Scheduling method and device for active power distribution network based on multiple time scales

Abstract

The invention provides a scheduling method and device for an active power distribution network based on multiple time scales. The method comprises the following steps: establishing a first satisfaction model for a user use mode according to an electricity price and a response model for representing the response of user power consumption to the electricity price, and establishing a comprehensive satisfaction model according to the first satisfaction model and a second satisfaction model for user cost; on the basis of the comprehensive satisfaction model, day-ahead scheduling data are determined through a day-ahead scheduling model, and the day-ahead scheduling model meets the minimum cost corresponding to the scheduling data; and determining scheduling optimization data of the active power distribution network at the next moment according to the real-time operation data and the day-ahead scheduling data of the active power distribution network at the intra-day stage. According to the scheme of the invention, the scheduling of the active power distribution network at the next moment is optimized based on the day-ahead scheduling data and the real-time operation data at the current moment, the prediction error is reduced, and the consumption of renewable energy sources is increased.

Description

Technical Field

[0001] The present application relates to the technical field of distribution network optimal scheduling, and particularly to a scheduling method and device for an active distribution network based on multiple time scales. Background Art

[0002] With the continuous increase of distributed energy resources such as wind power, photovoltaic power, and energy storage in the distribution network, it has promoted the transformation of the traditional distribution network into an active distribution network. An active distribution network usually includes various resources on the power generation side and the demand side, and effectively optimizes these resources. However, the random output of distributed energy and the different response characteristics of these resources make the scheduling of the active distribution network more complex and difficult. Therefore, how to reduce the random fluctuation of electricity, make full use of the resources on the power generation side and the demand side, and coordinate the resources according to their response characteristics is a challenge faced by the scheduling of the active distribution network. Due to the strong randomness and volatility of renewable energy, it is difficult to accurately predict the output, and the long-term scheduling result may not match the actual operation state. Summary of the Invention

[0003] The purpose of the present application is to provide a scheduling method and device for an active distribution network based on multiple time scales, so as to optimize the scheduling of the next moment of the active distribution network based on the day-ahead scheduling data and the real-time operation data at the current moment, avoiding the drawbacks of the scheduling scheme in the related technology and also solving the limitation of optimizing the scheduling scheme only through the day-ahead scheduling data.

[0004] To achieve the above object, an embodiment of the present application provides a scheduling method for an active distribution network based on multiple time scales, including:

[0005] According to the electricity price and a response model representing the response of user power consumption to the electricity price, establish a first satisfaction model for the user usage mode, and establish a comprehensive satisfaction model according to the first satisfaction model and a second satisfaction model for the user cost;

[0006] Based on the comprehensive satisfaction model, determine the day-ahead scheduling data through a day-ahead scheduling model, and the day-ahead scheduling model satisfies the minimum cost corresponding to the scheduling data;

[0007] According to the real-time operation data of the active distribution network in the intraday stage and the day-ahead scheduling data, determine the scheduling optimization data of the next moment of the active distribution network.

[0008] Optionally, the response model is established by the following method:

[0009] According to the response data of user power consumption to the electricity price, determine the original load demand, original electricity price data, load change amount, and electricity price change amount of the active distribution network;

[0010] Determine a first ratio of the load change amount to the original load demand, and a second ratio of the original electricity price data to the electricity price change amount;

[0011] Determine an elasticity coefficient for price response according to a first product of the first ratio and the second ratio;

[0012] Construct a response model characterizing the response of the user's power consumption to the electricity price according to the elasticity coefficient.

[0013] Optionally, the first satisfaction model is:

[0014]

[0015] where S u represents the first satisfaction, and are respectively the load value and the load change amount of the controllable distributed energy source i at time period t, N represents the number of subsets of the controllable distributed energy sources, and T represents the total number of time periods within the preset time.

[0016] Optionally, the second satisfaction model is established by the following method:

[0017] Determine an electricity price upper limit threshold and an electricity price lower limit threshold according to the allowable change range of the electricity price and the base electricity price;

[0018] Determine the maximum electricity price change rate according to the electricity price upper limit threshold and the electricity price lower limit threshold;

[0019] Construct a second satisfaction model for the user cost according to the electricity price upper limit threshold, the electricity price lower limit threshold and the maximum electricity price change rate.

[0020] Optionally, establishing the comprehensive satisfaction model according to the first satisfaction model and the second satisfaction model for the user cost includes:

[0021] Obtain a first weight corresponding to the first satisfaction and a second weight corresponding to the second satisfaction for the user cost;

[0022] Establish a comprehensive satisfaction model in a weighted manner according to the first satisfaction model, the first weight, the second satisfaction model for the user cost and the second weight.

[0023] Optionally, the day-ahead scheduling model satisfies the minimum sum of the costs of purchasing electricity in the controllable distribution network, the costs of controllable distributed generation, the energy storage cost, the cost of interruptible load and the cost of switched shunt capacitors.

[0024] Optionally, the constraint conditions of the day-ahead scheduling model include at least one of the following:

[0025] Power flow equation constraint conditions;

[0026] Safe operation constraint conditions;

[0027] On-load tap-changing transformer constraint conditions;

[0028] Distributed generator constraint conditions;

[0029] Capacitor bank constraint conditions;

[0030] Energy storage element constraint conditions;

[0031] Interruptible load constraint conditions.

