Distributed energy storage aggregation participating in power grid peak regulation strategy under multi-energy Internet

Abstract

The present invention relates to a method for coordinated operation of distributed energy storage aggregation participating in a power grid, and in particular to a distributed energy storage aggregation participating in a power grid peak-shaving strategy under a multi-energy Internet. It improves the access capacity of renewable energy in the distribution network, increases the economic benefits of the power grid and users, and provides strong support for the development of smart distribution networks. Specifically, the following steps are included: Step 1, establishing an overall control and operation framework for distributed energy storage aggregation participating in power grid peak-shaving; Step 2, establishing a multi-objective optimization model for the coordinated operation of distributed energy storage aggregation and multiple energy sources; Step 3, optimizing and solving the multi-objective optimization model for the coordinated operation of distributed energy storage aggregation and multiple energy sources in stages based on a heuristic intelligent algorithm.

Description

Technical Field

[0001] The present invention relates to a method for coordinated operation of distributed energy storage aggregation participating in a power grid, and in particular to a peak-shaving strategy for distributed energy storage aggregation participating in a power grid under a multi-energy Internet. Background Art

[0002] In the future, the high proportion of distributed energy access will have an increasingly significant impact on the power grid. The energy storage system that organically integrates multiple energy storage to achieve multi-energy complementarity will be the main form of application of energy storage systems in the power grid in the future. As a key technology for the interconnection of new power grids and energy, distributed energy storage can improve the stability of system operation, improve power quality, improve the flexibility of power grid regulation and power supply reliability, realize load peak elimination and valley filling, participate in system demand response, peak regulation, etc. However, the capacity of distributed energy storage units is small, the distribution is uneven, and there are problems such as system operation, management and operation difficulties. How to solve the unified management and application of distributed energy storage and study the economic operation mechanism of different types of distributed energy storage resources. To this end, a large-scale distributed energy storage aggregation method is proposed. Based on cloud platform technology, the coordinated operation of distributed energy storage and multiple energy sources in the power grid is realized. It is a key issue and of great significance to enhance the active support capability of distributed energy storage in the power grid.

[0003] At present, the research on distributed energy storage aggregation peak-shaving at home and abroad is mainly focused on the operation of new energy peak-shaving, and there is relatively little research on distributed energy storage aggregates combined with multi-energy peak-shaving. Through the research of the present invention, it is possible to realize the comprehensive application of various types of distributed energy storage, achieve the matching of multi-energy and power grid peaks and valleys, promote the balance of power grid supply and demand, and ensure the stable operation of the power grid. Relying on the distributed energy storage aggregation method, it is possible to realize the cascade utilization of various energy storage power sources with different characteristics, maximize the active participation of multi-energy in the interactive operation of the power grid, and promote the efficient, stable and economical operation of the distribution network. Relying on the distributed energy storage operation mechanism and cloud energy storage platform studied by the project, it is possible to solve the decentralized grid connection and centralized application mechanism of multiple types of distributed energy, thereby realizing high-density distributed energy storage access and global optimization management of multiple energy sources.

[0004] Domestic and foreign research on distributed energy storage aggregation peak load regulation is currently mainly focused on the participation of a small number of distributed energy storage aggregations in grid operation, and there is relatively little research on large-scale distributed energy storage aggregation combined with multi-energy peak load regulation. In existing research, most of them use battery energy storage technology to track planned power generation, smooth wind power output, improve wind power access to the grid, reduce wind farm wind abandonment, increase wind power utilization hours, accept grid dispatch during new energy grid-connected operation, and play a role in transient active output emergency response and transient voltage emergency support. The application in the field of grid peak load regulation is also limited to local and small-scale areas, and there is relatively little research in the field of large-scale peak load regulation. Summary of the invention

[0005] The present invention aims to address the defects of the prior art and provide a distributed energy storage aggregation participating in the peak load regulation strategy of the power grid under the multi-energy Internet. It can effectively improve the operation reliability of the system, improve the power quality of the system, improve the access capacity of renewable energy in the distribution network, increase the economic benefits of the power grid and users, and provide strong support for the development of smart distribution networks.

[0006] The present invention aims at the problem of distributed energy storage aggregation optimization scheduling participating in grid peak regulation in regional power grid under multi-energy interconnection. Firstly, the peak regulation operation characteristics and influencing factors of existing distributed energy storage in regional power grid are studied, and a distributed energy storage aggregate is established by integrating operation parameters and constraints. A phased optimization solution strategy is proposed, and a heuristic intelligent optimization algorithm method is used. The algorithms and data of the existing scheduling system are fully utilized to provide a practical technical theoretical basis for the implementation of distributed energy storage participating in regional power grid peak regulation.

[0007] To achieve the above-mentioned purpose, the present invention adopts the following technical scheme: in a large-scale regional power grid system, in view of the fact that distributed energy storage is currently unable to participate in power grid peak regulation; distributed energy storage monomers are aggregated into several groups, and through phased optimization control and coordinated control of the dispatching system, distributed energy storage aggregation and multi-energy joint operation are realized to participate in power grid peak regulation in real time, thereby realizing the coordinated control and stable operation of distributed energy storage aggregation participating in the power grid.

