A Day-Ahead and Intra-Day Scheduling Method Based on Dynamic Partitioning

Abstract

The present invention proposes a day-ahead and intra-day scheduling method based on dynamic partitioning, including the following steps: Step S1: Evaluate the flexibility resources on the power supply side, load side, and grid side of the distribution network, and define the flexibility deficiency rate based on dynamic partitioning; Step S2: Construct a day-ahead scheduling model based on dynamic partitioning, formulate the electricity purchase plan for the next day according to the predicted results of the intermittent distributed power output and load demand on the day-ahead, coordinate various controllable devices to achieve the optimal operation of the system, and make the day-ahead scheduling result play a guiding role in the intra-day scheduling; Step S3: Construct an intra-day scheduling model based on dynamic partitioning, and make a rolling adjustment to the day-ahead scheduling plan according to the errors in the intermittent distributed power generation and load demand prediction in the system, with the actual operation situation of the distribution network access device and the ultra-short-term prediction results, so that the intra-day scheduling can adjust the day-ahead scheduling and improve the utilization efficiency of the flexibility resources within the partition; The present invention can coordinate various controllable devices to achieve the optimal operation of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of optimal operation of distribution networks, and in particular to a day-ahead and intra-day scheduling method based on dynamic partitioning. Background Art

[0002] Due to the continuous changes in the wind, light, and load connected to the distribution network, problems such as power supply falling short of demand within a partition and insufficient power control capabilities of the power sources within the partition may occur in the fixed partitioning scheme. The distributed power sources connected to the distribution network are spatially distributed at different positions and their outputs change over time. The optimization based on the fixed partitioning of the distribution network cannot adapt to this dynamic characteristic. Dynamic partitioning refers to the partitioning scheme of the distribution network changing over time to improve the power control capabilities of the power sources within the partition. There are many related studies in this field. This patent realizes day-ahead and intra-day scheduling on the basis of the dynamic partitioning of the distribution network. The dynamic partitioning of the distribution network can ensure that the partition has a certain power reserve and the internal power sources have strong control capabilities, laying a foundation for suppressing the net load fluctuation based on the dynamic partitioning scheme. In order to improve the ability of the partition to quickly respond to fluctuations during actual operation, it is necessary to construct a day-ahead and intra-day scheduling model for the distribution network based on dynamic partitioning. Summary of the Invention

[0003] The present invention proposes a day-ahead and intra-day scheduling method based on dynamic partitioning. The day-ahead scheduling formulates the electricity purchase plan for the next day according to the predicted results of the output of intermittent distributed power sources and the load demand on the day-ahead, and coordinates various controllable devices to achieve the optimal operation of the system.

[0004] The present invention adopts the following technical solutions.

[0005] A day-ahead and intra-day scheduling method based on dynamic partitioning includes the following steps;

[0006] Step S1: Taking diesel engines, energy storage systems, and interruptible loads as flexibility resources, evaluate the flexibility resources on the power source side, load side, and grid side of the distribution network, and define the flexibility deficiency rate based on dynamic partitioning according to the flexibility requirements of the distribution network;

[0007] Step S2: Construct a day-ahead scheduling model based on dynamic partitioning, formulate the electricity purchase plan for the next day according to the predicted results of the output of intermittent distributed power sources and the load demand on the day-ahead, and coordinate various controllable devices to achieve the optimal operation of the system, so that the day-ahead scheduling result plays a guiding role in the intra-day scheduling;

[0008] Step S3: Construct an intra-day scheduling model based on dynamic partitioning, and make a rolling adjustment to the day-ahead scheduling plan according to the errors in the prediction of the power generation of intermittent distributed power sources and the load demand in the system, and the actual operation situation and ultra-short-term prediction results of the distribution network access devices, so that the intra-day scheduling can adjust the day-ahead scheduling and improve the utilization efficiency of the flexibility resources within the partition.

[0009] In step S1, the flexibility resources of the diesel engine are expressed by the formula

[0010]

[0011]

[0012] where: and are the upward and downward flexibilities of the diesel engine in zone i at time t. If there is no diesel engine in zone i, then and both take the value of 0; Ω DE,i is the set of diesel engines included in zone i; are the maximum and minimum outputs of diesel engine g respectively; P DE,g,t is the output of diesel engine g at time t; and are the upward and downward ramping constraints of diesel engine g; Δt is the time interval;

[0013] Based on the start-stop constraint and minimum output constraint of the diesel engine, its output is regulated according to the intermittent distributed power sources and load prediction in the zone;

[0014] In step S1, the flexibility resources of the energy storage system are expressed by the formula

[0015]

[0016]

[0017] where: and are the upward and downward flexibilities of the energy storage system in zone i at time t. If there is no energy storage system in zone i, then and both take the value of 0; Ω ESS,i is the set of energy storage systems included in zone i; are the maximum and minimum state of charge of energy storage system g respectively; S SOC,g,t is the state of charge of energy storage system g at time t; and are the maximum charge and discharge rates of energy storage g; Δt is the time interval. When the state of charge of the energy storage system is maintained at half, its upward and downward flexibilities will reach relatively high levels simultaneously, but this strategy cannot fully utilize the role of the energy storage system;

[0018] In step S1, the flexibility resources of the interruptible load are expressed by the formula;

[0019]

[0020] where: For the downward flexibility of interruptible loads in zone i, if there are no interruptible loads in zone i, then the value is 0; Ω IL,i is the set of interruptible loads included in zone i; P IL,g,t is the load shedding power of interruptible load g at time t; the interruptible load can have downward flexibility by restoring its load shedding power;

[0021] In step S1, the flexibility demand of the distribution network is generated by the intermittent distributed power sources and loads formed by the connected wind turbines and photovoltaic generators. The flexibility demand requires the distribution network to have the ability to suppress fluctuations, which is expressed by the formula

[0022]

[0023]

[0024] In the formula: is the net load fluctuation amount of zone i at time t, Ω WT,i 、Ω PV,i 、Ω load,i are the node sets where wind turbines, photovoltaic generators, and loads are connected in zone i respectively; P WT,g,t 、P PV,g,t 、P load,g,t are the predicted power generation of the wind turbine and photovoltaic generator installed at node g and the predicted demand of the load at time t; are the upward and downward prediction errors of the predicted power generation of the wind turbine and photovoltaic generator and the predicted demand of the load respectively;

[0025] In step S1, if the flexibility resources of the distribution network are greater than the flexibility demand, it means that the flexibility meets the requirements; otherwise, it means that the flexibility resources are insufficient. To further quantify the ability of the day-ahead scheduling scheme to cope with the uncertainty of the source-load, the flexibility deficiency rate is set as the evaluation index of the day-ahead scheduling scheme, which is expressed by the formula

[0026]

[0027]

[0028]

[0029]

[0030] In the formula: F1, F i flex 、F i flex,up 、F i flex,downThey are respectively the flexibility deficiency rate of the distribution network, the flexibility deficiency rate of zone i, the upward flexibility deficiency rate of zone i, and the downward flexibility deficiency rate of zone i; and They are respectively the upward and downward flexibility resources of zone i, provided by each controllable distributed power source and controllable load; is the number of zones of the distribution network at time t;

[0031] The flexibility deficiency rate requirements of each zone of the distribution network are different. The zones directly connected to the main network of the distribution network can adjust the power purchased from the main network in real time to suppress internal fluctuations, so there is no need for too much flexibility resources. If an important load with high power supply reliability requirements is connected within a certain zone, its power supply reliability needs to be improved, that is, different weights σ are set for each zone, and the weight of the zone with important load access is higher;

[0032] The flexibility deficiency rate is used to quantify the degree to which flexibility resources meet the demand. That is, in day-ahead scheduling, considering the flexibility deficiency rate makes the scheduling plan have the ability to suppress fluctuations, laying a foundation for the autonomous operation within the distribution network area.

