Power supply unit division method considering demand response master-slave game

Abstract

The invention discloses a power supply unit division method considering a demand response master-slave game, relates to the technical field of power supply partition planning and division of a power distribution network, and provides a power supply unit division combination solving framework considering the demand response master-slave game by considering the influence of demand response on power supply unit division. An incentive demand response master-slave game model of a power grid company and a power consumer is established, and a fast particle swarm optimization nested mixed integer linear programming method is provided for solving; carrying out feeder line block division by using a rotation center line distance weighted alternative positioning algorithm, and combining the blocks into power supply units according to the bordering degree; and finally, a power supply unit division overall model considering the demand response master-slave game is established by integrating the contents, an optimal division method is provided, and an optimal power supply unit division scheme and a corresponding demand response strategy are obtained through multiple iterations. And the economical efficiency of the power supply unit division scheme is improved on the premise of ensuring the N-1 safety criterion.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply zoning planning and division of a distribution network, and in particular to a power supply unit division method considering a master-slave game of demand response. Background Art

[0002] The power supply unit is a power supply level further subdivided on the basis of the power supply grid. It is usually composed of 1-3 groups of 10kV feeder interconnections. Factors such as load characteristics and communication methods need to be considered when dividing. The power supply unit is the basic unit for grid analysis, line communication structure determination and project planning. The target grid of the power supply grid and the wiring plan of the medium-voltage trunk line need to be implemented at this level.

[0003] Traditional unit division relies on the experience of planners. Unit division can be carried out according to the principles of "complete terrain, clear boundaries, and moderate management". It can also be divided through three division methods: power grid method, road network method, and plot method. However, these division methods do not have strict standards and may lead to different planning results. In order to make the results of unit division more unified and standardized, researchers have proposed different standardized division ideas. The most common one is the advanced feeder block division, which is then combined to form a unit. Some scholars have considered the access of distributed new energy sources, load peak-valley coupling, and uncertainty in the new distribution system in their research on power supply unit division, but there are few literatures studying the impact of demand response regulation potential on power supply unit division.

[0004] With the continuous development of new distribution systems, a large number of flexible and adjustable loads are connected to the distribution network, and the digitalization and intelligence level of the user side is constantly improving. Demand response will inevitably develop rapidly. Demand response refers to the situation where electricity users respond according to prices or incentives when electricity prices are high or system reliability is at risk, changing their own electricity consumption characteristics to achieve the effect of peak shaving and valley filling. Therefore, in the grid planning of distribution networks, fully tapping the demand response regulation capabilities of flexible loads can achieve peak shaving and valley filling of energy consumption curves, which is of great significance for optimizing planning results. However, how to consider the interest game of multiple subjects in demand response in the division of power supply units, obtain the price and optimal scheduling plan of demand response, and then make the net load of the feeder meet the N-1 safety criterion and achieve the optimal economic cost is a problem that needs to be considered in unit division.

[0005] Therefore, the optimal division method of power supply units considering the master-slave game of demand response has important research significance while taking into account both economic cost and N-1 safety criterion. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In view of the shortcomings of the prior art, the present invention provides a power supply unit division method considering the master-slave game of demand response, establishes a power supply unit division model considering the master-slave game of demand response and proposes a solution method, which solves the problems raised in the background technology.

[0008] (II) Technical solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0010] A method for dividing power supply units considering demand response master-slave game includes the following steps:

[0011] S1. According to the constraint conditions, the initial number of power supply units is set to a, that is, the minimum number of power supply units to be divided Set the initial demand response contract ratio to 0;

[0012] S2. Build a master-slave game model for demand response;

[0013] S3, using the rotation centerline distance weighted alternating positioning algorithm to divide the feeder blocks, and check whether the division result meets the N-1 criterion; under the condition that the N-1 criterion is not met, the demand response contract ratio of the model in S2 is increased by a step size of ε, and S2 is repeated until the demand response scheduling result obtained by S2 makes the feeder block division result meet the N-1 criterion; under the condition that the N-1 criterion is met, the feeder blocks are combined into power supply units according to the degree of feeder bordering;

[0014] S4, save the results obtained in S2 and S3, establish the overall model of power supply unit division and calculate the objective function value of each result;

[0015] S5. Increase the number of power supply units, initialize the demand response contract ratio to 0, repeat S2 to S4, and obtain results corresponding to different numbers of units;

[0016] S6. Compare the results obtained in S5 and select the one with the best model objective function as the final solution.

[0017] Furthermore, in the overall framework of S1 to S6, in order to establish a power supply unit division and combination solution framework for demand response master-slave game and perform data initialization, the framework decouples planning and operation, and continuously iterates the three optimization problems of demand response master-slave game, feeder block division and combination to obtain the optimal power supply unit division result and its corresponding demand response strategy; the power supply unit is composed of feeder blocks, and the feeder extends from the substation. The unit composed of adjacent feeder blocks of the same substation is called an intra-station unit, and the unit composed of feeder blocks of different substations is called an inter-station unit.

[0018] Furthermore, in S2, the upper layer of the demand response master-slave game model is the DNO decision layer, which is used to transmit the demand response incentive strategy to the lower layer; the lower layer is the load decision layer, which is used to derive the demand response scheduling plan with the lowest cost based on the incentive strategy of the upper layer, and then feed it back to the upper layer, and continuously iterate this process until an equilibrium solution is obtained.