[0008] The specific steps include:

[0009] Step 1: Establish an overall control and operation framework for distributed energy storage aggregation to participate in grid peak regulation;

[0010] Step 2: Establish a multi-objective optimization model for distributed energy storage aggregation and multi-energy coordinated operation;

[0011] Step 3: Based on the heuristic intelligent algorithm, the multi-objective optimization model of distributed energy storage aggregation and multi-energy coordinated operation is optimized and solved in stages.

[0012] (Step 4: Establish a Matlab / Simulink platform for distributed energy storage aggregation and multi-energy joint participation in power grid peak regulation strategy;

[0013] Step 5: Conduct simulation analysis on the effectiveness of distributed energy storage aggregation and multi-energy joint participation in power grid peak load regulation strategy, and verify that under multi-energy interconnection, more than 1,000 distributed energy storage monomers can achieve the economy and stability of participating in power system peak load regulation after aggregation.

[0014] Furthermore, the step 1 comprises:

[0015] Step 1.1, establish a multi-objective optimization model for distributed energy storage aggregation and multi-energy coordinated operation, including: distributed energy storage monomer model, distributed energy storage aggregation scheduling model, and multi-energy scheduling model;

[0016] Step 1.2: (Since the established power system multi-source joint peak load optimization model is a nonlinear, multi-stage, dynamic coupling relationship, and large-scale optimization model with mixed continuous and discrete variables;) propose distribution and staged solution rules when performing solution strategy analysis;

[0017] Step 1.3: Use the heuristic intelligent optimization algorithm and the existing dispatching system algorithms and data to prove the feasibility and stability of distributed energy storage aggregation and multi-energy interconnection in participating in grid peak regulation. (Further verify the feasibility of the present invention.)

[0018] Furthermore, the step 2 comprises:

[0019] Step 2.1, establish a charging and discharging model of energy storage monomers, and aggregate multiple energy storages into an equivalent group;

[0020] Step 2.2, after aggregating each energy storage, the equivalent centralized rated power, equivalent centralized rated capacity and equivalent centralized charging and discharging efficiency are used to represent the entire distributed energy storage system;

[0021] Step 2.3, (the energy storage polymer system is combined with other energy sources to jointly participate in the system peak regulation. When the energy storage capacity reaches a certain scale, it can also participate in the system peak regulation alone;) For the multi-source joint peak regulation optimization dispatch of the power grid, the following objective function is established:

[0022] J=Min

[0023]

[0024] In the objective function: the first item: represents the power generation cost of thermal power units in peak-shaving scheduling; the second item: represents the power generation cost of hydropower units in peak-shaving scheduling; the third item: represents the power generation cost of wind power in peak-shaving scheduling, to ensure that the system can accept the maximum wind power under the condition that the operating conditions are met, without sacrificing the high-cost peak-shaving power source; the fourth item: represents the peak-shaving cost of pumped storage units in peak-shaving scheduling; the fifth item: represents the peak-shaving cost of energy storage aggregates in peak-shaving scheduling; the sixth item: represents the peak-shaving cost of thermal power units in peak-shaving scheduling; the seventh item: represents the peak-shaving cost of heat storage devices in peak-shaving scheduling;

[0025] Where: f Gi : thermal power generation cost function; f Hi : hydropower generation cost function; f wi : Wind power generation cost function; C Pu: Peak load regulation cost coefficient of pumped storage unit; C N : Energy storage polymer cost coefficient; C TOf : Start-up and shutdown cost coefficient of thermal power units; f R : cost function of heat storage and heat release power; λ GA , H , w , P , λ N , GB , R : The weight coefficient of the peak-shaving cost of various peak-shaving power sources in the objective function; Pumped storage unit operation identification parameters; Thermal power unit start and stop identification parameters; They are thermal power units, hydropower units, wind power generation capacity, pumped storage unit peak-shaving capacity, energy storage polymer operating capacity, and heat storage or release capacity of heat storage devices.

[0026] Furthermore, the step 3 comprises:

[0027] Step 3.1, by obtaining relevant parameters including: daily load forecast results, regional power receiving plan, photovoltaic power forecast parameters, hydropower generation plan, energy storage battery plan, thermal power unit and other parameters, to calculate the corresponding equivalent parameters, and use the control variable aggregation method to optimize and solve the distributed energy storage aggregation and multi-energy coordinated operation multi-objective optimization model in stages;

[0028] Step 3.2, determine the dispatching strategy of high-cost peak-shaving power sources, ignore the operating cost of conventional power sources, take the lowest peak-shaving cost of high-cost peak-shaving power sources as the goal, and determine the dispatching strategy of each peak-shaving power source based on a heuristic intelligent algorithm to minimize the peak-shaving dispatching cost of the entire high-cost peak-shaving power source;

[0029] Step 3.3: Under the condition that the high-cost peak-shaving power source has been determined, the optimal scheduling of the conventional peak-shaving power source is solved with the lowest operating cost of the conventional peak-shaving power source as the optimization goal; based on the basic principle of staged optimization, the global optimal solution strategy for the optimization problem is obtained.