[0032] Optionally, determining the scheduling optimization data of the active distribution network at the next moment according to the real-time operation data of the active distribution network in the intraday stage and the day-ahead scheduling data includes:

[0033] Obtain the real-time operation data of the active distribution network in the intraday stage, determine the first difference between the output values of distributed energy in each time period in the real-time operation data and the solution amounts in the corresponding time periods in the day-ahead scheduling data, and the overall fluctuation average value of distributed energy in the day-ahead scheduling data;

[0034] Determine the weighted coefficient of distributed energy as the ratio of the first difference to the average value;

[0035] Construct a second objective function for smoothing power fluctuations according to the real-time operation data, the day-ahead scheduling data, and the weighted coefficient;

[0036] Determine the scheduling optimization data of the active distribution network at the next moment according to the second objective function.

[0037] Optionally, the second objective function is:

[0038]

[0039] where N is the number of subsets of controllable distributed energy, T is the total number of time periods within a preset time, is the decision variable of the i-th controllable distributed energy in the real-time operation data, is the optimization result of the i-th controllable distributed energy in the day-ahead scheduling data; ζ i is the weighted coefficient of the i-th controllable distributed energy.

[0040] To achieve the above object, an embodiment of the present application further provides a scheduling device for an active distribution network based on multiple time scales, including:

[0041] A model establishment module, configured to establish a first satisfaction model for the user's usage pattern according to the electricity price and a response model characterizing the response of the user's power consumption to the electricity price, and establish a comprehensive satisfaction model according to the first satisfaction model and a second satisfaction model for the user's cost;

[0042] A day-ahead scheduling data determination module, configured to determine day-ahead scheduling data based on the comprehensive satisfaction model through a day-ahead scheduling model, where the day-ahead scheduling model satisfies the minimum cost corresponding to the scheduling data;

[0043] A scheduling optimization module, configured to determine the scheduling optimization data for the next moment of the active distribution network according to the real-time operation data of the active distribution network in the intraday stage and the day-ahead scheduling data.

[0044] To achieve the above object, an embodiment of the present application further provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps in the scheduling method for an active distribution network based on multiple time scales as described in any one of the above are implemented.

[0045] The beneficial effects of the above technical solutions of the present application are as follows:

[0046] In the embodiment of the present application, a first satisfaction model for the user's usage pattern is established according to the electricity price and a response model characterizing the response of the user's power consumption to the electricity price, and a comprehensive satisfaction model is established according to the first satisfaction model and a second satisfaction model for the user's cost; based on the output result of the comprehensive satisfaction model, day-ahead scheduling data is determined through a day-ahead scheduling model, and the day-ahead scheduling model satisfies the minimum cost corresponding to the scheduling data; according to the real-time operation data of the active distribution network in the intraday stage and the day-ahead scheduling data, the scheduling optimization data for the next moment of the active distribution network is determined. In the above technical solution, based on the day-ahead scheduling data and the real-time operation data at the current moment, the scheduling for the next moment of the active distribution network is optimized, that is, the scheduling reference given in the day-ahead stage is tracked, the power fluctuation is smoothed, the prediction error is reduced, the system operation cost is refinedly reduced, and the consumption of renewable energy is increased. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic flowchart of a scheduling method for an active distribution network based on multiple time scales provided by an embodiment of the present application;

[0048] Figure 2 It is a schematic topology diagram of a test system provided by an embodiment of the present application;

[0049] Figure 3 It is an intraday load level diagram of a test system provided by an embodiment of the present application;

[0050] Figure 4The in-day active power output level diagram of the test system provided by the embodiments of the present application;

[0051] Figure 5 The in-day reactive power output level diagram of the test system provided by the embodiments of the present application;

[0052] Figure 6 The system net load of the optimization results of different methods provided by the embodiments of the present application;

[0053] Figure 7 The system line loss of the optimization results of different methods provided by the embodiments of the present application;

[0054] Figure 8 The module schematic diagram of the scheduling device for the active distribution network based on multiple time scales provided by the embodiments of the present application. Detailed implementation manners

[0055] To make the technical problems, technical solutions and advantages to be solved by the present application clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0056] It should be understood that the "one embodiment" or "an embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, the appearances of "in one embodiment" or "in an embodiment" throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner.

[0057] In various embodiments of the present application, it should be understood that the magnitudes of the serial numbers of the following processes do not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0058] The various methods based on model predictive control in the embodiments of the present application are also used to study the optimization problems of highly renewable energy integrated power systems, such as wind power scheduling, voltage regulation and energy management.

[0059] Referring to Figure 1 As shown, the embodiments of the present application provide a scheduling method for an active distribution network based on multiple time scales, including:

[0060] Step 11, according to the electricity price and the response model representing the response of user power consumption to the electricity price, establish a first satisfaction model for user usage patterns, and establish a comprehensive satisfaction model according to the first satisfaction model and the second satisfaction model for user costs.

[0061] Step 12, based on the comprehensive satisfaction model, determine the day-ahead scheduling data through the day-ahead scheduling model.

[0062] Step 13: Determine the scheduling optimization data for the next moment of the active distribution network according to the real-time operation data and the day-ahead scheduling data of the active distribution network during the intraday stage.

[0063] According to the response data of the user's power consumption to the electricity price, a linearized response curve of the user to the price can be determined. The response curve includes the original load demand P0 of the active distribution network, the original electricity price data ρ0, the load change ΔP, and the electricity price change Δρ; determine the first ratio of the load change to the original load demand, and the second ratio of the original electricity price data to the electricity price change; determine the elasticity coefficient of the user's power consumption response to the price according to the first product of the first ratio and the second ratio.