[0019] Furthermore, in S2, the mathematical expression of the demand response master-slave game model is as follows:

[0020] Upper level, DNO decision-making level:

[0021] Objective function: The upper layer aims to minimize the operating cost of DNO, and its mathematical expression is:

[0022] min C DNO =C DR +C VTP +C buyup -C sell

[0023]

[0024]

[0025]

[0026]

[0027] In this formula, C DNO , C DR , C VTP , C buyup , C sell They represent the total operating cost of DNO, the demand response incentive cost, the peak-valley difference penalty, the cost of purchasing electricity from the upper power grid, and the income from selling electricity to lower-level users; c dr,t 、c VTP 、c buyup 、c sell They represent the unit demand response incentive cost, the unit peak-valley difference penalty, the unit cost of purchasing electricity from the upper power grid, and the unit profit of selling electricity to lower-level users; P t and P a,t They represent the net load at time t before and after demand response, respectively; and They represent the peak and valley values ​​of the net load energy consumption curve after demand response respectively;

[0028] Constraints:

[0029]

[0030]

[0031] In this formula, and are the minimum and maximum incentive prices respectively; T p The peak period of electricity consumption; dr,t represents the incentive price at time t;

[0032] Lower level, power user decision-making level:

[0033] Objective function: The lower layer includes transferable loads and reducible loads, with the goal of minimizing its own operating cost. The mathematical expression is:

[0034]

[0035] In this formula, C load is the operating cost of the power user; α and β are both demand response dispatch cost coefficients; ΔP load,t It indicates the change in electricity consumption at time t before and after the user demand response scheduling. When it is greater than 0, it means that the electricity consumption at time t is reduced; c sell The unit price of electricity purchased by power users; P aload,t is the electricity consumption of the power user at time t after demand response;

[0036] Constraints:

[0037] P load,t -P aload,t =ΔP load,t

[0038]

[0039] 0≤ΔP loadcut,t ≤η dr P loadcut,t

[0040] In this formula, P load,t , P aload,t They represent the power consumption before and after the power user demand response dispatch at time t; η dr The proportion of demand response contracts signed between DNO and electricity users.

[0041] Furthermore, S2 also includes: using the fast particle swarm algorithm to nest the linear programming master-slave game model solution process, taking the upper DNO demand response incentive strategy as the particle, and the upper objective function as the fitness of the particle population; the particle position update formula is:

[0042]

[0043] In this formula, S i,j represents the j-th dimension variable of the ith particle; fbest jrepresents the j-th dimension variable of the global optimal particle; β represents the convergence coefficient, and its value range is [0, 1]; α is the particle attenuation coefficient, and the α of each iteration is smaller than that of the previous iteration; rand() represents a normally distributed random number; the lower-level power user model is embedded in the population fitness solution process as a mixed integer linear programming model, and can be solved using the solver.

[0044] Furthermore, in S3, the feeder block division is performed using the rotation centerline distance weighted alternating positioning algorithm, and the steps are as follows:

[0045] S301. According to the number of power supply units m, the initial positions of the center lines of 2m feeder blocks are randomly selected. The angles between the feeder lines are

[0046] S302, the group of initial feeder center lines is stepped Rotate clockwise to obtain the position of the center line of n groups of feeders;

[0047] S303, starting from the center line of the first group of feeders, dividing according to the Euclidean distance from the load to the center line of the feeder block, then updating the division result by weighted distance, updating the center line position according to the division result, and continuously iterating this process until the center line position no longer changes;

[0048] S304, dividing the n groups of feeder center lines in S303 to obtain n power supply unit division schemes;

[0049] S305. Select the optimal solution based on the comprehensive indicators, and take the power supply unit division solution with the best comprehensive indicators as the optimal solution.

[0050] Furthermore, in S303, the weighted distance calculation formula from the load to the center line of the feeder block is as follows:

[0051] l i ' ,j =ω 1,i,j ω 2,i,j l i,j

[0052]

[0053]

[0054] In this formula, l i ' ,j and l i,j are the weighted distance and Euclidean distance from load j to the center line of feeder block i, respectively; ω 1,i,j and ω 2,i,j are the peak and variance weighting coefficients, respectively; and They respectively represent the peak value and variance of the net load curve within the block when load j is divided into feeder block i; k1 and k2 are both weighting factors.

[0055] Furthermore, in S305, the comprehensive index calculation formula of the scheme is as follows:

[0056] F=αLR+βLB+γPVD

[0057]

[0058]

[0059]

[0060] In this formula, F is the comprehensive evaluation index of the unit division scheme, α, β and γ are weight coefficients respectively; LR is the average feeder load rate index; LB is the maximum load balance index; PVD is the average feeder peak-to-valley difference index; n is the number of feeder blocks; λ i is the feeder load rate of block i; S i is the maximum load in block i, S max is the maximum load in n blocks; and is the peak and valley value of the net load in block i; the comprehensive index F is positively correlated with the optimization degree of the feeder block division scheme.