[0030] Compared with the prior art, the present invention has beneficial effects.

[0031] The distributed energy storage aggregation participating in the grid peak-shaving strategy based on the heuristic intelligent optimization algorithm under the multi-energy interconnection of the present invention refers to that in a large-scale regional power grid system, in view of the situation that the distributed energy storage is currently unable to participate in the grid peak-shaving, the distributed energy storage monomers are aggregated into several groups, and through phased optimization control and coordinated control of the dispatching system, the distributed energy storage aggregation and multi-energy joint operation are realized to participate in the grid peak-shaving in real time, thereby realizing the coordinated control and stable operation of the distributed energy storage aggregation participating in the grid.

[0032] The present invention is convenient for commercial development. With the increasing application of distributed energy storage technology, the requirements for large-scale distributed energy storage aggregation platforms and multiple energy interconnections to participate in power grid operations are higher. The development of coordinated control methods for this system is bound to have a large demand. The present invention has good commercial development prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. The protection scope of the present invention is not limited to the following description.

[0034] Figure 1 It is a block diagram of the distributed energy storage aggregation scheduling structure.

[0035] Figure 2 It is a diagram of the peak load regulation mechanism of the power system.

[0036] Figure 3 It is a basic step block diagram of high-cost peak-shaving scheduling.

[0037] Figure 4 It is a solution strategy diagram based on heuristic intelligent algorithm.

[0038] Figure 5 This is a diagram showing the effect of distributed energy storage aggregation participating in grid peak regulation. DETAILED DESCRIPTION

[0039] The basic idea of ​​the distributed energy storage aggregation participating in the coordinated operation of the power grid under the interconnection of multiple energy sources proposed in the present invention is: for the aggregation of energy storage monomers and the joint participation of multiple energy sources in the grid-connected peak-shaving operation, multiple energy storage monomers are aggregated into groups, and phased optimization control and existing dispatching system control are used to achieve the purpose of distributed energy storage aggregation and multiple energy interconnection participating in the peak-shaving control of the power grid on a large scale.

[0040] like Figure 1-5 As shown, the present invention comprises the following steps:

[0041] Step 1) Establishing the overall control and operation framework for distributed energy storage aggregation to participate in power grid peak regulation;

[0042] (1) Establish a multi-objective optimization model for distributed energy storage aggregation and multi-energy coordinated operation: It is necessary to establish a distributed energy storage monomer model, a distributed energy storage aggregation scheduling model, and a multi-energy scheduling model.

[0043] ①Distributed energy storage monomer model:

[0044]

[0045] Where: SOC0 is the SOC value at the initial moment of energy storage; P ch and P dis are the charge and discharge power from t to t-1, at most one of them can be non-zero; η ch and η dis are the charge and discharge efficiency respectively; Δt is the duration of charge and discharge; S rate is the rated capacity of the energy storage.

[0046] Constraints for energy storage:

[0047] Charge:

[0048]

[0049] Discharge:

[0050]

[0051] Where: P irate is the rated charge and discharge power of energy storage i; P ich and P idis are the charging and discharging power of energy storage i; SOC i is the current SOC value of energy storage i.

[0052] After aggregating the various energy storages, only the equivalent centralized rated power, equivalent centralized rated capacity, and equivalent centralized charging and discharging efficiency are needed to represent the entire distributed energy storage system. When the various energy storages have the same charging and discharging duration at rated power, the corresponding equivalent centralized energy storage parameters are defined as follows:

[0053] The equivalent concentrated rated power is:

[0054]

[0055] Where: P allrate is the rated power P of the equivalent centralized energy storage irate is the rated power of the i-th energy storage.

[0056] The equivalent concentrated rated capacity is

[0057]

[0058] Where: Sallrate is the rated capacity of the equivalent centralized energy storage; S irate is the rated capacity of the i-th energy storage.

[0059] The equivalent concentrated charging efficiency and the equivalent concentrated discharging efficiency are

[0060]

[0061] Where: η allch and η alldis are the equivalent concentrated charging efficiency and the equivalent concentrated discharging efficiency respectively; η ich and η idis are the charging efficiency and discharging efficiency of the i-th energy storage respectively.

[0062] After defining the corresponding equivalent centralized parameters, only one equivalent centralized energy storage is needed to represent all energy storage during scheduling, and the power of each energy storage is allocated using the following formula.

[0063] The charging and discharging power constraints of the distributed energy storage system are:

[0064]

[0065] The SOC constraint of the distributed energy storage system is:

[0066] SOC min ≤SOC all ≤SOC max

[0067] SOC first =SOC last

[0068] Where: P jiess and Q jiess are the active power and reactive power of the energy storage installed at node i in the jth load power interval; SOC all is the SOC value of equivalent centralized energy storage; SOC min and SOC max They are the minimum and maximum values ​​allowed by the distributed energy storage system SOC; SOC first and SOC last are the initial SOC and final SOC of the equivalent centralized energy storage, respectively.