[0064] The elasticity coefficient ε is expressed by the following formula (1):

[0065]

[0066] where ε is the elasticity coefficient of the user's power consumption response to the price, P0 and ρ0 are the original load demand and the original electricity price; ΔP and Δρ are the load and electricity price changes respectively.

[0067] In some cases, the price change at a certain time will cause changes in the electricity consumption behavior at other times. Therefore, according to the elasticity coefficient, a response model representing the user's power consumption response to the electricity price can be constructed, which can be defined as an elasticity matrix and expressed by the following formula (2):

[0068]

[0069] Taking ε 23 as an example, first set all to zero (other loads except the third load remain unchanged), and then only change the third load to obtain At this time

[0070] In order to reflect the degree of the user's change in electricity consumption behavior, a first satisfaction model for the user's usage pattern is established. The first satisfaction model is expressed by the following formula (3):

[0071]

[0072] where S u represents the first satisfaction, that is, the user's usage pattern satisfaction, and are the load value and the load change of the controllable distributed energy source i at time t respectively, N represents the number of subsets of the controllable distributed energy sources, and T represents the total number of time periods within the preset time.

[0073] The upper bound of the user's satisfaction with the usage method is 1, and the lower bound is 0, representing the most satisfied and the most dissatisfied situations respectively. In the above formula (3), the less the user load changes, the higher the satisfaction with the usage method.

[0074] Optionally, a second satisfaction model for the user cost can be established according to the proportion of the electricity price change in the allowable range of the electricity price, including:

[0075] Determine the upper threshold and the lower threshold of the electricity price according to the allowable change range of the electricity price and the basic electricity price;

[0076] Determine the maximum electricity price change rate according to the upper threshold and the lower threshold of the electricity price;

[0077] Construct a second satisfaction model for the user cost according to the upper threshold of the electricity price, the lower threshold of the electricity price, and the maximum electricity price change rate.

[0078] Here, obtaining the allowable change range of the electricity price and the basic electricity price, and determining the upper threshold and the lower threshold of the electricity price can be represented by the following formula (4) and formula (5).

[0079] That is, assume the upper and lower limits of the electricity price are as follows:

[0080]

[0081]

[0082] Among them, in the above formula (4) and formula (5), and represent the upper bound and the lower bound of the electricity price respectively, that is, they represent the upper threshold and the lower threshold of the electricity price respectively; represents the basic electricity price; is the allowable change range of the electricity price; according to the upper threshold and the lower threshold of the electricity price, the total electricity price change rate can be determined and expressed as

[0083] Furthermore, according to the total electricity price change rate, determine the maximum electricity price change rate, which is represented by formula (6), and the maximum electricity price change rate is expressed as:

[0084]

[0085] In order to describe the change of the user's satisfaction with the electricity price in the range of [0, 1], and define the satisfaction with the basic electricity price as 0.5, then the satisfaction with the electricity price is expressed by formula (7) as:

[0086]

[0087] Among them, the upper bound of the user cost satisfaction is 1, and the lower bound is 0, which represent the most satisfied and the most dissatisfied situations respectively. That is, when the electricity price decreases, the logic of the increase in user cost satisfaction is as follows: the upper bound of 1 represents the most satisfied; the lower bound of 0 represents the most dissatisfied. The basic cost satisfaction is 50%, that is, when the electricity price does not change within a day, the user's satisfaction with the electricity bill is 50%.

[0088] Optionally, the above method further includes:

[0089] Obtain the first weight corresponding to the first satisfaction and the second weight corresponding to the second satisfaction with respect to the user cost;

[0090] Establish a comprehensive satisfaction model by weighted means according to the first satisfaction model, the first weight, the second satisfaction model with respect to the user cost, and the second weight.

[0091] That is, the user's comprehensive satisfaction is formed by weighted summation of the user usage pattern satisfaction and the user cost satisfaction, where the user comprehensive satisfaction model is represented by formula (8).

[0092] S a =γ u S u +γ c S c , formula (8);

[0093] Among them, S a is the user comprehensive satisfaction, S u is the first satisfaction, γ u is the first weight corresponding to the first satisfaction; S c is the second satisfaction, γ c is the second weight corresponding to the second satisfaction, and γ u +γ c = 1. The first weight and the second weight can be determined by the decision maker according to the local regional load characteristics to suit different load response requirements.

[0094] The comprehensive satisfaction model is to establish a model of the user for the active distribution network on the principle of ensuring the user's interests. The comprehensive satisfaction model can improve the flexibility of the user to participate in scheduling while transferring the electricity demand.

[0095] In the embodiments of the present application, after linearizing the response curve of the user to the electricity price, an electricity price elasticity matrix is established, and a response model is established based on the electricity price elasticity matrix. This response model can be used to measure the demand response of users within a day, and the daily demand changes of users when the electricity changes are measured through the demand response. According to the load change data of the main line of the active distribution network, a first satisfaction model for the user's usage pattern is established, and, according to the proportion of the electricity price change in the allowable interval of the electricity price, a second satisfaction model for the user's cost is established. The satisfaction situation of the user's usage pattern can be obtained through the first satisfaction model; the satisfaction situation of the user's cost can be obtained through the second satisfaction model; finally, the first satisfaction model and the second satisfaction model are combined to establish a comprehensive satisfaction model, and a comprehensive satisfaction value is determined according to the different weight ratios of the user's usage pattern and the user's cost through the comprehensive satisfaction model.