[0061] Furthermore, in S3, under the condition of only considering a single connection, that is, two feeders are connected at the end by a tie switch, and two feeder blocks form an inter-station unit or an intra-station unit. In the feeder block division stage, the number of feeder blocks in each substation is set to be an even number; before the feeder blocks are combined, the number of inter-station and intra-station units is determined; in the feeder block combination process, the feeder blocks are first combined into inter-station units according to the degree of bordering, and then combined into intra-station units according to the adjacent relationship;

[0062] The objective function of combining feeder blocks into inter-station units is as follows:

[0063]

[0064] Where, BOR is the boundary degree index of the inter-station unit combination scheme; len j is the length of the two feeder blocks of the jth inter-station unit under the combined scheme; tlen j is the total length of the feeder block boundary of the jth inter-station unit under this combination scheme.

[0065] There are several ways to combine feeder blocks into inter-station energy units. After listing all inter-station unit combination schemes, their border degree index BOR is compared. The one with the largest BOR is the optimal combination scheme, that is, the combination scheme that determines the adjacent energy units in the station in pairs.

[0066] Furthermore, in S4, the overall model of power supply unit division is established. The decision variables of the model are the number of energy units, the division results and the corresponding optimal demand response strategy. The overall objective function of the model is the economic cost of the unit division scheme, including the feeder construction cost, the demand response incentive cost and the power sales revenue loss of the power grid company due to demand response. The calculation formula is as follows:

[0067] minC=C line +C DR +C loss

[0068]

[0069]

[0070]

[0071] In the formula, c line 、c dr,t 、c buyup and c sell are the unit construction cost of the feeder, the unit incentive price of demand response, the unit cost of purchasing electricity from the upper power grid, and the unit cost of selling electricity to the load; b is the number of years the equipment has been in operation, r is the discount rate; L ij represents the line length in the jth feeder block of the i-th substation; P t and P a,t They correspond to the net load power of the grid at time t before and after the demand response respectively; μ represents the proportion of the days of demand response in the whole year;

[0072] The constraints include N-1 safety criterion constraint and feeder quantity constraint. The specific formula is as follows:

[0073]

[0074]

[0075] In the formula, represents the maximum load of the jth feeder of the i-th substation; S line Indicates the capacity of the feeder; and P max They are the net load of the grid after the maximum demand response scheduling and the net load of the grid without demand response respectively; the number of power supply units is m, and the total number of feeder blocks is 2m.

[0076] Also included is an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, the processor being used to call and run the computer program stored in the memory to execute a method for optimizing the division of power supply units considering the master-slave game of demand response;

[0077] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements any step in the above method for optimizing the division of power supply units taking into account the master-slave game of demand response.

[0078] (III) Beneficial effects

[0079] The present invention provides a method for dividing power supply units considering the master-slave game of demand response, which has the following beneficial effects:

[0080] (1) The power supply unit optimization division method designed in this scheme considering demand response can improve the load energy curve by performing a certain degree of demand response, thereby reducing the number of power supply units and improving the economy of the power supply unit division scheme while ensuring the N-1 safety criterion.

[0081] (2) The demand response master-slave game model established in this scheme can take into account the response willingness of power users. Through the game, the user response amount can meet the power supply unit division requirements, and the demand response cost paid by the power grid company is optimal.

[0082] (3) The fast particle swarm nested mixed integer linear programming method designed in this scheme can make full use of the characteristics of each layer of the model and solve the demand response master-slave game model quickly and accurately. The convergence speed is improved compared with the traditional particle swarm algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 This is a power supply unit division and combination solution framework diagram considering the master-slave game of demand response of the present invention;

[0084] Figure 2 This is a framework diagram of the demand response master-slave game model proposed by the present invention;

[0085] Figure 3 A flowchart for solving the demand response master-slave game model proposed by the present invention;

[0086] Figure 4 This is a flow chart of the power supply unit optimization division method considering the master-slave game of demand response proposed by the present invention;

[0087] Figure 5 The substation and load distribution within the power supply grid to be divided;

[0088] Figure 6 This is a schematic diagram of the power supply unit division results;

[0089] Figure 7 Grid net load curve corresponding to different demand response contract ratios. DETAILED DESCRIPTION

[0090] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0091] Research and development concept:

[0092] The present invention relates to a method for optimizing the division of power supply units considering the master-slave game of demand response. Firstly, considering the influence of demand response on the division of power supply units, a combined solution framework for the division of power supply units considering the master-slave game of demand response is proposed. Then, an incentive-type master-slave game model of demand response between power grid companies and power users is established, and a method for solving the problem by nesting mixed integer linear programming of a fast particle swarm algorithm is proposed. Thirdly, a weighted alternating positioning algorithm based on the distance of the rotating center line is used to divide feeder blocks, and the blocks are combined into power supply units according to the degree of connection. Finally, based on the above contents, an overall model for the division of power supply units considering the master-slave game of demand response is established, and an optimization division method is proposed. After multiple iterations, the optimal power supply unit division scheme and the corresponding demand response strategy are obtained.