[0069] ② Pumped storage power station dispatching model:

[0070] E f =δ Pu E p ≤E m

[0071] Where: E f and Ep are the power generation of the pumped storage unit and the power required for pumping load; E m is the maximum allowable power generation of the pumped storage power station subject to reservoir capacity constraints; Pu is the conversion efficiency of the pumped storage power station. Then, for each unit of electricity stored, 1 / δ Pu The amount of electricity, that is, the cost corresponding to the unit peak load capacity is: (1 / δ Pu -1)C Pu , where C Pu is the average power generation cost of the system.

[0072] Constraints of the pumped storage power station dispatching model:

[0073] Output constraints of pumped storage units:

[0074] Under power generation conditions, the

[0075]

[0076] Under pumping conditions, the

[0077]

[0078] in: and are the generating power and pumping power of the pumped storage unit, P PuMin and P PuMax are the minimum and maximum power generation of the pumped storage unit, is the power point of the pumped storage unit during pumping operation, and m is the number of power points.

[0079] Upper reservoir capacity constraints:

[0080] For any time period t, it satisfies:

[0081]

[0082] Where: η f and η p are the average water / electricity conversion constants for power generation and pumping respectively; C0 is the initial water volume of the upper reservoir, C Max and C Min are the maximum and minimum water volumes of the upper reservoir of the pumped storage power station, respectively. The above formula expresses the capacity constraint of the upper reservoir during the dispatching period. Since the sum of the total storage capacity of the upper and lower reservoirs remains unchanged during the whole process, it is actually equivalent to limiting the storage capacity of the lower reservoir.

[0083] Water balance constraints:

[0084]

[0085] It means that the pumping volume during the dispatching cycle is balanced with the water consumption for power generation.

[0086] In the above expression, the output of the pumped-storage unit in each period of the dispatching cycle is used as a variable. In peak-shaving dispatching, it is necessary to determine the operating power of each unit in each period under the condition of the minimum operating capacity of the pumped-storage unit.

[0087] ③Heat storage adjustable load scheduling model:

[0088] According to the laws of thermodynamics, the calculation formula is:

[0089] Q t =C×M t ×ΔT

[0090] Among them: Q: represents the heat storage load of the heat storage tank; C: represents the specific heat of the heat storage medium; M: represents the mass of the heat storage medium; ΔT: represents the temperature difference that causes the heat load change; t: represents a certain time period.

[0091] The mass of the heat storage medium is closely related to the volume of the heat storage load. Assuming that the heat storage load is a regular shape, if the bottom area of ​​the heat storage load is S, the water level in the heat storage load at time t is, and the equivalent temperature of the water in the heat storage load at time t is T t , the reference temperature is T0, then

[0092] M t =ρ·S·h

[0093] ΔT=T t -T0

[0094] Substituting the above two equations into

[0095] Q t =C·ρ·S·h·ΔT

[0096] Assuming that the water temperature in the thermal storage load is distributed in a gradient form, the sensors from top to bottom are T t 1 , T t 2 , T t 3 , T t n , then the above formula can be transformed into

[0097]

[0098] After sorting, we get

[0099]

[0100] Define the equivalent temperature of water in the thermal storage load at time t

[0101]

[0102] Then we have:

[0103] Q t =C·ρ·S·h·ΔT t (2-27)

[0104] In the above formula, the heat storage load is related to the medium's specific heat, density, volume and other parameters, among which the medium height can be measured and the equivalent temperature can be controlled. The electric boiler control system formally controls the heat storage load by controlling the equivalent temperature.

[0105] Maximum heat storage capacity of thermal storage load

[0106] Since the specific heat, density and bottom area of ​​water in the thermal storage load do not change during the heat storage and heating process, according to the calculation, when the water level in the thermal storage load reaches the maximum value, the water temperature in the thermal storage load reaches the maximum value, that is,

[0107] Q max =C·ρ·S·h max ·(T max -T0) (2-28)

[0108] Among them, h max : The water level in the heat storage tank when the device is full of water; T max : The temperature value when all the water in the heat storage tank is hot water and the temperature of each layer is consistent with the maximum value.

[0109] Heating power of thermal storage load

[0110] According to the calculation formula of power and energy:

[0111] P·t=Q

[0112] Where: P: heating power of thermal storage load; t: heating power output period of thermal storage load; Q: heating load of thermal storage load.