[0096] According to the obtained comprehensive operating cost of the system, a first objective function for the day-ahead stage is determined; the comprehensive operating cost of the system is: the sum of the cost of purchasing electricity by the transmission network, the cost of controllable distributed generation, the energy storage cost, the cost of interruptible load, and the cost of switched shunt capacitors; according to the first objective function, a day-ahead scheduling model of the active distribution network is constructed. That is, the day-ahead scheduling model satisfies that the sum of the cost of purchasing electricity by the controllable distribution network, the cost of controllable distributed generation, the energy storage cost, the cost of interruptible load, and the cost of switched shunt capacitors is minimized.

[0097] The first objective function is represented by formula (9), and each parameter in formula 9 is represented by formulas (10) to (14):

[0098]

[0099]

[0100]

[0101]

[0102]

[0103]

[0104] In the above formulas (9) to (14), T is preferably 24 hours ahead of schedule. This application assumes that the active distribution network is an independent entity and can purchase electricity from the transmission network. is the cost of purchasing electricity for the transmission network; is the cost of controllable distributed generation; is the energy storage cost; is the cost of switched shunt capacitors; is the cost of interruptible load. is the unit cost of purchasing electricity from the transmission grid, P t Grid is the power of electricity purchased by the active distribution network from the transmission grid at time t; is the unit cost of controllable distributed generation, and are the active power and reactive power output by the controllable distributed generation at time t, respectively; is the unit cost of energy storage charging and discharging, and are the active power of energy storage charging and discharging at time t, respectively; is the operating cost of switched capacitor banks, is the working state of switched capacitor banks; is the unit compensation cost of interruptible load, is the load reduction of the interruptible load at time t.

[0105] According to the first objective function determined by the above formula, a day-ahead scheduling model of the active distribution network can be constructed, and the day-ahead scheduling model can be used to determine day-ahead scheduling data.

[0106] From formulas (9) to (14), the scheduling data includes the power of electricity P purchased by the active distribution network from the transmission grid at time t t Grid , the active power output by the controllable distributed generation at time t and reactive power the active power of energy storage charging and discharging at least one of them.

[0107] Of course, the above first objective formula includes various constraint conditions. Optionally, the constraint conditions of the day-ahead scheduling model include at least one of the following:

[0108] (1), power flow equation constraint conditions;

[0109] (2), safe operation constraint conditions;

[0110] (3), on-load tap-changer constraint conditions;

[0111] (4), distributed generator constraint conditions;

[0112] (5), capacitor bank constraint conditions;

[0113] (6), energy storage element constraint conditions;

[0114] (7), interruptible load constraint conditions.

[0115] In one implementation, the power flow equation constraint condition can be expressed by the following formula (15):

[0116]

[0117] where U and D are respectively subsets of the branches with bus j as the starting segment and the ending segment; P j,t and Q j,t are respectively the active or reactive power injected into bus j; P jk,t and Q jk,t are respectively the active or reactive power of branch jk; r ij and x ij are respectively the resistance and reactance values of branch ij; l ij is the square of the current value of branch ij; v i,t is the square of the voltage value of bus i at time t.

[0118] In another implementation, the safe operation constraint condition can be expressed by the following formula (16):

[0119]

[0120] where, to ensure the safety of the distribution network, the voltages of each bus and the powers of each branch should be limited within a certain range. In formula (16), and are respectively the lower bound and the upper bound of the square of the voltage value of bus i; and are respectively the lower bound and the upper bound of the active power P ij,t transmitted on branch ij; and are respectively the lower bound and the upper bound of the reactive power Q ij,t transmitted on branch ij.

[0121] In another implementation, for branch ij, if there is a on-load tap-changer connected to the front end of the bus, the voltage at the front end of the transformer can be defined as v m, t, and the on-load tap-changer constraint condition can be expressed by the following formula (17):

[0122]

[0123] where, is the tap ratio of the on-load tap-changer; and are respectively the lower bound and the upper bound of the tap ratio of the on-load tap-changer; is the current tap position of the on-load tap-changer; N OLTC represents the total number of taps of the current on-load tap-changer.

[0124] In another implementation, the distributed generators can be divided into two parts: controllable distributed generation (such as gas turbines) and distributed renewable energy. These generators can be integrated into the power grid through inverters, and reactive power can be generated through the inverters. For controllable distributed generation, the output constraints of a gas turbine, for example, can be expressed by the following formula (18):

[0125]

[0126] Where is the rated capacity of the inverter of the controllable distributed generator. and are the lower and upper bounds of the active power output of the distributed generator at time t, and are the lower and upper bounds of the reactive power output of the distributed generator at time t, respectively.

[0127] For distributed renewable energy, the output constraints can be expressed by the following formula (19):

[0128]

[0129] Where and are the active-to-reactive power output of the renewable energy at time t, is the predicted power of the renewable energy, is the rated capacity of the inverter of the renewable energy.