[0093] The optimal division method of power supply units considering the master-slave game of demand response proposed in this scheme includes:

[0094] (1) A framework for solving the combination of power supply units considering the master-slave game of demand response;

[0095] (2) Demand response master-slave game model and solution method (the first optimization problem, i.e., demand response master-slave game model and solution method);

[0096] (3) Feeder block division method based on the rotation centerline distance weighted alternating positioning algorithm (the second optimization problem, i.e., feeder block division method based on the rotation centerline distance weighted alternating positioning algorithm);

[0097] (4) Block combination method based on block adjoining degree (the third optimization problem, i.e., block combination method based on block adjoining degree);

[0098] (5) Overall model and method of power supply unit division considering the master-slave game of demand response (integrating the above three optimization problems)

[0099] See also Figures 1 to 7 This embodiment provides a method for dividing power supply units considering the master-slave game of demand response, and the specific steps include:

[0100] S1. According to the constraint conditions, the initial number of power supply units is set to a, that is, the minimum number of power supply units to be divided The initial demand response contract ratio is set to 0; data is initialized based on problem point (1) "power supply unit division and combination solution framework considering demand response master-slave game";

[0101] In the overall framework of S1 to S6, this step proposes a power supply unit partition combination solution framework considering the master-slave game of demand response;

[0102] The framework decouples planning and operation, splits the solution of complex problems into the combination of multiple optimization problems, and continuously iterates the three optimization problems of demand response master-slave game, feeder block division and combination to obtain the optimal power supply unit division result and its corresponding demand response strategy; the specific contents are as follows:

[0103] (1) The master-slave game of demand response is carried out at the power supply grid level, taking into account the respective interests of the power grid company and the power users, and optimizing the demand response of the reducible load and the transferable load within the grid; on the premise that the power supply unit division result can meet the N-1 criterion, the demand response contract ratio, incentive price at peak power consumption time and dispatch strategy of the game equilibrium are solved; the meaning of game equilibrium is that the contract ratio and incentive price formulated by the power grid company can minimize the demand response cost of the power grid company, and the dispatch strategy of the power users under the contract ratio and incentive price also optimizes the interests of the power users; the load curve after demand response is used as the boundary condition for the division of the power supply unit.

[0104] (2) The power supply unit is divided based on the power supply grid according to the load curve after demand response;

[0105] The power supply unit is composed of feeder blocks. The feeders extend from the substation. The unit composed of adjacent feeder blocks of the same substation is called an intra-station unit, and the unit composed of feeder blocks of different substations is called an inter-station unit. Therefore, the division of power supply units is ultimately a problem of feeder block division and combination. When the number of power supply units is certain, if the division result cannot meet the N-1 criterion, the demand response contract ratio will be increased to continue to optimize the demand response result.

[0106] S2. Build a master-slave game model for demand response. Based on problem point (2) "master-slave game model and solution method for demand response", the master-slave game for demand response is carried out. First, a master-slave game model for demand response between power grid companies and power users is established. Second, the model is solved using the fast particle swarm nested mixed integer linear programming method. Finally, the incentive pricing for demand response and the load curve after dispatch are obtained.

[0107] In this step, it is assumed that there are multiple types of power loads in the power grid, mainly transferable loads and curtailable loads; the distribution network operator adopts an incentive-based demand response mechanism, signs a demand response contract with load users, and determines the demand response contract ratio and the demand response unit incentive subsidy price; power users will perform the optimal demand response according to the incentive price set by the distribution network operator;

[0108] Based on the above analysis, a master-slave game model between DNO and power users (including curtailable load and transferable load) is established; the upper layer of the model is the DNO decision layer, which takes its own cost minimum as its goal and transmits the demand response incentive strategy to the lower layer; the lower layer is the load decision layer, which will derive the demand response scheduling plan with the lowest cost according to the incentive strategy of the upper layer, and then feed it back to the upper layer, and continuously iterate this process until an equilibrium solution is obtained;

[0109] The mathematical expression of the model is as follows:

[0110] Upper layer: DNO decision-making level

[0111] (1) Objective function: The upper layer aims to minimize the operating cost of DNO, and its mathematical expression is:

[0112] min C DNO =C DR +C VTP +C buyup -C sell

[0113]

[0114]

[0115]

[0116]

[0117] In the formula, C DNO , C DR , C VTP , C buyup , C sell They represent the total operating cost of DNO, the demand response incentive cost, the peak-valley difference penalty, the cost of purchasing electricity from the upper power grid, and the income from selling electricity to lower-level users; cdr,t 、c VTP 、c buyup 、c sell They represent the unit demand response incentive cost, the unit peak-valley difference penalty, the unit cost of purchasing electricity from the upper power grid, and the unit profit of selling electricity to lower-level users; P t and P a,t They represent the net load at time t before and after demand response, respectively; and They represent the peak and valley values ​​of the net load energy consumption curve after demand response respectively;

[0118] (2) Constraints:

[0119]

[0120]

[0121] In the formula, and are the minimum and maximum incentive prices respectively; T p The peak period of electricity consumption; dr,t represents the incentive price at time t; the second constraint is to determine the peak and valley values ​​of the net load curve, and this constraint is a linear constraint;

[0122] Lower level: Electricity user decision-making level

[0123] (1) Objective function: The lower layer includes transferable loads and reducible loads, with the goal of minimizing its own operating cost. The mathematical expression is:

[0124]

[0125] In the formula, C load is the operating cost of the power user; α and β are both demand response dispatch cost coefficients; ΔP load,t It indicates the change in electricity consumption at time t before and after the user demand response scheduling. When it is greater than 0, it means that the electricity consumption at time t is reduced; c sell The unit price of electricity purchased by power users; P aload,t is the electricity consumption of the power user at time t after demand response;

[0126] (2) Constraints:

[0127] P load,t -P aload,t =ΔP load,t

[0128]

[0129] 0≤ΔP loadcut,t ≤η drP loadcut,t

[0130] Where P load,t , P aload,t They represent the electricity consumption at time t before and after the power user's demand response scheduling; ηdr is the demand response ratio signed between the DNO and the power user; the second constraint is the demand response constraint for transferable loads and the total electricity consumption constraint, and the third constraint is the demand response constraint for curtailable loads.