[0113] Heating power P of thermal storage load in period t g(t) It can be calculated by the following formula:

[0114] P g(t) Δt=Qo ut(t) -Q in(t)

[0115] Among them, Q out(t) , Q in(t) are the heat load injection and heat load outflow of the heat storage load at time

[0116] Assume that at time t, the flow rates of the heat storage load inlet and outlet are G in(t) , G in(t) The unit is generally Kg / h, and the water flow rate is v in(t) 、v out(t) , the tube diameters are D in , D out , then

[0117]

[0118]

[0119] According to the relationship between hot water flow rate and flow rate

[0120]

[0121] Integrating the above formula, we get

[0122]

[0123]

[0124]

[0125] During the off-peak hours at night, the operating mode to be adopted depends on the load situation. Generally, the electric boiler single heat storage mode or heat storage and heating mode is adopted. However, the night load should be controlled, otherwise excessive night load will affect the system heat storage, which may cause excessive operation of the electric boiler the next day and increase the operating cost.

[0126] During the day, the combined heating mode with the priority of the hot water storage tank is adopted, and the hot water storage tank is used to release heat at a constant rate, so as to ensure that the hot water storage tank releases heat evenly and that the heat of the water tank is used up during the working period. When calculating the constant heat release of the water tank, the operation of the electric heat storage during the peak electricity period should be considered. During this period, the hot water storage tank should provide heat at full capacity (i.e., the hot water storage tank single heating mode), so as to minimize the operating costs of the electric boiler.

[0127] (2) Since the established power system multi-source joint peak-shaving optimization model is a nonlinear, multi-stage, dynamically coupled, large-scale optimization model that mixes continuous and discrete variables, a distributed and staged solution rule is proposed when analyzing the solution strategy, which can greatly reduce the complexity of the optimization problem.

[0128] (3) Using a heuristic intelligent optimization algorithm and making full use of the algorithms and data of the existing dispatching system, the feasibility and stability of distributed energy storage aggregation and multi-energy interconnection in participating in grid peak regulation are demonstrated, further verifying the feasibility of the present invention.

[0129] Step 2) Establish a multi-objective optimization model for distributed energy storage aggregation and multi-energy coordinated operation

[0130] (1) Establish a charging and discharging model for energy storage units and aggregate multiple energy storage units into an equivalent group;

[0131] (2) After aggregating the various energy storages, only the equivalent centralized rated power, equivalent centralized rated capacity, and equivalent centralized charging and discharging efficiency are needed to represent the entire distributed energy storage system;

[0132] (3) Energy storage aggregate systems are generally combined with other energy sources to jointly participate in system peak regulation. When the energy storage capacity reaches a certain scale, they can also participate in system peak regulation alone. For the problem of multi-source joint peak regulation optimization scheduling of power grid, the following objective function is established:

[0133] J=Min

[0134]

[0135] In the objective function: the first item: represents the power generation cost of thermal power units in peak-shaving scheduling; the second item: represents the power generation cost of hydropower units in peak-shaving scheduling; the third item: represents the wind power generation cost in peak-shaving scheduling, to ensure that the system can accept the maximum amount of wind power under the conditions of meeting the operating conditions without sacrificing high-cost peak-shaving power sources; the fourth item: represents the peak-shaving cost of pumped storage units in peak-shaving scheduling; the fifth item: represents the peak-shaving cost of energy storage aggregates in peak-shaving scheduling; the sixth item: represents the peak-shaving cost of starting and stopping thermal power units in peak-shaving scheduling; the seventh item: represents the peak-shaving cost of heat storage devices in peak-shaving scheduling.

[0136] Where: f Gi : thermal power generation cost function; f Hi : hydropower generation cost function; f wi : Wind power generation cost function; C Pu : Peak load regulation cost coefficient of pumped storage unit; C N : Energy storage polymer cost coefficient; C TOf : Start-up and shutdown cost coefficient of thermal power units; f R : cost function of heat storage and heat release power; λ GA , H , w , P , λ N , GB , R : The weight coefficient of the peak-shaving cost of various peak-shaving power sources in the objective function; Pumped storage unit operation identification parameters; Thermal power unit start and stop identification parameters; They are thermal power units, hydropower units, wind power generation capacity, pumped storage unit peak-shaving capacity, energy storage polymer operating capacity, and heat storage or release capacity of heat storage devices.

[0137] Step 3) Based on the heuristic intelligent algorithm, the multi-objective optimization model of distributed energy storage aggregation and multi-energy coordinated operation is optimized and solved in stages.

[0138] (1) Obtain relevant parameters such as daily load forecast results, regional power receiving plans, photovoltaic power forecast parameters, hydropower generation plans, energy storage battery plans, thermal power unit equivalent parameters, etc. to calculate the corresponding equivalent compliance parameters, and use the control variable aggregation method to optimize and solve the distributed energy storage aggregation and multi-energy coordinated operation multi-objective optimization model in stages;

[0139] (2) Determine the dispatching strategy of high-cost peak-shaving power sources, ignore the operating costs of conventional power sources, take the lowest peak-shaving cost of high-cost peak-shaving power sources as the goal, and determine the dispatching strategy of each peak-shaving power source based on a heuristic intelligent algorithm to minimize the peak-shaving dispatching cost of the entire high-cost peak-shaving power source;