[0033] The specific content of step S2 is as follows:

[0034] Select the flexibility deficiency rate as the evaluation index for the ability of each zone to suppress fluctuations. A preset amount of flexibility resources need to be reserved within each zone to suppress the net load fluctuations, that is, the first objective function;

[0035] If aiming to reduce the flexibility deficiency rate causes the power of controllable distributed power sources to be transmitted across zones, then in order to suppress the net load fluctuations locally while reducing the mutual influence between zones, select the tie-line power of dynamic zones as the second objective function; in order to avoid excessive overall operating costs of the distribution network, set the operating cost as the third objective function;

[0036] The first objective function is based on the flexibility deficiency rate, that is, reducing the flexibility deficiency rate to improve the ability of each zone to cope with the uncertainties of wind and light power output and load demand prediction, and laying a foundation for suppressing fluctuations and adjusting the output of the day-ahead plan in intraday scheduling by formulating a flexible day-ahead scheduling plan;

[0037] The second objective function is based on the tie-line power, that is, in order to reduce the coupling between zones and reduce the impact of a zone on other zones when suppressing fluctuations, take the minimum absolute value of the tie-line power of the zone as the objective function, expressed by the formula as;

[0038]

[0039] In the formula: is the number of tie-lines at time t, P link,i,t is the power of tie-line i at time t;

[0040] The third objective function is based on the operating cost of the day-ahead scheduling model, which includes the operating cost C of the diesel engine DE , the operating cost C of the energy storage system ESS , the compensation cost C of the interruptible load IL and the electricity purchase cost C grid ; Expressed by the formula as

[0041]

[0042] In the formula: P grid,t is the electricity purchase amount from the main grid at time t; P DE,t is the output of the diesel engine at time t; P ESS,t is the output of the energy storage system at time t; P IL,t is the load shedding power of the interruptible load at time t; χ1, χ2, χ3 are the fuel cost coefficients of the diesel generator, which vary according to the diesel engine model; C IN is the initial investment cost, C RF is the capital recovery factor; P BD is the rated power of the energy storage system, E BD is the rated capacity of the energy storage system; K b is the power cost coefficient, K e is the energy cost coefficient; r is the interest rate, l is the service life of the energy storage system; c IL is the interruptible load compensation coefficient, C grid,t is the electricity purchase cost from the main grid at time t;

[0043] The constraint conditions of the intra-day scheduling model include power balance constraint,, energy storage system operation constraint, interruptible load operation constraint;

[0044] The power balance constraint is expressed by the formula as

[0045] P grid,t +P DE,t +P ESS,t +P IL,t +P WT,t +P PV,t =P load,t Formula XIV;

[0046] In the formula: P WT,t is the output of the wind turbine at time t; P PV,t is the output of the photovoltaic generator at time t; P load,t is the total load at time t;

[0047] The diesel engine unit constraint is expressed by the formula as

[0048]

[0049] -v down≤P DE,t+1 -P DE,t ≤v up Formula XVI;

[0050]

[0051] Where: and are the minimum and maximum outputs of the diesel engine unit respectively; v down and v up are the ramp rates of the diesel engine unit; η t is the 0-1 state variable of the start-stop of the diesel engine unit at time t; T on,t-1 and T off,t-1 are the continuous start-up and shutdown times; T on and T off are the shortest start-up and shutdown times;

[0052] The operating constraints of the energy storage system are expressed by the formula

[0053]

[0054]

[0055] S SOC,0 = S SOC,T Formula XX;

[0056] Where: S SOC,t is the state of charge of the energy storage system at time t; and are the lower and upper bounds of the state of charge of the energy storage system; and are the maximum charge and discharge rates of the energy storage system; the t values equal to 0 and T represent the initial and end states of the energy storage respectively;

[0057] The operating constraints of the interruptible load are expressed by the formula

[0058]

[0059] Where: represents the maximum cuttable power of the interruptible load;

[0060] The day-ahead optimization model of the distribution network is a multi-objective nonlinear programming problem. A multi-objective optimization algorithm based on fast non-dominated sorting is used to solve the model to solve its discontinuous and non-differentiable problems, and to solve the problem that most single-objective algorithms solve multi-objective problems by using weights to transform multi-objective problems into single-objective problems.

[0061] In step S3, during the intraday scheduling phase, the partition has autonomous capabilities, that is, when there are deviations between the predicted and actual wind and solar power outputs and load demands, the flexibility resources within the partition can be used to quickly suppress fluctuations.

[0062] In step S3, a single-objective intraday scheduling model is constructed with the lowest operating cost to improve the utilization rate and solution speed of flexibility resources. Its objective function is expressed as

[0063]

[0064] The operating costs of intraday scheduling include the operating costs of diesel engines, energy storage systems, the compensation costs of interruptible loads, and the costs of purchasing electricity from the main grid. The calculation methods of the energy storage system, interruptible load, and the costs of purchasing electricity from the main grid are the same as those of the day-ahead scheduling. The diesel engine operating cost calculation function is transformed into a piecewise function to improve the calculation speed and lay a foundation for subsequent linearization methods. It is expressed by the formula as

[0065]

[0066] Based on the Distflow branch power flow model, the power flow constraints of the distribution network of the present invention are constructed. It is expressed by the formula as

[0067]

[0068]

[0069]

[0070]

[0071] Q j,t = Q L,j,t - Q DE,j,t - Q PV,j,t - Q WT,j,t - Q grid,j,t Formula twenty-eight;

[0072] In the formula: u(j) is the set of head nodes with j as the tail node in the middle; v(j) is the set of tail nodes with j as the head node in the middle; r i,j , x i,j are the resistance and reactance of branch i-j; I ij,t is the current of branch i-j at time t; U i,t is the node voltage at time t; P L,j,t , P DE,j,t , P IL,j,t , P PV,j,t and P WT,j,t are the active power demands of the loads connected to node j, the active power output of the diesel engine, the load shedding power of the interruptible load, the active power output of the photovoltaic generator, and the active power output of the wind turbine at time t; PESS,j,t is the discharge power of the energy storage system; P grid,j,t and Q grid,j,t are the active power and reactive power exchanged between the distribution network and the main grid; Q L,j,t 、Q DE,j,t 、Q PV,j,t 、Q WT,j,t are respectively the reactive power demand of the load connected to node j, the reactive power output of the diesel engine, the photovoltaic generator, and the wind turbine at time t;

[0073] The security constraints are expressed by the formula as

[0074]

[0075]

[0076] In the formula: and are respectively the upper and lower voltage limits of node i; and are respectively the upper and lower current limits of branch i-j;

[0077] The difference in form between the diesel engine unit constraint, the energy storage system operation constraint, and the interruptible load operation constraint of the intraday scheduling model and the day-ahead scheduling is the different resolution of the two;

[0078] The day-ahead scheduling scheme restricts that the deviations of the diesel engine output, the state of charge of the energy storage system, and the load shedding power of the interruptible load in the intraday scheduling from the day-ahead scheme do not exceed the threshold, so as to ensure that the day-ahead scheduling scheme can effectively guide the intraday scheduling. The states of the controllable distributed power sources and the response amounts of the controllable loads in the intraday scheduling cannot deviate too much from the day-ahead scheduling; The day-ahead scheduling scheme restricts the deviation of the tie-line power in the intraday scheduling from the day-ahead scheme to avoid that the intraday scheduling model reduces the output of the controllable distributed power sources to reduce the operation cost, resulting in an increase in the power coupling degree of each partition and a decrease in the autonomy ability. Specifically as follows;

[0079] The diesel engine output constraint is expressed by the formula as

[0080]