[0131] According to the above model, a fast particle swarm algorithm nested linear programming master-slave game model solving method is proposed. This method takes the upper DNO demand response incentive strategy as the particle and the upper objective function as the fitness of the particle population. The particle position update formula is:

[0132]

[0133] In the formula, S i,j represents the j-th dimension variable of the ith particle; fbest j represents the j-th dimension variable of the global optimal particle; β represents the convergence coefficient, and its value range is [0, 1]. The closer it is to 1, the faster the convergence speed; α is the particle attenuation coefficient, and the α of each iteration is smaller than that of the previous iteration; rand() represents a random number with a normal distribution; the lower-level power user model is embedded in the population fitness solution process as a mixed integer linear programming model, and can be solved using the solver;

[0134] It should be noted that the corresponding letters in different formulas represent corresponding meanings. For coefficients or factors such as α, β and γ, different formulas have different corresponding meanings. Therefore, different formulas give specific instructions and explanations for α, β and γ in this formula, and there will be no ambiguity.

[0135] S3, using the rotating center line distance weighted alternating positioning algorithm to divide the feeder blocks, and check whether the division result meets the N-1 criterion; under the condition that the N-1 criterion is not met, increase the demand response contract ratio of the model in S2 by a step size of ε, and repeat S2 until the demand response scheduling result obtained by S2 makes the feeder block division result meet the N-1 criterion; under the condition that the N-1 criterion is met, the feeder blocks are combined into power supply units according to the degree of feeder adjacency; based on problem point (3) "feeder block division method based on rotating center line distance weighted alternating positioning algorithm" and problem point (4) "block combination method based on block adjacency degree", feeder block division and combination are performed;

[0136] Among them, the feeder block division is performed using the rotation centerline distance weighted alternating positioning algorithm. The specific steps are as follows:

[0137] S301. According to the number of power supply units m, the initial positions of the center lines of 2m feeder blocks are randomly selected. The angles between the feeder lines are

[0138] S302, the center line of this group of initial feeders is stepped Rotate clockwise to obtain the position of the center line of n groups of feeders;

[0139] S303, starting from the center line of the first group of feeders, divide the feeder block according to the Euclidean distance from the load to the center line, then update the division result by weighted distance, update the center line position according to the division result, and iterate this process until the center line position no longer changes; wherein, the weighted distance calculation formula from the load to the center line of the feeder block is as follows:

[0140] l i ' ,j =ω 1,i,j ω 2,i,j l i,j

[0141]

[0142]

[0143] In the formula, l i ' ,j and l i,j are the weighted distance and Euclidean distance from load j to the center line of feeder block i, respectively; ω 1,i,j and ω 2,i,j are the peak and variance weighting coefficients, respectively; and They represent the peak value and variance of the net load curve in the feeder block i when load j is divided into the feeder block i; k1 and k2 are weighting factors; weighting factor ω 1,i,j and ω 2,i,j The meaning of is that when load j is divided into feeder block i, the larger the peak value and variance of the load curve of the block, ω 1,i,j and ω 2,i,j The larger it is, the more disadvantageous it is for the load to be allocated to this feeder block;

[0144] S304, performing the division in step 3 on the center lines of n groups of feeders to obtain n power supply unit division schemes;

[0145] S305, selecting a scheme based on comprehensive indicators, and taking the power supply unit division scheme with the best comprehensive indicators as the best scheme;

[0146] The calculation formula of the comprehensive index of the scheme is as follows:

[0147] F=αLR+βLB+γPVD

[0148]

[0149]

[0150]

[0151] Where F is the comprehensive evaluation index of the unit division scheme; LR is the average feeder load rate index, which indicates the average daily load rate of all feeders, so as to evaluate the utilization rate of feeders; LB is the maximum load balance index, which indicates the balance of load scales in different feeder blocks; PVD is the average feeder peak-to-valley difference index, which indicates the volatility of the net load curve in the feeder block; n is the number of feeder blocks; λ i is the feeder load rate of block i; S i is the maximum load in block i, S max is the maximum load in n blocks; and are the peak and valley values ​​of the net load in block i;

[0152] It can be seen from the above formula that the value range of the three evaluation indicators is [0, 1]; the closer the LR, LB, and PVD values ​​of the feeder block division scheme are to 1, the higher the utilization rate of each feeder, the more balanced the load scale within the feeder block, and the smaller the peak-to-valley difference of the net load curve of the feeder block; therefore, the larger the comprehensive index F is, the better the feeder block division scheme is;