[0140] Heuristic intelligent algorithm is a computational model of biological evolution process that simulates the natural selection and genetic mechanism of Darwin's theory of biological evolution. It is a method of searching for the optimal solution by simulating the natural evolution process. Heuristic intelligent algorithm starts from a population that represents the potential solution set of the problem, and a population is composed of a certain number of individuals encoded by genes. Each individual is actually an entity with characteristics of chromosomes. As the main carrier of genetic material, chromosomes are a collection of multiple genes. Their internal expression is a certain gene combination, which determines the external expression of the shape of the individual. After the first generation of population is generated, according to the principle of survival of the fittest and survival of the fittest, it evolves from generation to generation to produce better and better approximate solutions. In each generation, individuals are selected according to the fitness of individuals in the problem domain, and combined crossover and mutation are performed with the help of genetic operators to generate a population representing a new solution set. This process will lead to the population being more adaptable to the environment than the previous generation, just like natural evolution. The best individuals in the last generation of population can be decoded and used as the approximate optimal solution to the problem.

[0141] 1) Continuously optimize the solution mode of control variables.

[0142] The power generation capacity of each generator in a thermal power unit is limited to P Gmin and P Gmax Since the maximum and minimum values ​​of each generator are different, the power range of each generator is generally different, and the range of values ​​is large, which increases the difficulty of the search performance of the genetic algorithm. The original data of the unit's power generation is changed as follows:

[0143] P G =(P Gmax -P Gmin )×P * -P Gmin

[0144] Among them, P G is the original power of the unit, P * is a number with a range of 0 to 1. After the above changes, the value of any generator at any time is limited to the same range. Binary encode each power variable parameter of the unit to be solved in the range of 0-1 to obtain substrings, and then connect these substrings into chromosomes.

[0145] 2) Calculation mode of unit start-stop optimization control vector.

[0146] For the optimization of unit start and stop, the start and stop of the unit can be represented by a 0-1 variable "1" for start and "0" for stop. As mentioned above, each unit start and stop variable can be represented by a vector with the same dimension as the study period. Since there are many constraints on the start and stop of the unit, it is very unfavorable to the solution operation if the group is still randomly generated. Therefore, the generated group is required to meet the start and stop constraints of the unit.

[0147] According to the characteristics of power grid peak regulation, the main problem is the pressure load problem of peak regulation. In order to generate feasible initial variables of the start and stop states of the units, the initial vector is generated according to the following rules:

[0148] 1. The initial element values ​​of the unit start and stop vector are all 1;

[0149] 2. Introduce the unit shutdown period variables toff, to ff ∈[1,T], randomly generate t off ;

[0150] 3. Introduce the unit shutdown time variable T off , To ff =TT G on -T G off , randomly generate T off ;

[0151] 4. Generate feasible initial variable vectors for the start and stop states of the unit based on the values ​​of random variables.

[0152] When the heuristic intelligent search algorithm is operated, the variable t of each start-stop unit is off and T off By using this encoding method, the minimum startup time and minimum shutdown time constraints of the unit are included in the binary string of decision variables, which can always be fully satisfied.

[0153] (3) Under the condition that the high-cost peak-shaving power source has been determined, the optimal scheduling of the conventional peak-shaving power source is solved with the lowest operating cost of the conventional peak-shaving power source as the optimization goal. According to the basic principle of staged optimization, the global optimal solution strategy for this optimization problem can be obtained;

[0154] Step 4) Establish a Matlab / Simulink platform for distributed energy storage aggregation and multi-energy joint participation in grid peak regulation strategy.

[0155] Step 5) Conduct simulation analysis on the effectiveness of the distributed energy storage aggregation and multi-energy joint participation in the grid peak load regulation strategy to verify that under the multi-energy interconnection, the distributed energy storage monomer aggregation can achieve the economy and stability of participating in the power system peak load regulation.

[0156] Preferably, distributed energy storage aggregation and multi-energy interconnection jointly participate in the coordinated operation of the power grid to ensure the economy and stability of peak load regulation of the power system.

[0157] Preferably, a multi-objective optimization model for peak-shaving dispatch of power grid considering the aggregation effect of distributed energy storage is established. The model considers various distributed energy forms such as distributed energy storage aggregates, pumped storage, thermal storage adjustable loads, etc., which can be used for day-ahead peak-shaving dispatch of power grid with a time resolution of less than 60 minutes.

[0158] Preferably, the peak-shaving operation characteristics and influencing factors of the existing distributed energy storage in the power grid are studied, a distributed energy storage aggregate is established based on the comprehensive operation parameters and constraints, and a phased optimization solution strategy is proposed.

[0159] Preferably, a heuristic intelligent search algorithm is used, and the algorithms and data of the existing dispatching system are fully utilized to provide a practical technical theoretical basis for the implementation of distributed energy storage participating in power grid peak regulation.