[0081] In the formula: is the output of diesel engine k at time t in the intraday scheduling stage; is the output of diesel engine k at time t in the day-ahead scheduling stage; ΔP DE,k,t is the threshold of the day-ahead to intraday output deviation of diesel engine k at time t;

[0082] The state of charge constraint of the energy storage system is expressed by the formula as:

[0083]

[0084] In the formula: represents the state of charge of energy storage system k at time t during the intraday scheduling stage; represents the state of charge of energy storage system k at time t during the day-ahead scheduling stage; ΔS SOC,k,t is the threshold of the state-of-charge deviation of energy storage system k from day-ahead to intraday at time t. Based on the result of the state of charge of the energy storage system in intraday rolling scheduling, the day-ahead scheduling state of charge is updated according to the day-ahead output of the energy storage system to guide the intraday scheduling model and ensure the effectiveness of the constraints;

[0085] The interruptible load constraint is expressed by the formula

[0086]

[0087] In the formula: represents the load shedding power of interruptible load k at time t during the intraday scheduling stage; represents the load shedding power of interruptible load k at time t during the day-ahead scheduling stage; ΔP IL,k,t is the threshold of the load shedding power deviation of interruptible load k from day-ahead to intraday at time t;

[0088] The tie-line power constraint is expressed by the formula

[0089]

[0090] In the formula: represents the active power transmitted by tie-line k at time t during the intraday scheduling stage; represents the active power transmitted by tie-line k at time t during the day-ahead scheduling stage; ΔP line,k,t is the threshold of the power deviation of tie-line k from day-ahead to intraday at time t;

[0091] To quickly obtain the global optimal solution, the second-order cone relaxation technique is used to transform the non-linear power flow constraint into a linear constraint. Specifically:

[0092] Define the square of the node voltage amplitude and the square of the branch current amplitude The relationship between the two is expressed by the formula as follows

[0093]

[0094]

[0095] Formula (36) is transformed into

[0096] Further transformed into

[0097] Based on the Distflow branch power flow model, Formulas 24 to 26 related to the power flow constraints of the distribution network of the present invention are constructed and transformed into

[0098]

[0099] The corresponding voltage constraints and current constraints are transformed into:

[0100]

[0101]

[0102] When solving the day-ahead - intra-day model, the resolution of the day-ahead scheduling model is 1h, and the resolution of the intra-day scheduling model is 15min. The reference output of the day-ahead scheduling model is adjusted through linear programming so that the diesel output of the day-ahead plan meets the ramping constraints at a 15min resolution and reduces the deviation from the output before adjustment;

[0103] The objective function used is

[0104]

[0105] The constraint conditions are:

[0106]

[0107]

[0108]

[0109] In the formula: K DE is the number of diesel engines; Formula (43) is the sign function. When is equal to 0, the value is 0. When is greater than 0, the value is 1; and are the maximum and minimum outputs of diesel engine k respectively; The DEpro obtained with the minimum of F is the day-ahead diesel reference output after adaptation processing, which meets the diesel ramping constraints at a 15min time resolution.

[0110] When solving the day-ahead - intra-day model, the linearization of the calculation of the diesel cost is used to linearize the intra-day scheduling model. The transformation method of the diesel cost function is as follows

[0111]

[0112] C DE,k,t = W 2,DE,k,t ·6 + W3,DE,k,t · 34.1313 + W 4,DE,k,t · Equation 47;

[0113] W 1,DE,k,t + W 2,DE,k,t + W 3,DE,k,t + W 4,DE,k,t = 1 Equation 48;

[0114] W 1,DE,k,t ≤ Z 1,DE,k,t Equation 49;

[0115] W 2,DE,k,t ≤ Z 1,DE,k,t + Z 2,DE,k,t Equation 50;

[0116] W 3,DE,k,t ≤ Z 2,DE,k,t + Z 3,DE,k,t Equation 51;

[0117] W 4,DE,k,t ≤ Z 3,DE,k,t Equation 52;

[0118] Z 1,DE,k,t + Z 2,DE,k,t + Z 3,DE,k,t = 1 Equation 53;

[0119] 0 ≤ W 1,DE,k,t Equation 54;

[0120] 0 ≤ W 2,DE,k,t Equation 55;

[0121] 0 ≤ W 3,DE,k,t Equation 56;

[0122] 0 ≤ W 4,DE,k,t Equation 57;

[0123] Where: W 1,DE,k,t , W 2,DE,k,t , W 3,DE,k,t , W 4,DE,k,t are 4 consecutive variables of the k-th diesel engine at time t; Z 1,DE,k,t , Z 2,DE,k,t , Z 3,DE,k,t are 3 0-1 variables of the k-th diesel engine at time t; Using 4 consecutive variables and 3 0-1 variables can achieve the linear conversion of the piecewise function; W 1,DE,k,t , W 2,DE,k,t , W 3,DE,k,t , W 4,DE,k,t achieve the linearization of the function, Z 1,DE,k,t , Z 2,DE,k,t , Z 3,DE,k,tAuxiliary to limit the value range of continuous variables, thereby restricting the interval where the continuous variables are located.

[0124] The tie-line power refers to the active power transmitted by the branch in the sub-region in the power flow distribution result; the day-ahead scheduling model takes the minimum absolute value of the tie-line power of the sub-region as one of the objective functions to schedule the output of controllable power sources, so that the day-ahead scheduling model can effectively guide the intra-day scheduling model, and at the same time, the deviation of the tie-line power of the intra-day scheduling model from the day-ahead scheduling plan does not exceed the threshold.

[0125] Compared with the prior art, the present invention has the following beneficial effects:

[0126] The present invention uses dynamic partitioning to optimize the scheduling of the distribution network. There is sufficient power reserve inside the dynamic partition. Considering the errors in the intermittent distributed power generation and load demand prediction in the system, the intra-day scheduling makes a rolling adjustment to the day-ahead scheduling plan according to the actual operation of the distribution network access devices and the ultra-short-term prediction results. Therefore, the day-ahead scheduling result plays a guiding role in the intra-day scheduling, and the intra-day scheduling can adjust the day-ahead scheduling and improve the utilization efficiency of flexible resources inside the partition. It has greater reference significance for actual engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0127] The present invention will be further described in detail below with reference to the drawings and specific embodiments:

[0128] Attached Figure 1 is a flow schematic diagram of the present invention;

[0129] Attached Figure 2 is a framework schematic diagram of the day-ahead and intra-day scheduling models. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0130] As shown in the figure, a day-ahead and intra-day scheduling method based on dynamic partitioning includes the following steps;

[0131] Step S1: Taking diesel engines, energy storage systems, and interruptible loads as flexible resources, evaluating the flexible resources on the power supply side, load side, and grid side of the distribution network, and defining the flexibility deficiency rate based on dynamic partitioning according to the flexibility requirements of the distribution network;

[0132] Step S2: Constructing a day-ahead scheduling model based on dynamic partitioning, formulating the electricity purchase plan for the next day according to the output of intermittent distributed power sources and the load demand prediction results for the day-ahead, coordinating various controllable devices to achieve the optimal operation of the system, so that the day-ahead scheduling result plays a guiding role in the intra-day scheduling;

[0133] Step S3: Build an intraday scheduling model based on dynamic partitioning. According to the errors in the intermittent distributed power generation and load demand forecasting in the system, make rolling adjustments to the day-ahead scheduling plan based on the actual operation of the distribution network access devices and the ultra-short-term forecasting results, so that the intraday scheduling can adjust the day-ahead scheduling and improve the utilization efficiency of flexible resources within the partition.