[0153] S4. Save the results obtained in S2 and S3, establish an overall model for power supply unit division and calculate the objective function values ​​of each result; save the power supply unit division scheme and demand response scheduling scheme obtained in S2 and S3, and calculate the objective function values ​​of each scheme according to the overall model for power supply unit division established in problem point (5) "Overall model and method for power supply unit division considering master-slave game of demand response";

[0154] In this step, an overall model is established by integrating the above three optimization problems. The decision variables of the model are the number of energy units, the division results and the corresponding optimal demand response strategy. The overall objective function of the model is the economic cost of the unit division scheme, including the feeder construction cost, the demand response incentive cost and the power sales revenue loss of the power grid company caused by demand response. The calculation formula is as follows:

[0155] minC=C line +C DR +C loss

[0156]

[0157]

[0158]

[0159] In the formula, c line 、c dr,t 、c buyup and c sell are the unit construction cost of the feeder, the unit incentive price of demand response, the unit cost of purchasing electricity from the upper power grid, and the unit cost of selling electricity to the load; b is the number of years the equipment has been in operation, r is the discount rate; L ij represents the line length in the jth feeder block of the i-th substation; P t and P a,t They correspond to the net load power of the grid at time t before and after the demand response respectively; μ represents the proportion of the number of days for demand response in the whole year; min represents the minimum value of the function, and the capital C is the economic cost of the unit division scheme. It should be noted that in the capital C, C line Refers to the equivalent annual cost of feeder construction. In the lowercase c, c l i ne is the unit construction cost of the feeder, that is, the feeder construction cost per kilometer (the unit construction cost of the feeder);

[0160] The constraints include N-1 safety criterion constraint and feeder quantity constraint. The specific formula is as follows:

[0161]

[0162]

[0163] In the formula, represents the maximum load of the jth feeder of the i-th substation; S line Indicates the capacity of the feeder; and P max are the net load of the grid after the maximum demand response scheduling and the net load of the grid without demand response respectively; the number of power supply units is m, and the total number of feeder blocks is 2m; in order to ensure the energy supply reliability of the energy unit, it is considered that the feeder needs to meet the N-1 safety criterion, that is, for the two interconnected feeders in a unit, if one feeder fails, all the loads in the unit will be transferred by the other feeder. Therefore, if the two feeders have the same capacity, the load rate of each feeder should not exceed 50% during normal operation.

[0164] S5. Increase the number of power supply units, initialize the demand response contract ratio to 0, repeat S2 to S4, and obtain results corresponding to different numbers of units;

[0165] S6. Compare the results obtained in S5 and select the one with the best model objective function as the final solution.

[0166] Embodiment: The proposed model and solution algorithm can be applied to the unit division of the power grid. The substations and load distribution in the grid are as follows: Figure 5 As shown; there are 243 loads and distributed power sources in the grid, among which distributed power sources are mainly distributed photovoltaics, and users include conventional electricity users. Conventional electricity users are divided into four categories: administrative, residential, commercial, and industrial. It is believed that residential loads have the ability to transfer demand response, and commercial and administrative loads have the ability to reduce demand response.

[0167] Combination Figure 1 The overall solution process of the power supply unit optimization division method considering the master-slave game of demand response proposed by the present invention is described in detail, and the specific steps are as follows:

[0168] Step 1: Input the spatial distribution and ownership information of loads, distributed generation and substations, as well as the energy consumption curves of various loads and the output curves of distributed generation and other example information; wherein the example information is the content described in the previous paragraph and the attached figure, including the spatial distribution and ownership information of loads, distributed generation and substations, as well as the energy consumption curves of various loads and the output curves of distributed generation;

[0169] Step 2: Data initialization, setting the initial number of power supply units and the initial demand response contract ratio;

[0170] Step 3: Conduct a master-slave game of demand response to obtain the incentive pricing of demand response and the load curve after dispatch;

[0171] Step 4: Use the rotating centerline distance weighted alternating positioning algorithm to divide the feeder blocks and check whether the division result meets the N-1 criterion. If not, increase the demand response contract ratio by a step length of ε and repeat step 2. If it meets the criterion, combine the feeder blocks into power supply units according to the degree of bordering;

[0172] Step 5: Save the power supply unit division scheme and demand response scheduling scheme obtained in steps 2 and 3, and calculate the objective function value of the scheme;

[0173] Step 6: Increase the number of power supply units, initialize the demand response contract ratio, repeat steps 3-5, and obtain the optimal power supply unit division scheme and the corresponding demand response scheduling scheme corresponding to different numbers of units;

[0174] Step 7: Compare the various solutions obtained in step 6 and select the one with the best model objective function as the final solution.

[0175] Among them, the final power supply unit division scheme obtained in step 7 is as follows Figure 6 As shown, the demand response dispatch curve corresponding to this scheme in step 3 is as follows Figure 7 As shown;

[0176] In this embodiment, compared with the unit division without considering demand response, the method proposed in the present invention reduces the total cost of the power supply unit division scheme by 20%, and increases the average load rate of the feeder by nearly 5%; it can be seen that the method proposed in the present invention can improve the net load curve through a certain degree of demand response, tap the power supply potential of the feeder, reduce the number of unit divisions, divide the power supply grid into a smaller number of power supply units under the premise of meeting the N-1 criterion, and improve the economy of the power supply unit division scheme.