[0130] In another implementation, considering the service life of the capacitor bank, the number of operations of the capacitor bank within the scheduling range is limited. The capacitor bank constraint conditions are expressed by the following formulas (20) to (22):

[0131]

[0132]

[0133]

[0134] Where, in the above formulas (20) to (22), is a 0-1 variable representing the operating state of the capacitor bank, is the reactive power output by the stepped switched capacitor at time t, is the number of steps of the stepped switched capacitor at time t, is the reactive power output per step of the stepped switched capacitor, is the total number of steps of the stepped switched capacitor, For the limit of the number of operations of the switched capacitor per unit time, is the maximum number of operations of the switched capacitor within a day. Equation (20) represents the output characteristics of the capacitor bank; Equation (21) uses 0-1 variables to represent the working state of the capacitor bank; Equation (22) limits the number of operations of the capacitor bank within the scheduling period.

[0135] In another implementation, within the scheduling period, the stored energy should be maintained within a certain range to avoid overcharging and over-discharging. The energy storage cannot be charged and discharged simultaneously. The established energy storage element constraint conditions are expressed as Equation (23):

[0136]

[0137] where, and are the lower and upper bounds of the current stored energy state of the energy storage device respectively; represents the charging state of the energy storage. 0 represents not in the charging state, and 1 represents in the charging state; represents the discharging state of the energy storage. 0 represents not in the discharging state, and 1 represents in the discharging state; and represent the maximum power of the energy storage charging and discharging respectively. η ch represents the net load at time t.

[0138] In another implementation, the interruptible load is taken as part of the demand response model. Customers sign interruptible load contracts with the distribution network operator, reduce part of the load during peak hours, and obtain electricity price compensation. The established interruptible load constraint conditions are expressed as Equation (24):

[0139]

[0140] where, is the load reduction amount, is the original active power of the interruptible load; is a 0-1 variable, representing the state of whether the interruptible load is interrupted. 0 represents interrupted, and 1 represents not interrupted.

[0141] By inputting the obtained response data and user satisfaction into the comprehensive satisfaction model, and inputting the output result of the comprehensive satisfaction model into the day-ahead scheduling model, the day-ahead scheduling data can be determined. That is, in the day-ahead stage, considering demand response and user satisfaction, the day-ahead scheduling data for optimizing the distribution network within a day is obtained, so as to minimize the system cost within a day, maximize the reduction of the operation cost, and reduce the impact of distributed energy access to the distribution network.

[0142] In the embodiments of the present application, by using real-time operation data to determine the operation status of the active distribution network in the intraday stage, and then combining the real-time operation data with the day-ahead scheduling data to determine the optimized data for the next moment of the active distribution network, it is possible to track the scheduling reference given in the day-ahead stage, roll and smooth the power fluctuations, reduce the prediction error, refine and reduce the system operation cost, and increase the consumption of renewable energy.

[0143] In the present application, first, a comprehensive satisfaction model of users for the active distribution network is established based on the principle of ensuring user interests, so as to improve the flexibility of users to participate in scheduling while transferring electricity demand. Secondly, in the day-ahead stage, considering demand response and user satisfaction, the distribution network within a day is optimized to minimize the cost of the system within a day, so as to minimize the operation cost to the greatest extent and reduce the impact of distributed energy access to the distribution network. Then, in the intraday stage, the operation status of the distribution network is optimized by combining the method of model predictive control, tracking the scheduling reference given in the day-ahead stage, rolling and smoothing the power fluctuations, reducing the prediction error, refining and reducing the system operation cost, and increasing the consumption of renewable energy. Finally, the feasibility of the optimized data for the next moment is verified, and the effectiveness of the proposed optimized data for the next moment is proved by the calculation results of the preset verification method, and the effects of reducing the prediction error, refining and reducing the system operation cost, and increasing the consumption of renewable energy can be achieved.

[0144] In an alternative embodiment of the present application, step 13 described above includes:

[0145] Obtain the real-time operation data of the active distribution network in the intraday stage, determine the first difference between the output values of distributed energy in each time period in the real-time operation data and the solution amounts in the corresponding time periods in the day-ahead scheduling data, and the overall fluctuation average value of distributed energy in the day-ahead scheduling data;

[0146] Determine the ratio of the first difference to the average value as the weighting coefficient of the distributed energy;

[0147] Construct a second objective function for smoothing power fluctuations according to the real-time operation data, the day-ahead scheduling data, and the weighting coefficient;

[0148] Determine the scheduling optimization data for the next moment of the active distribution network according to the second objective function.

[0149] In the embodiments of the present application, in the real-time stage, the active distribution network tracks the output reference given in the previous day-ahead stage and smooths the power fluctuations. According to minimizing the weighted sum of power deviations within a time range, a second objective function is established. Specifically, it can be expressed by the following formulas (25) to (27):

[0150]

[0151]

[0152]

[0153] where N is the number of subsets of controllable distributed energy resources, and T is the total number of time periods within a preset time period. is the decision variable of the i-th type of controllable distributed energy resource in the real-time operation data. is the optimization result of the i-th type of controllable distributed energy resource in the day-ahead scheduling data; ζ i is the weighting coefficient of the i-th type of controllable distributed energy resource.

[0154] Equation (25) is for the operation in the real-time stage. The prediction model uses the second objective function for control, aiming to reduce power fluctuations.