[0134] In step S1, the flexible resources of the diesel engine are expressed by the formula

[0135]

[0136]

[0137] In the formula: and are the upward and downward flexibility of the diesel engine in partition i at time t. If there is no diesel engine in partition i, then and both take values of 0; Ω DE,i is the set of diesel engines included in partition i; are the maximum and minimum outputs of diesel engine g respectively; P DE,g,t is the output of diesel engine g at time t; and are the upward and downward ramp constraints of diesel engine g; Δt is the time interval;

[0138] Based on the start-stop constraint and minimum output constraint of the diesel engine, regulate its output according to the intermittent distributed power source and load forecasting situation within the partition;

[0139] In step S1, the flexible resources of the energy storage system are expressed by the formula

[0140]

[0141]

[0142] In the formula: and are the upward and downward flexibility of the energy storage system in partition i at time t. If there is no energy storage system in partition i, then and both take values of 0; Ω ESS,i is the set of energy storage systems included in partition i; are the maximum and minimum state of charge of energy storage system g respectively; S SOC,g,t is the state of charge of energy storage system g at time t; and is the maximum charge and discharge rate of the energy storage g; Δt is the time interval. When the state of charge of the energy storage system is maintained at half, its upward and downward flexibility will reach relatively high levels simultaneously, but this strategy cannot fully utilize the role of the energy storage system;

[0143] In step S1, the flexibility resource of the interruptible load is expressed by the formula;

[0144]

[0145] In the formula: is the downward flexibility of the interruptible load in zone i. If there is no interruptible load in zone i, then takes the value of 0; Ω IL,i is the set of interruptible loads included in zone i; P IL,g,t is the load shedding power of the interruptible load g at time t; the interruptible load can restore its load shedding power to make the interruptible load have downward flexibility;

[0146] In step S1, the flexibility demand of the distribution network is generated by the intermittent distributed power sources and loads formed by the connected wind turbines and photovoltaic generators. The flexibility demand requires the distribution network to have the ability to suppress fluctuations, and is expressed by the formula

[0147]

[0148]

[0149] In the formula: is the net load fluctuation amount in zone i at time t, Ω WT,i 、Ω PV,i 、Ω load,i are the node sets where wind turbines, photovoltaic generators, and loads are connected in zone i respectively; P WT,g,t 、P PV,g,t 、P load,g,t are the predicted power generation of the wind turbine and photovoltaic generator installed at node g and the predicted demand of the load at time t; are the upward and downward prediction errors of the predicted power generation of the wind turbine and photovoltaic generator and the predicted demand of the load respectively;

[0150] In step S1, if the flexibility resource of the distribution network is greater than the flexibility demand, it means that the flexibility meets the requirements; otherwise, it means that the flexibility resource is insufficient. In order to further quantify the ability of the day-ahead scheduling plan to cope with the uncertainty of the source-load, the flexibility deficiency rate is set as the evaluation index of the day-ahead scheduling plan, and is expressed by the formula

[0151]

[0152]

[0153]

[0154]

[0155] Where: F1, F i flex 、F i flex,up 、F i flex,down are respectively the flexibility deficiency rate of the distribution network, the flexibility deficiency rate of zone i, the upward flexibility deficiency rate of zone i, and the downward flexibility deficiency rate of zone i; and are respectively the upward and downward flexibility resources of zone i, provided by each controllable distributed power source and controllable load; is the number of zones of the distribution network at time t;

[0156] The distribution network's various zones have different demands for the flexibility deficiency rate. The zones directly connected to the main network of the distribution network can adjust the electricity quantity purchased from the main network in real time to suppress internal fluctuations, so there is no need for excessive flexibility resources. If an important load with high power supply reliability requirements is connected within a certain zone, its power supply reliability needs to be improved, that is, different weights σ are set for each zone, and the zone with important load access has a higher weight;

[0157] The flexibility deficiency rate is used to quantify the degree to which flexibility resources meet the demand. That is, in day-ahead scheduling, by considering the flexibility deficiency rate, the scheduling plan has the ability to suppress fluctuations, laying a foundation for the autonomous operation within the distribution network area.

[0158] The specific content of step S2 is as follows:

[0159] Select the flexibility deficiency rate as the evaluation index for the ability of the zone to suppress fluctuations. Each zone needs to reserve a preset amount of flexibility resources to suppress the net load fluctuation, that is, the first objective function;

[0160] If aiming to reduce the flexibility deficiency rate causes the power of the controllable distributed power source to be transmitted across zones, then in order to suppress the net load fluctuation locally while reducing the mutual influence between zones, select the tie-line power of the dynamic zone as the second objective function; in order to avoid excessive overall operating costs of the distribution network, set the operating cost as the third objective function;

[0161] The first objective function is based on the flexibility deficiency rate, that is, by reducing the flexibility deficiency rate to improve the ability of the zone to cope with the uncertainty of wind and light output and load demand prediction, and by formulating a flexible day-ahead scheduling plan to lay a foundation for suppressing fluctuations and adjusting the output of the day-ahead plan during intraday scheduling;

[0162] The second objective function is based on the tie-line power. That is, in order to reduce the coupling between regions and minimize the impact of one region on others when suppressing fluctuations, the objective function is to minimize the absolute value of the tie-line power between regions, which can be expressed by the formula as follows;

[0163]

[0164] In the formula: is the number of tie-lines at time t, and P link,i,t is the power of tie-line i at time t;

[0165] The third objective function is based on the operating cost of the day-ahead scheduling model, which includes the operating cost C DE of diesel generators, the operating cost C ESS of energy storage systems, the compensation cost C IL for interruptible loads, and the power purchase cost C grid ; which can be expressed by the formula as

[0166]

[0167] In the formula: P grid,t is the power purchased from the main grid at time t; P DE,t is the output of the diesel generator at time t; P ESS,t is the output of the energy storage system at time t; P IL,t is the load shedding power of the interruptible load at time t; χ1, χ2, and χ3 are the fuel cost coefficients of the diesel generators, which vary according to the diesel generator models; C IN is the initial investment cost, C RF is the capital recovery factor; P BD is the rated power of the energy storage system, E BD is the rated capacity of the energy storage system; K b is the power cost coefficient, K e is the energy cost coefficient; r is the interest rate, l is the service life of the energy storage system; c IL is the compensation coefficient for interruptible loads, C grid,t is the power purchase cost from the main grid at time t;

[0168] The constraint conditions of the intra-day scheduling model include power balance constraints, energy storage system operation constraints, and interruptible load operation constraints;

[0169] The power balance constraint can be expressed by the formula as

[0170] P grid,t +P DE,t +P ESS,t +P IL,t +P WT,t +P PV,t =P load,t Formula XIV;

[0171] In the formula: P WT,t is the output of the wind turbine in period t; P PV,t is the output of the photovoltaic generator in period t; P load,t is the total load in period t;

[0172] The constraints of the diesel engine unit are expressed by the formula as

[0173]

[0174] -v down ≤P DE,t+1 -P DE,t ≤v up Formula XVI;

[0175]

[0176] In the formula: and are the minimum and maximum outputs of the diesel engine unit respectively; v down , v up is the ramp rate of the diesel engine unit; η t is the 0-1 state variable of the start / stop of the diesel engine unit in period t; T on,t-1 , T off,t-1 are the continuous on and off times; T on , T off are the shortest on and off times;

[0177] The operating constraints of the energy storage system are expressed by the formula as

[0178]

[0179]

[0180] S SOC,0 =S SOC,T Formula XX;

[0181] In the formula: S SOC,t is the state of charge of the energy storage system in period t; and are the lower and upper bounds of the state of charge of the energy storage system; and are the maximum charge and discharge rates of the energy storage system; The t values equal to 0 and T represent the initial and end states of the energy storage respectively;

[0182] The operating constraints of the interruptible load are expressed by the formula as

[0183]

[0184] In the formula: Indicates the maximum curtailment power of interruptible load;

[0185] The day-ahead optimization model of the distribution network is a multi-objective non-linear programming problem. A multi-objective optimization algorithm based on fast non-dominated sorting is used to solve the model to address its discontinuous and non-differentiable problems, as well as the problem that most single-objective algorithms solve multi-objective problems by converting them into single-objective problems using weights.