[0177] In the application, the several formulas involved are all calculated by taking their numerical values ​​after removing the dimensions, and the formula is established by collecting a large amount of data and performing software simulation to obtain a formula for the most recent actual situation. Some coefficients or weights in the formula are set by technical personnel in this field according to actual conditions, and the weight coefficient can also be determined by the coefficient of variation method. The coefficient of variation method is a method of weighting each indicator according to the degree of variation between the current value and the target value of each evaluation indicator; if the numerical difference of an indicator is large, it can clearly distinguish the evaluated objects, indicating that the indicator has rich discrimination information, and therefore the indicator should be given a larger weight; conversely, if the numerical difference of each evaluated object on a certain indicator is small, then the ability of this indicator to distinguish between the evaluated objects is weak, and therefore the indicator should be given a smaller weight; this method directly uses the information contained in each indicator to obtain the weight of the indicator through calculation, so it is objective.

[0178] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0179] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0180] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be covered by the protection scope of the present application.

Claims

1. A method for dividing power supply units considering demand response master-slave game, characterized in that: The steps include: S1. According to the constraint conditions, the initial number of power supply units is set to a, that is, the minimum number of power supply units to be divided Set the initial demand response contract ratio to 0; S2. Build a master-slave game model for demand response; S3, using the rotation centerline distance weighted alternating positioning algorithm to divide the feeder blocks, and check whether the division result meets the N-1 criterion; under the condition that the N-1 criterion is not met, the demand response contract ratio of the model in S2 is increased by a step size of ε, and S2 is repeated until the demand response scheduling result obtained by S2 makes the feeder block division result meet the N-1 criterion; under the condition that the N-1 criterion is met, the feeder blocks are combined into power supply units according to the degree of feeder bordering; S4, save the results obtained in S2 and S3, establish the overall model of power supply unit division and calculate the objective function value of each result; S5. Increase the number of power supply units, initialize the demand response contract ratio to 0, repeat S2 to S4, and obtain results corresponding to different numbers of units; S6. Compare the results obtained in S5 and select the one with the best model objective function as the final solution.

2. A method for dividing power supply units considering demand response master-slave game according to claim 1, characterized in that: In the overall framework of S1 to S6, in order to establish the power supply unit division and combination solution framework of the demand response master-slave game and perform the data initialization process, the framework decouples planning and operation, and continuously iterates the three optimization problems of demand response master-slave game, feeder block division and combination to obtain the optimal power supply unit division result and its corresponding demand response strategy; the power supply unit is composed of feeder blocks, and the feeder extends from the substation. The unit composed of adjacent feeder blocks of the same substation is called an intra-station unit, and the unit composed of feeder blocks of different substations is called an inter-station unit.

3. A method for dividing power supply units considering demand response master-slave game according to claim 2, characterized in that: In S2, the upper layer of the demand response master-slave game model is the DNO decision layer, which is used to transmit the demand response incentive strategy to the lower layer; the lower layer is the load decision layer, which is used to derive the demand response scheduling plan with the lowest cost based on the incentive strategy of the upper layer, and then feed it back to the upper layer, and continuously iterate this process until an equilibrium solution is obtained.

4. A method for dividing power supply units considering demand response master-slave game according to claim 3, characterized in that: In S2, the mathematical expression of the demand response master-slave game model is as follows: Upper level, DNO decision-making level: Objective function: The upper layer aims to minimize the operating cost of DNO, and its mathematical expression is: my C DNO =C DR +C VTP +C buyup -C sell In this formula, C DNO , C DR , C VTP , C buyup , C sell They represent the total operating cost of DNO, the demand response incentive cost, the peak-valley difference penalty, the cost of purchasing electricity from the upper power grid, and the income from selling electricity to lower-level users; c dr,t 、c VTP 、c buyup 、c sell They represent the unit demand response incentive cost, the unit peak-valley difference penalty, the unit cost of purchasing electricity from the upper power grid, and the unit profit of selling electricity to lower-level users; P t and P a,t They represent the net load at time t before and after demand response, respectively; and They represent the peak and valley values ​​of the net load energy consumption curve after demand response respectively; Constraints: In this formula, and are the minimum and maximum incentive prices respectively; T p The peak period of electricity consumption; dr,t represents the incentive price at time t; Lower level, power user decision-making level: Objective function: The lower layer includes transferable loads and reducible loads, with the goal of minimizing its own operating cost. The mathematical expression is: In this formula, C load is the operating cost of the power user; α and β are both demand response dispatch cost coefficients; ΔP load,t It indicates the change in electricity consumption at time t before and after the user demand response scheduling. When it is greater than 0, it means that the electricity consumption at time t is reduced; c sell The unit price of electricity purchased by power users; P aload,t is the electricity consumption of the power user at time t after demand response; Constraints: P load,t -P aload,t =ΔP load,t 0≤ΔP loadcut,t ≤η dr P loadcut,t In this formula, P load,t , P aload,t They represent the power consumption before and after the power user demand response dispatch at time t; η dr The proportion of demand response contracts signed between DNO and electricity users.