[0160] Figure 1 In the control mode, the aggregator obtains the control authority of the power terminal equipment by signing a contract with the user, issues control instructions on demand, and adjusts the charging and discharging behavior of the distributed energy storage. At this time, the operation, decision-making and control execution of the controlled equipment are all completed by the aggregator. This control mode is bound by the contract and shows high controllability and reliability by taking into account the operating characteristics and charge state of different types of energy storage. For example, the distributed energy storage with insufficient power is charged first, and the distributed energy storage with sufficient energy is discharged first.

[0161] Figure 2In the above, the difference between the upper peak-shaving limit and the daily peak load is the positive spinning reserve reserved by the system, which is generally 3%-5% of the rated capacity of the largest unit or the total load. The difference between the lower peak-shaving limit and the daily minimum load is the negative spinning reserve of the system, which is generally 3%-5% of the minimum load. The critical peak-shaving threshold is the dividing line between the conventional pressure load peak-shaving capacity and the high-cost peak-shaving capacity, which means that the minimum output level that the system can achieve when all conventional pressure load peak-shaving capacities are put into use. When the lower peak-shaving limit is below the critical peak-shaving threshold, it means that high-cost peak-shaving means (such as pumped storage, energy storage polymer peak-shaving, unit start-up and shutdown, etc.) are required to meet the system peak-shaving requirements.

[0162] Figure 3 In the first step, the optimization problem is a large-scale discrete variable optimization problem, which is solved by intelligent optimization search strategy; the second step is the same as the conventional power grid peak-shaving dispatching, which can be solved by using the software package in the dispatching automation system, so as to make the best use of the existing dispatching platform. The key to the problem is how to find the optimal solution of the variable set for the first step optimization problem, that is, to minimize the peak-shaving cost of high-cost peak-shaving power sources, while satisfying various constraints of peak-shaving dispatching.

[0163] Figure 4 In the , heuristic intelligent algorithm is a computational model of biological evolution process that simulates the natural selection and genetic mechanism of Darwin's biological evolution theory. It is a method of searching for the optimal solution by simulating the natural evolution process. The heuristic intelligent algorithm starts from a population that represents the potential solution set of the problem, and a population is composed of a certain number of individuals encoded by genes. Each individual is actually an entity with characteristics of chromosomes. As the main carrier of genetic material, the chromosome is a collection of multiple genes. Its internal expression is a certain gene combination, which determines the external expression of the shape of the individual. After the first generation of population is generated, according to the principle of survival of the fittest and survival of the fittest, it evolves from generation to generation to produce better and better approximate solutions. In each generation, individuals are selected according to the fitness of individuals in the problem domain, and combined crossover and mutation are performed with the help of genetic operators to generate a population representing a new solution set. This process will lead to the population being more adapted to the environment than the previous generation, just like natural evolution. The best individuals in the last generation of population can be decoded and used as the approximate optimal solution to the problem.

[0164] Figure 5 This is a diagram showing the effect of distributed energy storage participating in grid peak regulation.

[0165] (1) Only distributed energy storage aggregation is considered to participate in grid peak regulation.

[0166] (2) Considering pumped storage and distributed energy storage with interconnected adjustable thermal storage devices to participate in grid peak regulation

[0167] In order to simplify the peak-shaving analysis of distributed energy storage in this scenario, the energy storage device will be dispatched and controlled according to the following strategy. For heat storage devices with small capacity and strong dispatching flexibility, the principle of priority input and priority exit is adopted. On the contrary, for devices with large capacity, the number of switching on and off times should be kept as small as possible. During the wind abandonment period, the power grid and wind turbines cooperate to store heat for the heat storage device. When the abandoned wind power is greater than the rated capacity of the heat storage device, the heat storage device is put into operation and the wind turbines store heat. Otherwise, the system abandons wind and obtains electricity from the power grid for heat storage. In the non-wind abandonment stage, the heat storage device only provides heat to meet the heat load demand. The simulation duration is 24 hours, and the time interval is 60 minutes.

[0168] (1) Before the energy storage polymer participated in the peak load regulation: the wind curtailment volume was 3359MWh; the daily load peak-to-valley difference was 3461MW, and the peak-to-valley difference rate was 25.82%. After the energy storage polymer participated in the peak load regulation: the wind curtailment volume was 2359MWh; the daily load peak-to-valley difference was 3051MW, which was 410MW lower than before the peak load regulation. The peak-to-valley difference rate was 23.04%.

[0169] (2) When other conditions remain unchanged, considering pumped storage, distributed energy storage aggregation with interconnected adjustable thermal storage devices participating in the peak load regulation of the power grid, and considering the participation of pumped storage power stations and energy storage aggregates in peak load regulation:; Daily load peak-to-valley difference: The peak-to-valley difference is 2335MW, which is 1126MW lower than before peak load regulation. The peak-to-valley difference rate is 17.42%. Considering thermal storage devices, pumped storage power stations and energy storage aggregates participating in peak load regulation:; Daily load peak-to-valley difference: The peak-to-valley difference is 1875MW, which is 1585MW lower than before peak load regulation. The peak-to-valley difference rate is 13.9%.