[0186] In step S3, during the intraday scheduling stage, the partition has autonomous capabilities, that is, when there are deviations between the predicted and actual wind-solar power generation and load demand, the flexibility resources within the partition can be used to quickly suppress fluctuations.

[0187] In step S3, a single-objective intraday scheduling model is constructed with the lowest operating cost to improve the utilization rate of flexibility resources and the solution speed. Its objective function is expressed as

[0188]

[0189] The operating costs of intraday scheduling include the operating costs of diesel engines, energy storage systems, interruptible load compensation costs, and main grid power purchase costs; the calculation methods of energy storage systems, interruptible loads, and main grid power purchase costs are the same as those of day-ahead scheduling. The diesel engine operating cost calculation function is converted into a piecewise function to improve the calculation speed and lay a foundation for subsequent linearization methods. It is expressed by the formula as

[0190]

[0191] The distribution network power flow constraint of the present invention is constructed based on the Distflow branch power flow model. It is expressed by the formula as

[0192]

[0193]

[0194]

[0195]

[0196] Q j,t =Q L,j,t -Q DE,j,t -Q PV,j,t -Q WT,j,t -Q grid,j,t Formula twenty-eight;

[0197] In the formula: u(j) is the set of head nodes with j as the tail node in ; v(j) is the set of tail nodes with j as the head node in ; r i,j , x i,j are the resistance and reactance of branch i-j; I ij,tis the current of branch i-j at time period t; U i,t is the node voltage at time period t; P L,j,t 、P DE,j,t 、P IL,j,t 、P PV,j,t and P WT,j,t are respectively the active power demand of the load connected to node j, the active power output of the diesel engine, the load shedding power of the interruptible load, the active power output of the photovoltaic generator and the active power output of the wind turbine at time period t; P ESS,j,t is the discharge power of the energy storage system; P grid,j,t and Q grid,j,t are the active power and reactive power exchanged between the distribution network and the main grid; Q L,j,t 、Q DE,j,t 、Q PV,j,t 、Q WT,j,t are respectively the reactive power demand of the load connected to node j, the reactive power outputs of the diesel engine, the photovoltaic generator and the wind turbine at time period t;

[0198] The security constraints are expressed by the formula as

[0199]

[0200]

[0201] In the formula: and are respectively the upper and lower voltage limits of node i; and are respectively the upper and lower current limits of branch i-j;

[0202] The difference in form between the diesel engine unit constraint, the energy storage system operation constraint, the interruptible load operation constraint of the intraday scheduling model and the day-ahead scheduling is that their resolutions are different;

[0203] The day-ahead scheduling scheme restricts that the deviations of the diesel engine output, the state of charge of the energy storage system and the load shedding power of the interruptible load in the intraday scheduling from the day-ahead scheme do not exceed the threshold, so as to ensure that the day-ahead scheduling scheme can effectively guide the intraday scheduling, and the states of each controllable distributed power source and the response amounts of the controllable loads in the intraday scheduling cannot deviate too much from the day-ahead scheduling; the day-ahead scheduling scheme restricts the deviation of the tie-line power in the intraday scheduling from the day-ahead scheme, so as to avoid that the intraday scheduling model reduces the operation cost by reducing the output of the controllable distributed power source, resulting in an increase in the power coupling degree of each partition and a decrease in the autonomy ability. Specifically as follows;

[0204] The diesel engine output constraint is expressed by the formula as

[0205]

[0206] In the formula: Output of diesel engine k at time t during the intraday scheduling stage; Output of diesel engine k at time t during the day-ahead scheduling stage; ΔP DE,k,t Threshold of the day-ahead to intraday output deviation of diesel engine k at time t;

[0207] The state of charge constraint of the energy storage system is expressed by the formula:

[0208]

[0209] In the formula: State of charge of energy storage system k at time t during the intraday scheduling stage; State of charge of energy storage system k at time t during the day-ahead scheduling stage; ΔS SOC,k,t Threshold of the day-ahead to intraday state of charge deviation of energy storage system k at time t; Based on the result of the state of charge of the energy storage system in the intraday rolling scheduling, update the state of charge of the day-ahead scheduling according to the day-ahead output of the energy storage system to guide the intraday scheduling model to ensure the effectiveness of the constraint;

[0210] The interruptible load constraint is expressed by the formula

[0211]

[0212] In the formula: Load shedding power of interruptible load k at time t during the intraday scheduling stage; Load shedding power of interruptible load k at time t during the day-ahead scheduling stage; ΔP IL,k,t Threshold of the day-ahead to intraday load shedding power deviation of interruptible load k at time t;

[0213] The tie-line power constraint is expressed by the formula

[0214]

[0215] In the formula: Active power transmitted by tie-line k at time t during the intraday scheduling stage; Active power transmitted by tie-line k at time t during the day-ahead scheduling stage; ΔP line,k,t Threshold of the day-ahead to intraday power deviation of tie-line k at time t;

[0216] To quickly obtain the global optimal solution, the second-order cone relaxation technique is used to transform the non-linear power flow constraint into a linear constraint, specifically:

[0217] Define the square of the node voltage magnitude and the square of the branch current magnitude The relationship between the two is expressed by the formula as follows

[0218]

[0219]

[0220] Formula 36 is transformed into

[0221] Further transformed into

[0222] Based on the Distflow branch power flow model, Formulas 24 to 26 related to the power flow constraints of the distribution network of the present invention are constructed and transformed into

[0223]

[0224] The corresponding voltage constraints and current constraints are transformed into:

[0225]

[0226]

[0227] When solving the day-ahead and intra-day models, the resolution of the day-ahead scheduling model is 1 h, and the resolution of the intra-day scheduling model is 15 min. The reference output of the day-ahead scheduling model is adjusted by linear programming so that the diesel output of the day-ahead plan meets the ramping constraint at a 15-min resolution and reduces the deviation from the output before adjustment;

[0228] The objective function used is

[0229]

[0230] The constraint conditions are:

[0231]

[0232]

[0233]

[0234] In the formula: K DE is the number of diesel engines; Formula (43) is the sign function. When is equal to 0, the value is 0. When is greater than 0, the value is 1; and are the maximum and minimum outputs of diesel engine k respectively; DEpro obtained with minimizing F is the day-ahead diesel reference output after adaptation processing, which meets the diesel ramping constraint at a 15-min time resolution.

[0235] When solving the day-ahead and intra-day model, the linearization of the calculation of the diesel engine cost is adopted to linearize the intra-day scheduling model. The transformation method of the diesel engine cost function is as follows

[0236]

[0237] C DE,k,t =W 2,DE,k,t ·6+W 3,DE,k,t ·34.1313+W 4,DE,k,t ·86.1 Formula 47;

[0238] W 1,DE,k,t +W 2,DE,k,t +W 3,DE,k,t +W 4,DE,k,t =1 Formula 48;

[0239] W 1,DE,k,t ≤Z 1,DE,k,t Formula 49;

[0240] W 2,DE,k,t ≤Z 1,DE,k,t +Z 2,DE,k,t Formula 50;

[0241] W 3,DE,k,t ≤Z 2,DE,k,t +Z 3,DE,k,t Formula 51;

[0242] W 4,DE,k,t ≤Z 3,DE,k,t Formula 52;

[0243] Z 1,DE,k,t +Z 2,DE,k,t +Z 3,DE,k,t =1 Formula 53;

[0244] 0≤W 1,DE,k,t Formula 54;

[0245] 0≤W 2,DE,k,t Formula 55;

[0246] 0≤W 3,DE,k,t Formula 56;

[0247] 0≤W 4,DE,k,t Formula 57;

[0248] Where: W 1,DE,k,t 、W 2,DE,k,t 、W 3,DE,k,t 、W 4,DE,k,t are the four consecutive variables of the kth diesel engine at time t; Z 1,DE,k,t 、Z 2,DE,k,t 、Z 3,DE,k,tThree 0-1 variables of the k-th diesel engine at time t; the linear conversion of the piecewise function can be realized by using four continuous variables and three 0-1 variables; W 1,DE,k,t 、W 2,DE,k,t 、W 3,DE,k,t 、W 4,DE,k,t Realize the linearization of the function, Z 1,DE,k,t 、Z 2,DE,k,t 、Z 3,DE,k,t Auxiliary to limit the value of the continuous variable, thereby restricting the interval where the continuous variable is located.