[0155] It should be noted that the real-time operation data is continuously updated during the real-time operation stage. Each in the second objective function (Equation (25)) is actually lower than the output data collected during real-time operation, that is, it satisfies Constraint (Equation (19)). Combining the real-time operation data corresponding to this decision variable generates Constraint (Equation (19)), and following the other operation constraints in Equations (15) to (24), with Equation (25) as the objective function, continuously updates the decision variables for each real-time stage, thereby realizing the optimization process in the real-time stage and continuously improving the operation state of the system. Each time the real-time operation data is updated, the prediction model is used for a rolling calculation once, continuously updating the decision variables for each real-time stage until all real-time data is updated, thereby realizing the optimization process in the real-time stage and continuously improving the operation state of the system.

[0156] In a further embodiment of the present application, after obtaining the optimization result in the day-ahead scheduling stage, that is, after determining the day-ahead scheduling data, in the real-time stage, the real-time operation data is combined with the prediction model, and in combination with the real-time operation data and following the operation constraints in Equations (15) to (24), the second objective function of Equation (25) is used to continuously update the decision variables for each real-time stage. In each unit of time, according to the current system state and short-term prediction data, the operation state of the distribution network from the current moment to the short-term future is optimized. Only the decision for the next moment is executed, and other decisions are discarded. By repeating the above process at each moment, the system gradually completes the optimization in the real-time stage, thereby ensuring the determined optimization data for the next moment, correcting the prediction errors in the day-ahead stage, and optimizing the scheduling effect.

[0157]

[0158] ​To prove the effectiveness of the multi-time scale active distribution network scheduling strategy that considers demand response and user comprehensive satisfaction, taking the "IEEE 33 standard test system" as an example, the effectiveness of the effects that can be achieved by the above method is illustrated. The topological schematic diagram of the IEEE 33 standard test system is referred to Figure 2 as shown.

[0159] This test system includes on-load tap-changing transformers, controllable gas turbines, wind turbines, photovoltaics, energy storage, capacitor banks, SVCs, and interruptible loads. In this paper, both the electricity cost of the upstream transmission network and the load electricity price adopt time-of-use electricity price modeling. A preset model is built for the basic electricity value of the distribution network, adding a certain cost to the transmission network. The data of this preset model is shown in Table 1.

[0160] Table 1: Electricity purchase cost

[0161]

[0162]

[0163] The adjustable range of the on-load tap-changing transformer is 0.95 - 1.05, and the step size is 0.025; the rated power of the gas turbine on bus 7 is 800 kVA; the rated power of the distributed wind generator on bus 16 is 600 kVA; the rated power of the photovoltaics on bus 24 is 500 kVA; the capacity of the energy storage on bus 31 is 500 kWh; the total capacity of the capacitor bank on bus 5 is 300 kvar, with each step being 100 kvar. The reactive power output range of the SVC is -300 - 300 kvar; the interruptible loads on buses 13 and 21 are directly controlled by the network operator, and the compensation price for each is 1.2 $ / (kWh).

[0164] Verify the effect of the proposed method when the minimum user satisfaction is 70%. The system's daily load is as Figure 3 shown, and the daily active power output and reactive power output of the distributed energy are as Figure 4 and Figure 5 shown.

[0165] From Figure 4 and Figure 5As can be seen, starting from 6:00, the active and reactive power of the load begins to increase. The active power of the photovoltaic also increases accordingly. Therefore, the active distribution network needs to coordinate the active and reactive power outputs of these distributed energy sources. For wind turbines, since the active power output during the day is relatively low compared to the night time periods (such as 0 - 5:00, 18 - 23:00), the reactive power generated is supplied to the grid through the inverter. In addition, during this period, the active and reactive power outputs of the gas turbine also change, which can be explained by the change in the purchase power cost of the upstream transmission network. During some peak load periods, such as 9:00 - 12:00 and 17:00 - 20:00, the cost of purchasing electricity increases. Therefore, the grid operator will use the gas turbine to generate active power instead of reactive power. Thus, the proposed method can improve the operating state of the system by controlling the active and reactive power outputs of different distributed power sources.

[0166] To study the effect of accommodating renewable energy and reducing network losses, the system net load is defined where N represents the total number of buses. The system network loss is defined In the formula, is the active power output of renewable energy at time t, is the active load of bus i at t, r ij is the resistance of branch ij, l ij,t is the square of the current value of branch ij.

[0167] For simplicity, the method without model predictive control and the method of this application are defined as Method 1 and Method 2. The difference between these two methods is that Method 1 can only be scheduled on a day-ahead time scale, while Method 2 can optimize the distribution network on both day-ahead and real-time time scales, that is, the method proposed in this invention. Taking the period with large fluctuations in renewable energy (9:00 - 16:00) as an example, the calculation results of the two methods are as Figure 6 and Figure 7 shown.

[0168] From Figure 6 and Figure 7 the results, it can be seen that the closer the net load curve of Method 2 is to the actual net load curve, the more renewable energy is consumed. This is because the actual net load curve does not consider the abandonment of wind power and photovoltaic power generation. Figure 6 shows that Method 1 (the method without model predictive control) deviates from the actual net load to a certain extent, while the curve of Method 2 (the method of this application) is closer to the actual net load. From Figure 7It can be further seen that by rolling and adjusting the system operating state in the real-time stage, network losses can be reduced, especially when the renewable energy fluctuates greatly. In this case, due to the large prediction errors, the optimization effect of the scheduling method with a single time scale (such as only day-ahead scheduling) may be limited. In contrast, the method based on model predictive control proposed by the present invention can correct the outputs of different distributed energy sources according to the latest prediction data and the system operating state, thereby improving the absorption of renewable energy and reducing network losses.