5. A method for dividing power supply units considering demand response master-slave game according to claim 4, characterized in that: S2 also includes: using the fast particle swarm algorithm to nest the linear programming master-slave game model solution process, taking the upper DNO demand response incentive strategy as the particle, and the upper objective function as the fitness of the particle population; the particle position update formula is: In this formula, S i,j represents the j-th dimension variable of the ith particle; fbest j represents the j-th dimension variable of the global optimal particle; β represents the convergence coefficient, and its value range is [0, 1]; α is the particle attenuation coefficient, and the α of each iteration is smaller than that of the previous iteration; rand() represents a normally distributed random number; the lower-level power user model is embedded in the population fitness solution process as a mixed integer linear programming model, and can be solved using the solver.

6. A method for dividing power supply units considering demand response master-slave game according to claim 1, characterized in that: In S3, the feeder block division is performed using the rotation centerline distance weighted alternating positioning algorithm, and the steps are as follows: S301. According to the number of power supply units m, the initial positions of the center lines of 2m feeder blocks are randomly selected. The angles between the feeder lines are S302, the group of initial feeder center lines is stepped Rotate clockwise to obtain the position of the center line of n groups of feeders; S303, starting from the center line of the first group of feeders, dividing according to the Euclidean distance from the load to the center line of the feeder block, then updating the division result by weighted distance, updating the center line position according to the division result, and continuously iterating this process until the center line position no longer changes; S304, dividing the n groups of feeder center lines in S303 to obtain n power supply unit division schemes; S305. Select the optimal solution based on the comprehensive indicators, and take the power supply unit division solution with the best comprehensive indicators as the optimal solution.

7. A method for dividing power supply units considering demand response master-slave game according to claim 6, characterized in that: In S303, the weighted distance calculation formula from the load to the center line of the feeder block is as follows: l i ' ,j =ω 1,i,j oh 2,i,j l i,j In this formula, l i ' ,j and l i,j are the weighted distance and Euclidean distance from load j to the center line of feeder block i, respectively; ω 1,i,j and ω 2,i,j are the peak and variance weighting coefficients, respectively; and They respectively represent the peak value and variance of the net load curve within the block when load j is divided into feeder block i; k1 and k2 are both weighting factors.

8. A method for dividing power supply units considering demand response master-slave game according to claim 6, characterized in that: In S305, the comprehensive index calculation formula of the scheme is as follows: F=αLR+βLB+γPVD In this formula, F is the comprehensive evaluation index of the unit division scheme, α, β and γ are weight coefficients respectively; LR is the average feeder load rate index; LB is the maximum load balance index; PVD is the average feeder peak-to-valley difference index; n is the number of feeder blocks; λ i is the feeder load rate of block i; S i is the maximum load in block i, S max is the maximum load in n blocks; P i max and P i min is the peak and valley value of the net load in block i; the comprehensive index F is positively correlated with the optimization degree of the feeder block division scheme.

9. A method for dividing power supply units considering demand response master-slave game according to claim 8, characterized in that: In S3, under the condition of only considering single contact, that is, two feeders are connected at the end by a tie switch, and two feeder blocks form an inter-station unit or an intra-station unit. In the feeder block division stage, the number of feeder blocks in each substation is set to an even number; before the feeder blocks are combined, the number of inter-station and intra-station units is determined; in the feeder block combination process, the feeder blocks are first combined into inter-station units according to the degree of bordering, and then combined into intra-station units according to the adjacent relationship; The objective function of combining feeder blocks into inter-station units is as follows: Where, BOR is the boundary degree index of the inter-station unit combination scheme; len j is the length of the two feeder blocks of the jth inter-station unit under the combined scheme; tlen j is the total length of the feeder block boundary of the jth inter-station unit under this combination scheme. There are several ways to combine feeder blocks into inter-station energy units. After listing all inter-station unit combination schemes, their border degree index BOR is compared. The one with the largest BOR is the optimal combination scheme, that is, the combination scheme that determines the adjacent energy units in the station in pairs.

10. The method for dividing power supply units considering the master-slave game of demand response according to claim 1, characterized in that: In S4, the overall model of power supply unit division is established. The decision variables of the model are the number of energy units, the division results and the corresponding optimal demand response strategy. The overall objective function of the model is the economic cost of the unit division scheme, including feeder construction cost, demand response incentive cost and the power company's power sales revenue loss due to demand response. The calculation formula is as follows: my C=C line +C DR +C loss In the formula, c line 、c dr,t 、c buyup and c sell are the unit construction cost of the feeder, the unit incentive price of demand response, the unit cost of purchasing electricity from the upper power grid, and the unit cost of selling electricity to the load; b is the number of years the equipment has been in operation, r is the discount rate; L ij represents the line length in the jth feeder block of the i-th substation; P t and P a,t They correspond to the net load power of the grid at time t before and after the demand response respectively; μ represents the proportion of the days of demand response in the whole year; The constraints include N-1 safety criterion constraint and feeder quantity constraint. The specific formula is as follows: In the formula, represents the maximum load of the jth feeder of the i-th substation; S line Indicates the capacity of the feeder; and P max They are the net load of the grid after the maximum demand response scheduling and the net load of the grid without demand response respectively; the number of power supply units is m, and the total number of feeder blocks is 2m.

Images