[0249] The tie-line power refers to the active power transmitted by the branch in the sub-region in the power flow distribution result; the day-ahead scheduling model takes the minimum absolute value of the tie-line power of the sub-region as one of the objective functions to schedule the output of the controllable power source, so that the day-ahead scheduling model can effectively guide the intra-day scheduling model, and at the same time make the deviation of the tie-line power of the intra-day scheduling model from the day-ahead scheduling plan not exceed the threshold.

[0250] Example:

[0251] In this embodiment, a day-ahead and intra-day optimal scheduling model for the distribution network is constructed based on dynamic partitioning.

[0252] (1) Establish a day-ahead scheduling model based on dynamic partitioning, reasonably control the output of the controllable distributed power sources and the response amount of the controllable loads within the partition, which can reduce the lack of flexibility rate of the distribution network and the coupling between sub-regions, and improve the autonomous ability of the partition.

[0253] (2) The adaptation method of the day-ahead to intra-day model lays a foundation for the day-ahead plan to guide the intra-day scheduling, and ensures the effectiveness of the constraints of the intra-day scheduling model.

[0254] (3) The day-ahead scheduling plan guides the intra-day scheduling model, and the intra-day scheduling adjusts the day-ahead scheduling output plan, so that the intra-day scheduling model can suppress the fluctuations within the partition, improve the utilization rate of flexibility resources, and reduce the tie-line power between sub-regions.

[0255] Preferably, the purpose of the day-ahead to intra-day scheduling model based on dynamic partitioning is: to improve the autonomous ability of the partition, absorb intermittent distributed power sources such as wind and light nearby and meet the load demand connected to the distribution network. The overall idea is: use the day-ahead scheduling model with global characteristics to reserve a certain amount of flexibility resources for the distribution network and reduce the coupling degree between sub-regions. According to the characteristics of the higher prediction algorithm accuracy in the intra-day scheduling stage, use the flexibility resources to suppress the fluctuations nearby and reduce the operation cost, which has greater reference significance for actual projects. Combining with the embodiment, the technical effects of this example are shown in Tables 1, 2 and 3

[0256] Table 1 Comparison of day-ahead scheduling results based on dynamic partitioning

[0257] Solution Cost / Yuan Interconnection line power / kW Insufficient flexibility rate / % Scheme 1 22766 5498 4.78 Scheme 2 24000 2602 8.71 Scheme 3 23246 4152 11.00

[0258] Table 2 Comparison of Day-ahead Scheduling Results Based on Fixed Partitioning

[0259] Solution Cost / Yuan Interconnection line power / kW Insufficient flexibility rate / % Scheme 1 21816 6477 16.42 Scheme 2 22977 8697 7.34 Scheme 3 22898 10054 5.89

[0260] Table 3 Comparison of Day-ahead and Intra-day Operating Costs

[0261]

[0262] In the day-ahead scheduling scheme based on fixed partitioning, the power sources within some partitions cannot meet the load demand. When the load demand within the partition is high, external power sources are required to provide power, which also causes strong power coupling between partitions. At this time, the scheduling scheme based on fixed partitioning requires higher power interaction between partitions to achieve a lower flexibility deficiency rate. It is difficult for the day-ahead scheduling based on fixed partitioning to achieve regional autonomy for each partition.

[0263] By establishing a day-ahead scheduling model based on dynamic partitioning and reasonably controlling the output of controllable distributed power sources and the response volume of controllable loads within the partition, the flexibility deficiency rate of the distribution network and the coupling between partitions can be reduced, and the partition autonomy ability can be improved. The adaptation method for the day-ahead and intra-day model lays a foundation for the day-ahead scheme to guide the intra-day scheduling and ensures the effectiveness of the constraints of the intra-day scheduling model. The day-ahead scheduling scheme guides the intra-day scheduling model, and the intra-day scheduling adjusts the output scheme of the day-ahead scheduling, enabling the intra-day scheduling model to suppress the fluctuations within the partition, improve the utilization rate of flexibility resources, and reduce the power of the tie lines between partitions. To improve the flexibility of the distribution network in the day-ahead optimization scheduling model, the output of the diesel engine is maintained at a certain value. In the intra-day scheduling stage, the prediction results of the wind and light output and the load demand are relatively accurate, and there is no need to increase the output of the diesel engine too much to maintain a high flexibility, which leads to an increase in the operating cost. Therefore, the overall output of the diesel engine in the intra-day stage will be lower than that in the day-ahead stage, the charge and discharge times of the energy storage are reduced, and the charge and discharge depth and operating cost are significantly reduced. The day-ahead scheduling scheme reserves a certain amount of flexibility resources for the distribution network. When there are net load fluctuations, controllable distributed power sources such as diesel engines and energy storage systems and controllable loads can achieve near-by suppression of fluctuations. However, when the load corresponding to the partition directly connected to the main grid fluctuates, it is still necessary to adjust the power purchase quantity to achieve power suppression. The intra-day power purchase cost increases by 12.7% compared with the day-ahead power purchase cost, and the total intra-day operating cost increases by 2.73% compared with the day-ahead. Due to the reduction of the operating costs of controllable distributed power sources and controllable loads, the overall cost change is not significant.

[0264] The above are only the preferred embodiments of the present invention. All equivalent changes and modifications made according to the scope of the patent application of the present invention shall fall within the scope of the present invention.

Claims

1. A day-ahead and intra-day scheduling method based on dynamic partitioning, characterized in that: Including the following steps; Step S1: Taking a diesel engine, an energy storage system, and an interruptible load as flexibility resources, evaluating the flexibility resources on the power supply side, load side, and grid side of the distribution network, and defining the flexibility deficiency rate based on dynamic partitioning according to the flexibility requirements of the distribution network; Step S2: Constructing a day-ahead scheduling model based on dynamic partitioning, formulating the next-day power purchase plan according to the predicted results of the intermittent distributed power generation and load demand on the day-ahead, coordinating various controllable devices to achieve the optimal operation of the system, and making the day-ahead scheduling result play a guiding role in the intra-day scheduling; Step S3: Constructing an intra-day scheduling model based on dynamic partitioning, making a rolling adjustment to the day-ahead scheduling plan according to the errors in the prediction of the intermittent distributed power generation and load demand in the system, and using the actual operation conditions of the distribution network access devices and the ultra-short-term prediction results, so that the intra-day scheduling can adjust the day-ahead scheduling and improve the utilization efficiency of the flexibility resources within the partition; In step S1, the flexibility resource of the interruptible load is expressed by the formula; In the formula: is the downward flexibility of the interruptible load in zone i. If there is no interruptible load in zone i, then takes a value of 0; Ω IL,i is the set of interruptible loads included in zone i; P IL,g,t is the load shedding power of interruptible load g at time t. The interruptible load can have downward flexibility by restoring its load shedding power; In step S1, the flexibility requirement of the distribution network is generated by the intermittent distributed power sources formed by the connected wind turbines and photovoltaic generators and the load. The flexibility requirement is that the distribution network is required to have the ability to suppress fluctuations, and is expressed by the formula Where: is the net load fluctuation of partition i at time t, Ω WT,i , Ω PV,i , Ω load,i are the node sets of wind turbines, photovoltaic generators, and loads connected in partition i respectively; P WT,g,t , P PV,g,t , P load,g,t are the predicted power generation of the wind turbine and photovoltaic generator installed at node g and the predicted demand of the load at time t; are the upward and downward prediction errors of the predicted power generation of the wind turbine and photovoltaic generator and the predicted demand of the load respectively; In step S1, if the flexibility resources of the distribution network are greater than the flexibility requirements, it means that the flexibility meets the requirements; otherwise, it means that the flexibility resources are insufficient. In order to quantify the ability of the day-ahead scheduling plan to cope with the source-load uncertainty, the flexibility deficiency rate is set as the evaluation index of the day-ahead scheduling plan, and is expressed by the formula Where: F1, F i flex , F i flex,up , F i flex,down are respectively the flexibility deficiency rate of the distribution network, the flexibility deficiency rate of zone i, the upward flexibility deficiency rate of zone i, and the downward flexibility deficiency rate of zone i; and are respectively the upward and downward flexibility resources of zone i, provided by each controllable distributed power source and controllable load; is the number of zones of the distribution network at time t.