[0169] Referring Figure 8 As shown, an embodiment of the present application further provides a scheduling device for an active distribution network based on multiple time scales, including:

[0170] A model establishment module 81, configured to establish a first satisfaction model for the user usage mode according to the electricity price and a response model characterizing the response of the user's power consumption to the electricity price, and establish a comprehensive satisfaction model according to the first satisfaction model and a second satisfaction model for the user cost;

[0171] A day-ahead scheduling data determination module 82, configured to determine day-ahead scheduling data based on the comprehensive satisfaction model through a day-ahead scheduling model, where the day-ahead scheduling model satisfies the minimum cost corresponding to the scheduling data;

[0172] A scheduling optimization module 83, configured to determine the scheduling optimization data for the next moment of the active distribution network according to the real-time operation data of the active distribution network in the intraday stage and the day-ahead scheduling data.

[0173] Optionally, the model establishment module 81 is configured to establish the response model in the following manner:

[0174] According to the response data of the user's power consumption to the electricity price, determine the original load demand, original electricity price data, load change amount, and electricity price change amount of the active distribution network; determine a first ratio of the load change amount to the original load demand, and a second ratio of the original electricity price data to the electricity price change amount; according to a first product of the first ratio and the second ratio, determine the elasticity coefficient for price response; and construct a response model characterizing the response of the user's power consumption to the electricity price according to the elasticity coefficient.

[0175] Optionally, the model establishment module 81 is configured to establish the second satisfaction model in the following manner:

[0176] According to the allowable change range of the electricity price and the base electricity price, determine the electricity price upper limit threshold and the electricity price lower limit threshold; according to the electricity price upper limit threshold and the electricity price lower limit threshold, determine the maximum electricity price change rate; and construct a second satisfaction model for the user cost according to the electricity price upper limit threshold, the electricity price lower limit threshold, and the maximum electricity price change rate.

[0177] Optionally, a model establishment module 81 is configured to establish a comprehensive satisfaction model based on the first satisfaction model and the second satisfaction model for user cost, and is used for:

[0178] Obtaining a first weight corresponding to the first satisfaction and a second weight corresponding to the second satisfaction for user cost; establishing a comprehensive satisfaction model in a weighted manner according to the first satisfaction model, the first weight, the second satisfaction model for user cost, and the second weight.

[0179] Optionally, a scheduling optimization module 83 is configured to determine the scheduling optimization data for the next moment of the active distribution network according to the real-time operation data of the active distribution network in the intraday stage and the day-ahead scheduling data, and is used for:

[0180] Obtaining the real-time operation data of the active distribution network in the intraday stage, determining a first difference between the output values of distributed energy in each period in the real-time operation data and the solution amounts in the corresponding periods in the day-ahead scheduling data, and the overall fluctuation average value of distributed energy in the day-ahead scheduling data; determining the ratio of the first difference to the average value as the weighted coefficient of distributed energy; constructing a second objective function for smoothing power fluctuations according to the real-time operation data, the day-ahead scheduling data, and the weighted coefficient; and determining the scheduling optimization data for the next moment of the active distribution network according to the second objective function.

[0181] Among them, the above implementation embodiments of the scheduling method for the active distribution network based on multiple time scales are all applicable to the embodiments of the scheduling device for the active distribution network based on multiple time scales, and can also achieve the same technical effects.

[0182] A readable storage medium according to an embodiment of the present application stores a program or instruction, and when the program or instruction is executed by a processor, the steps in the scheduling method for the active distribution network based on multiple time scales as described above are implemented, and the same technical effects can be achieved. To avoid repetition, it will not be elaborated here.

[0183] Among them, the processor is the processor in the scheduling method for the active distribution network based on multiple time scales in the above embodiment. The readable storage medium includes computer-readable storage media, such as a computer read-only memory (Read-Only Memory, abbreviated as ROM), a random access memory (Random Access Memory, abbreviated as RAM), a magnetic disk, or an optical disc, etc.

[0184] In the embodiments of the present application, the module can be implemented in software so as to be executed by various types of processors. For example, an identified executable code module can include one or more physical or logical blocks of computer instructions. For example, it can be constructed as an object, a procedure, or a function. Nevertheless, the executable code of the identified module does not need to be physically located together, but can include different instructions stored in different locations. When these instructions are logically combined together, they constitute the module and achieve the specified purpose of the module.

[0185] In fact, the executable code module can be a single instruction or many instructions, and can even be distributed over multiple different code segments, distributed in different programs, and distributed across multiple memory devices. Similarly, the operation data can be identified within the module, and can be implemented in any appropriate form and organized within any appropriate type of data structure. The operation data can be collected as a single data set, or can be distributed at different locations (including on different storage devices), and can at least partially exist only as electronic signals in the system or network.

[0186] The above exemplary embodiments are described with reference to these drawings. Many different forms and embodiments are feasible without departing from the spirit and teachings of the present application. Therefore, the present application should not be construed as being limited to the exemplary embodiments presented herein. Rather, these exemplary embodiments are provided so that the present application will be complete and perfect, and will convey the scope of the present application to those skilled in the art. In these drawings, the component sizes and relative sizes may be exaggerated for clarity. The terms used herein are for the purpose of describing specific exemplary embodiments only and are not intended to be limiting. As used herein, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms as well. It will be further understood that the terms "comprising" and / or "including" when used in this specification, indicate the presence of the stated features, integers, steps, operations, components, and / or groups thereof, but do not preclude the presence or addition of one or more other features, integers, steps, operations, components, groups thereof, and / or others. Unless otherwise indicated, when stating a value range, the range includes the upper and lower limits thereof and any sub-ranges therebetween.

