moment on the internal load curve. The most severe net load curve refers to the net load curve corresponding to the maximum sum of the parts whose climbing rates in each period of the entire dispatching cycle are greater than the maximum climbing rate of the total adjustable power of the distribution network within the fluctuation range of the net load curve. Its severity is measured by the index S, which represents the sum of the areas of the parts whose climbing rates in each period of the entire dispatching cycle are greater than the maximum climbing rate of the total adjustable power of the distribution network within the fluctuation range of the net load curve, that is: After the upper optimization model determines the optimal dispatching strategy with the minimum operating cost of the distribution network, the ramp rate of the adjustable power in each period of the entire cycle can be obtained. The lower optimization model searches for the most severe net load curve within the fluctuation range of the net load curve, and its expression is: Where: It represents the ramp rate of the most severe net load curve in time period t; Indicates the maximum and minimum power values ​​of the net load curve within the fluctuation range of time period t; The flexibility constraint in step 4 means that within the fluctuation range of wind-solar combined output, the maximum ramp rate of the most severe net load curve in each period cannot exceed the maximum ramp rate of the total adjustable power of the distribution network. Its expression is: Where: Indicates the net load size of the most severe net load curve in time period t; The fuzzy form corresponding to the flexibility constraint is: Where: The objective function in the upper optimization model considering flexibility constraints is converted into a corresponding fuzzy form using an intuitive fuzzy programming method, namely: Where: z0, z1, y0, y1 are the parameters of the intuitive fuzzy programming model; Intuitive fuzzy programming considers membership function and non-membership function, that is, the difference between overall satisfaction and dissatisfaction is used as the objective function, so as to obtain an optimal scheduling strategy that satisfies both aspects at the same time; Overall satisfaction α optimization model: Overall dissatisfaction β optimization model: The upper optimization model considering flexibility constraints can be transformed into an intuitive fuzzy programming form, namely: At this point, the upper-level optimization model considering flexibility constraints is transformed into an intuitive fuzzy programming model, and optimization iterations are performed based on it.

2. The two-layer optimization scheduling method for a distributed renewable energy distribution network considering flexibility constraints as claimed in claim 1 is characterized in that: The flexibility of each system in the distribution network is reflected in the climbing rate of the adjustable power in the distribution network, that is, the fluctuation amplitude of the net load curve cannot exceed the maximum climbing rate of the adjustable total power in the distribution network. The net load curve refers to the difference between the corresponding points in the time series between the daily load curve and the wind-solar combined output curve, and the corresponding curve in the time series composed of the difference.

3. The two-layer optimization scheduling method for a distributed renewable energy distribution network considering flexibility constraints as claimed in claim 1 is characterized in that: In step 5, z0 represents a conservative solution within the acceptable range, and z1 represents the optimal solution. The calculation model is shown in the following optimization problem: Where: H1≤b represents the conventional constraint; H 2i ≤ b i and H 2i ≤ b i +p i represents a fuzzy constraint, where b i represents the threshold of the membership function corresponding to the i-th intuitionistic fuzzy constraint, p i represents the maximum acceptable deviation of the i-th fuzzy membership constraint; The calculation model of y0 and y1 is shown in the following optimization problem: Where: as well as represents a fuzzy constraint, where represents the threshold of the membership function corresponding to the i-th intuitionistic fuzzy constraint, represents the maximum acceptable deviation of the i-th fuzzy non-membership constraint.

Images