2. The day-ahead and intra-day scheduling method based on dynamic partitioning according to claim 1, wherein: In step S1, the flexibility resource of the diesel engine is expressed by the formula Where: and are the upward and downward flexibility of the diesel engine in zone i at time t. If there is no diesel engine in zone i, then and both take the value of 0; Ω DE,i is the set of diesel engines included in zone i; are the maximum and minimum outputs of diesel engine g respectively; P DE,g,t is the output of diesel engine g at time t; and are the upward and downward ramping constraints of diesel engine g; Δt is the time interval; Based on the start-stop constraint and minimum output constraint of the diesel engine, its output is regulated according to the prediction of the intermittent distributed power source and load within the partition; In step S1, the flexibility resource of the energy storage system is expressed by the formula In the formula: and are the upward and downward flexibilities of the energy storage system in zone i at time t. If there is no energy storage system in zone i, then and both take values of 0; Ω ESS,i is the set of energy storage systems included in zone i; are the maximum and minimum state of charge of energy storage system g respectively; S SOC,g,t is the state of charge of energy storage system g at time t; and are the maximum charge and discharge rates of energy storage g; Δt is the time interval; when the state of charge of the energy storage system is maintained at half, its upward and downward flexibilities will reach high levels simultaneously, and the role of the energy storage system cannot be fully exerted; The requirements for the flexibility deficiency rate of each partition of the distribution network are different. The partitions directly connected to the main network of the distribution network suppress internal fluctuations by adjusting the power purchased from the main network in real time, so there is no need for too much flexibility resources; if an important load with high power supply reliability requirements is connected within a certain partition, its power supply reliability needs to be improved, that is, different weights σ are set for each partition, and the partition connected to the important load has a high weight; The flexibility deficiency rate is used to quantify the degree to which the flexibility resources meet the requirements, that is, in the day-ahead scheduling, the scheduling plan has the ability to suppress fluctuations by considering the flexibility deficiency rate.

3. The day-ahead and intra-day scheduling method based on dynamic partitioning according to claim 2, characterized in that: The specific content of step S2 includes the following: Selecting the flexibility deficiency rate as the evaluation index of the partition's ability to suppress fluctuations. A preset flexibility resource needs to be reserved within each partition to suppress the net load fluctuation, that is, the first objective function; If aiming to reduce the flexibility deficiency rate will cause the cross-regional transmission of the power of the controllable distributed power source, then in order to reduce the mutual influence between partitions while achieving the in-situ suppression of the net load fluctuation, the tie-line power of the dynamic partition is selected as the second objective function; in order to avoid the overall operation cost of the distribution network being too high, the operation cost is set as the third objective function; The first objective function is based on the flexibility deficiency rate, that is, by reducing the flexibility deficiency rate to improve the ability of the partition to cope with the uncertainties of wind and light output and load demand forecasting. By formulating a flexible day-ahead scheduling plan, it lays a foundation for suppressing fluctuations and adjusting the output of the day-ahead plan during the intraday scheduling; The second objective function is based on the tie-line power, that is, in order to reduce the coupling between partitions and minimize the impact of the internal part of the partition on other partitions when suppressing fluctuations, the absolute value of the tie-line power of the partition is used as the objective function, which is expressed by the formula; Where: is the number of tie lines at time t, and P link,i,t is the power of tie line i at time t; The third objective function is based on the operating cost of the day-ahead scheduling model, which includes the operating cost of diesel engines \(C\) DE DE , the operating cost of energy storage systems \(C\) ESS ESS , the compensation cost of interruptible loads \(C\) IL IL and the electricity purchase cost \(C\) grid grid ; It is expressed by the formula as Where: P grid,t is the electricity purchase volume from the main grid at time t; P DE,t is the output of the diesel engine at time t; P ESS,t is the output of the energy storage system at time t; P IL,t is the load shedding power of the interruptible load at time t; χ1, χ2, χ3 are the fuel cost coefficients of the diesel generator, which vary according to different diesel engine models; C IN is the initial investment cost, C RF is the capital recovery factor; P BD is the rated power of the energy storage system, E BD is the rated capacity of the energy storage system; K b is the power cost coefficient, K e is the energy cost coefficient; r is the interest rate, l is the service life of the energy storage system; c IL is the interruptible load compensation coefficient, C grid,t is the cost of purchasing electricity from the main grid at time t; The constraint conditions of the intraday scheduling model include power balance constraint, energy storage system operation constraint, and interruptible load operation constraint; The power balance constraint is expressed by the formula P grid,t +P DE,t +P ESS,t +P IL,t +P WT,t +P PV,t =P load,t ; Where: P WT,t is the output of the wind turbine in period t; P PV,t is the output of the photovoltaic generator in period t; P load,t is the total load in period t; The diesel engine unit constraint is expressed by the formula -v down ≤P DE ,t +1 -P DE ,t≤v up ; Wherein: and are respectively the minimum and maximum output of the diesel engine unit; v down , v up are the ramp rates of the diesel engine unit; η t is the 0-1 state variable of the start-stop of the diesel engine unit at time t; T on,t-1 , T off,t-1 are the continuous start-up and shutdown times; T on and T off are the shortest start-up and shut-down times; The energy storage system operation constraint is expressed by the formula S SOC,0 = S SOC , T; Where: S SOC,t is the state of charge of the energy storage system at time t; and are the lower and upper bounds of the state of charge of the energy storage system; and are the maximum charge and discharge rates of the energy storage system; The t values equal to 0 and T represent the initial and end states of the energy storage respectively; The interruptible load operation constraint is expressed by the formula In the formula: represents the maximum cuttable power of the interruptible load; The said distribution network day-ahead optimization model is a multi-objective nonlinear programming problem, and a multi-objective optimization algorithm based on fast non-dominated sorting is used to solve the model to solve its discontinuous and non-differentiable problems, and to solve the problem that most single-objective algorithms use weights to transform multi-objective problems into single-objective problems.