[0187] The above is the preferred embodiment of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle described in the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

Claims

1. A scheduling method for an active distribution network based on multiple time scales, characterized in that, Including: Based on the electricity price and the response model that characterizes the response of user power consumption to the electricity price, establish a first satisfaction model for the user's usage pattern. Based on the first satisfaction model and the second satisfaction model for the user's cost, establish a comprehensive satisfaction model. Based on the comprehensive satisfaction model, through the day-ahead scheduling model, determine the day-ahead scheduling data, and the day-ahead scheduling model satisfies the minimum cost corresponding to the scheduling data. Based on the real-time operation data of the active distribution network in the intraday stage and the day-ahead scheduling data, determine the scheduling optimization data for the next moment of the active distribution network.

2. The method according to claim 1, wherein The response model is established by the following method: Based on the response data of user power consumption to the electricity price, determine the original load demand, original electricity price data, load change amount, and electricity price change amount of the active distribution network. Determine the first ratio of the load change amount to the original load demand, and the second ratio of the original electricity price data to the electricity price change amount. Based on the first product of the first ratio and the second ratio, determine the elasticity coefficient for price response. Based on the elasticity coefficient, construct a response model that characterizes the response of user power consumption to the electricity price.

3. The method according to claim 2, wherein The first satisfaction model is: Among them, S u represents the first satisfaction degree, and are respectively the load value and the load change amount of the controllable distributed energy source i at time period t. N represents the number of subsets of the controllable distributed energy sources, and T represents the total number of time periods within the preset time.

4. The method according to claim 1, characterized in that, The second satisfaction model is established by the following method: Based on the allowable change range of the electricity price and the base electricity price, determine the electricity price upper limit threshold and the electricity price lower limit threshold. Based on the electricity price upper limit threshold and the electricity price lower limit threshold, determine the maximum electricity price change rate. Based on the electricity price upper limit threshold, the electricity price lower limit threshold, and the maximum electricity price change rate, construct a second satisfaction model for the user's cost.

5. The method according to claim 1, characterized in that The establishment of the comprehensive satisfaction model based on the first satisfaction model and the second satisfaction model for the user's cost includes: Obtain the first weight corresponding to the first satisfaction and the second weight corresponding to the second satisfaction for the user's cost. Based on the first satisfaction model, the first weight, the second satisfaction model for the user's cost, and the second weight, establish a comprehensive satisfaction model by weighting.

6. The method according to claim 1, wherein The day-ahead scheduling model satisfies the minimum sum of the costs of purchasing electricity for the controllable distribution network, the costs of controllable distributed generation, the energy storage cost, the cost of interruptible load, and the cost of sectionalized switched capacitors.

7. The method according to claim 6, wherein The constraint conditions of the day-ahead scheduling model include at least one of the following: Power flow equation constraint conditions; Safe operation constraint conditions; On-load tap-changer constraint conditions; Distributed generator constraint conditions; Capacitor bank constraint conditions; Energy storage element constraint conditions; Interruptible load constraint conditions.

8. The method according to claim 1, characterized in that The determination of the scheduling optimization data for the next moment of the active distribution network based on the real-time operation data of the active distribution network in the intraday stage and the day-ahead scheduling data includes: Obtain the real-time operation data of the active distribution network in the intraday stage, and determine the first difference between the output values of distributed energy in each period in the real-time operation data and the solution amount in the corresponding period in the day-ahead scheduling data, and the overall fluctuation average value of distributed energy in the day-ahead scheduling data. Determine the weighted coefficient of distributed energy as the ratio of the first difference to the average value. Construct a second objective function for smoothing power fluctuations based on the real-time operation data, the day-ahead scheduling data, and the weighting factor; Determine the scheduling optimization data for the next moment of the active distribution network according to the second objective function.

9. The method according to claim 8, wherein The second objective function is: where N is the number of subsets of controllable distributed energy sources, and T is the total number of time periods within a preset time; is the decision variable of the i-th controllable distributed energy source at time period t in the real-time operation data; is the optimization result of the i-th controllable distributed energy source at time period t in the day-ahead scheduling data; ζ i is the weighting coefficient of the i-th controllable distributed energy source.

10. A scheduling device for an active distribution network based on multiple time scales, characterized in that, Including: A model establishment module, configured to establish a first satisfaction model for user usage patterns according to electricity prices and a response model characterizing the response of user power consumption to electricity prices, and establish a comprehensive satisfaction model according to the first satisfaction model and a second satisfaction model for user costs; A day-ahead scheduling data determination module, configured to determine day-ahead scheduling data based on the comprehensive satisfaction model through a day-ahead scheduling model, where the day-ahead scheduling model satisfies the minimum cost corresponding to the scheduling data; A scheduling optimization module, configured to determine the scheduling optimization data for the next moment of the active distribution network according to the real-time operation data of the active distribution network in the intraday stage and the day-ahead scheduling data.

11. A readable storage medium having a program or instructions stored thereon, characterized in that, When the program or instruction is executed by a processor, the steps in the scheduling method for an active distribution network based on multiple time scales according to any one of claims 1 to 9 are implemented.

Images