[0170] The distributed energy storage aggregation under the multi-energy Internet of the present invention participates in the peak load regulation strategy of the power grid. It can realize the comprehensive application of various distributed energy storage, achieve the matching of multi-energy and power grid peaks and valleys, promote the balance of power grid supply and demand, and ensure the stable operation of the power grid; relying on the distributed energy storage aggregation method studied in the project, it can realize the cascade utilization of various energy storage power sources with different characteristics, maximize the active participation of multi-energy in the interactive operation of the power grid, and promote the efficient, stable and economical operation of the distribution network.

[0171] It can be understood that the above specific description of the present invention is only used to illustrate the present invention and is not limited to the technical solutions described in the embodiments of the present invention. Those skilled in the art should understand that the present invention can still be modified or replaced by equivalents to achieve the same technical effects; as long as the use requirements are met, they are within the protection scope of the present invention.

Claims

1. Distributed energy storage aggregation participates in the grid peak load regulation strategy under the multi-energy Internet, which is characterized by: The following steps are involved: Step 1: Establish an overall control and operation framework for distributed energy storage aggregation to participate in grid peak regulation; Step 2: Establish a multi-objective optimization model for distributed energy storage aggregation and multi-energy coordinated operation; Step 3: Based on the heuristic intelligent algorithm, the multi-objective optimization model of distributed energy storage aggregation and multi-energy coordinated operation is optimized and solved in stages; The step 1 comprises: Step 1.1, establish a multi-objective optimization model for distributed energy storage aggregation and multi-energy coordinated operation, including: distributed energy storage monomer model, distributed energy storage aggregation scheduling model, and multi-energy scheduling model; Step 1.2: Propose distribution and phased solution rules when analyzing the solution strategy; Step 1.3: Use the heuristic intelligent optimization algorithm and the algorithms and data of the existing dispatching system to prove the feasibility and stability of distributed energy storage aggregation and multi-energy interconnection to participate in grid peak regulation; The step 2 comprises: Step 2.1, establish a charging and discharging model of energy storage monomers, and aggregate multiple energy storages into an equivalent group; Step 2.2, after aggregating each energy storage, the equivalent centralized rated power, equivalent centralized rated capacity and equivalent centralized charging and discharging efficiency are used to represent the entire distributed energy storage system; Step 2.3: For the optimal dispatching of multi-source joint peak load regulation of power grid, the following objective function is established: In the objective function: the first item: represents the power generation cost of thermal power units in peak-shaving scheduling; the second item: represents the power generation cost of hydropower units in peak-shaving scheduling; the third item: represents the power generation cost of wind power in peak-shaving scheduling, so as to ensure that the system can accept the maximum wind power under the condition that the operating conditions are met, without sacrificing the high-cost peak-shaving power source; the fourth item: represents the peak-shaving cost of pumped storage units in peak-shaving scheduling; the fifth item: represents the peak-shaving cost of energy storage aggregates in peak-shaving scheduling; the sixth item: represents the peak-shaving cost of thermal power units in peak-shaving scheduling; the seventh item: represents the peak-shaving cost of heat storage devices in peak-shaving scheduling; Where: f Gi : thermal power generation cost function; f Hi : hydropower generation cost function; f wi : Wind power generation cost function; C Pu : Peak load regulation cost coefficient of pumped storage unit; C N : Energy storage polymer cost coefficient; C TOf : Start-up and shutdown cost coefficient of thermal power units; f R : Cost function of heat storage and heat release power; λ GA , H , w , P , N , GB , R : The weight coefficient of the peak-shaving cost of various peak-shaving power sources in the objective function; Pumped storage unit operation identification parameters; Thermal power unit start and stop identification parameters; They are thermal power units, hydropower units, wind power generation capacity, pumped storage unit peak-shaving capacity, energy storage aggregate operating capacity, and heat storage or heat release capacity of heat storage devices; The step 3 comprises: Step 3.1, by obtaining relevant parameters including: daily load forecast results, regional power receiving plan, photovoltaic power forecast parameters, hydropower generation plan, energy storage battery plan, thermal power unit and other parameters, to calculate the corresponding equivalent parameters, and use the control variable aggregation method to optimize and solve the distributed energy storage aggregation and multi-energy coordinated operation multi-objective optimization model in stages; Step 3.2, determine the dispatching strategy of high-cost peak-shaving power sources, ignore the operating cost of conventional power sources, take the lowest peak-shaving cost of high-cost peak-shaving power sources as the goal, and determine the dispatching strategy of each peak-shaving power source based on a heuristic intelligent algorithm to minimize the peak-shaving dispatching cost of the entire high-cost peak-shaving power source; Step 3.3: Under the condition that the high-cost peak-shaving power source has been determined, the optimal scheduling of the conventional peak-shaving power source is solved with the lowest operating cost of the conventional peak-shaving power source as the optimization goal; based on the basic principle of staged optimization, the global optimal solution strategy for the optimization problem is obtained.

Images