4. A day-ahead and intra-day scheduling method based on dynamic partitioning according to claim 1, characterized in that: In step S3, during the intraday scheduling stage, the partition has autonomous capabilities, that is, when there are deviations between the wind and light output, load demand forecasting and the actual situation, the flexibility resources within the partition are used to quickly suppress fluctuations; In step S3, a single-objective intraday scheduling model is constructed with the lowest operating cost to improve the utilization rate and solution speed of flexibility resources, and its objective function is expressed as The operating costs of intraday scheduling include the operating costs of diesel engines, energy storage systems, interruptible load compensation costs, and main network power purchase costs; the calculation methods of energy storage systems, interruptible loads, and main network power purchase costs are the same as those of day-ahead scheduling; The diesel engine operating cost calculation function is transformed into a piecewise function to improve the calculation speed and lay a foundation for subsequent linearization methods, which is expressed by the formula Based on the Distflow branch power flow model, the power flow constraint of the distribution network is constructed, which is expressed by the formula Q j,t = Q L,j,t -Q DE,j,t -Q PV,j,t -Q WT,j,t -Q grid,j,t ; where: u(j) is the set of head nodes with j as the tail node in the middle; v(j) is the set of tail nodes with j as the head node in the middle; r i,j , x i,j are the resistance and reactance of branch i-j; I ij,t is the current of branch i-j at time t; U i,t is the node voltage at time t; P L,j,t , P DE,j,t , P IL,j,t , P PV,j,t and P WT,j,t are respectively the active power demand of the load connected to node j, the active power output of the diesel engine, the load shedding power of the interruptible load, the active power output of the photovoltaic generator, and the active power output of the wind turbine at time t; P ESS,j,t is the discharge power of the energy storage system; P grid,j,t and Q grid,j,t are the active power and reactive power exchanged between the distribution network and the main grid; Q L,j,t , Q DE,j,t , Q PV,j,t , Q WT,j,t are respectively the reactive power demand of the load connected to node j, the reactive power output of the diesel engine, the photovoltaic generator, and the wind turbine at time t; The security constraint is expressed by the formula where: and are the upper and lower voltage limits of node i, respectively; and are the upper and lower current limits of branch i-j, respectively; The differences in the form of the diesel engine unit constraint, energy storage system operation constraint, and interruptible load operation constraint between the intraday scheduling model and the day-ahead scheduling are the different resolutions of the two; The day-ahead scheduling plan restricts that the deviations of the diesel engine output, the state of charge of the energy storage system, and the load shedding power of the interruptible load during the intraday scheduling from the day-ahead plan do not exceed the threshold, so as to ensure that the day-ahead scheduling plan can effectively guide the intraday scheduling, and the states of each controllable distributed power source and the controllable load response amount during the intraday scheduling cannot deviate too much from the day-ahead scheduling; The day-ahead scheduling plan restricts the deviation of the tie-line power during the intraday scheduling from the day-ahead plan to avoid that the intraday scheduling model reduces the output of controllable distributed power sources in order to reduce the operating cost, resulting in an increase in the power coupling degree and a decrease in the autonomous ability of each partition. Specifically as follows; The diesel engine output constraint is expressed by the formula Wherein: is the output of diesel engine k at time t during the intraday scheduling stage; is the output of diesel engine k at time t during the day-ahead scheduling stage; ΔP DE,k,t is the threshold value of the day-ahead to intraday output deviation of diesel engine k at time t; The state of charge constraint of the energy storage system is expressed as: Wherein: is the state of charge of energy storage system k at time t during the intraday scheduling stage; is the state of charge of energy storage system k at time t during the day-ahead scheduling stage; ΔS SOC,k,t is the threshold of the day-ahead to intraday state-of-charge deviation of energy storage system k at time t; Based on the state-of-charge result of the energy storage system for intraday rolling scheduling, the day-ahead scheduling state of charge is updated according to the day-ahead output of the energy storage system to guide the intraday scheduling model to ensure the effectiveness of the constraints; The interruptible load constraint is expressed by the formula Where: is the load shedding power of interruptible load k at time t during the intraday scheduling stage; is the load shedding power of interruptible load k at time t during the day-ahead scheduling stage; ΔP IL,k,t is the deviation threshold of the day-ahead to intraday load shedding power of interruptible load k at time t; The tie-line power constraint is expressed by the formula Wherein: is the active power transmitted by tie line k during the intraday scheduling stage t; is the active power transmitted by tie line k during the day-ahead scheduling stage t; ΔP line,k,t is the threshold value of the power deviation of tie line k from day-ahead to intraday at time t; To quickly obtain the global optimal solution, the second-order cone relaxation technology is used to transform the nonlinear power flow constraint into a linear constraint. Specifically: Define the square of the node voltage amplitude and the square of the branch current amplitude The relationship between the two is expressed by the following formula Deformed into Converted to Based on the Distflow branch power flow model, formulas related to the power flow constraints of the distribution network are constructed and deformed into The corresponding voltage constraint and current constraint are transformed into:

5. The day-ahead and intra-day scheduling method based on dynamic partitioning according to claim 4, wherein: When solving the day-ahead and intra-day model, the resolution of the day-ahead scheduling model is 1h, and the resolution of the intra-day scheduling model is 15min. The reference output of the day-ahead scheduling model is adjusted through linear programming to make the diesel output of the day-ahead plan meet the ramping constraint at a 15min resolution and reduce the deviation from the output before adjustment. The objective function used is The constraint conditions are: where: K DE is the number of diesel engines; Equation (43) is the sign function. When is equal to 0, the value is 0. When is greater than 0, the value is 1; and are the maximum and minimum output powers of diesel engine k, respectively; Taking F DEpro as the minimum objective function, the obtained is the reference output power of the diesel engine for the day-ahead after adaptation processing, which satisfies the ramp constraint of the diesel engine at a 15-minute time resolution.

6. The day-ahead and intra-day scheduling method based on dynamic partitioning according to claim 4, wherein: When solving the day-ahead and intra-day model, the linearization of the diesel cost calculation is used to linearize the intra-day scheduling model. The conversion method of the diesel cost function is as follows C DE,k,t = W 2,DE,k,t · 6 + W 3,DE,k,t · 34.1313 + W 4,DE,k,t · 86.1; W 1,DE,k,t +W 2,DE,k,t +W 3,DE,k,t +W 4,DE,k,t = 1; W 1,DE,k,t ≤Z 1,DE,k,t ; W 2,DE,k,t ≤Z 1,DE,k,t +Z 2,DE,k,t ; W 3,DE,k,t ≤Z 2,DE,k,t +Z 3,DE,k,t ; W 4,DE,k,t ≤Z 3,DE,k,t ; Z 1,DE,k,t +Z 2,DE,k,t +Z 3,DE,k,t = 1; 0≤W 1,DE,k,t ； 0≤W 2,DE,k,t ； 0≤W 3,DE,k,t ； 0≤W 4,DE,k,t ； Where: W 1,DE,k,t 、W 2,DE,k,t 、W 3,DE,k,t 、W 4,DE,k,t are the four consecutive variables of the k-th diesel engine at time t; Z 1,DE,k,t 、Z 2,DE,k,t 、Z 3,DE,k,t are the three 0-1 variables of the k-th diesel engine at time t; The linear conversion of the piecewise function can be achieved by using four consecutive variables and three 0-1 variables; W 1,DE,k,t 、W 2,DE,k,t 、W 3,DE,k,t 、W 4,DE,k,t realize the linearization of the function, and Z 1,DE,k,t 、Z 2,DE,k,t 、Z 3,DE,k,t assist in restricting the values of the continuous variables, thereby restricting the intervals where the continuous variables are located.

7. A day-ahead and intra-day scheduling method based on dynamic partitioning according to claim 3, characterized in that: The tie-line power refers to the active power transmitted by the branch between partitions in the power flow distribution result. The day-ahead scheduling model schedules the output of controllable power sources with the minimum absolute value of the tie-line power between partitions as one of the objective functions, enabling the day-ahead scheduling model to effectively guide the intra-day scheduling model and ensuring that the deviation of the tie-line power of the intra-day scheduling model from the day-ahead scheduling plan does not exceed the threshold.

Images