A demand response optimization method considering 5G base station energy storage participating in grid interaction

Abstract

The present invention discloses a demand response optimization method considering the participation of 5G base station energy storage in grid interaction, including the following steps: S1, establishing a two-stage interval optimization model based on the first stage day-ahead stage and the second stage intraday stage, and proposing an objective function; S2, establishing price-type demand response constraints; S3, establishing 5G base station energy storage constraints; S4, establishing other microgrid operation constraints; S5, solving the objective function. Based on the two-stage coordination method of PBDR and energy storage, in order to minimize the operating cost and improve the operating reliability, the first stage determines the day-ahead electricity price of PBDR, and further guides users to transfer load demand to the time period with low electricity price. At the same time, considering the degradation cost, the hourly 5G energy storage SoC limit is optimized to solve the problem of limited time scale of uncertainty prediction. In the second stage, the energy storage is scheduled within the SoC limit optimized day-ahead to track the uncertainty realization and compensate for the PBDR decision deviation in the first stage.

Description

Technical Field

[0001] The present invention relates to the field of microgrid technology, and in particular to a demand response optimization method that considers the participation of 5G base station energy storage in grid interaction. Background Art

[0002] Microgrids can effectively integrate distributed generation (DG) to supply power to local loads. However, the uncertainty of renewable energy and loads may lead to increased system operating costs or violation of system operating constraints.

[0003] The fifth generation mobile communication technology (5G) is a new generation of broadband mobile communication technology with high speed, low latency and large connection characteristics. It is the network infrastructure for realizing the interconnection of man, machine and things. The International Telecommunication Union (ITU) has defined three major application scenarios for 5G, namely enhanced mobile broadband (eMBB), ultra-reliable and low-latency communication (uRLLC) and massive machine-type communication (mMTC). Enhanced mobile broadband (eMBB) is mainly aimed at the explosive growth of mobile Internet traffic, providing mobile Internet users with a more extreme application experience; ultra-reliable and low-latency communication (uRLLC) is mainly aimed at the application needs of vertical industries such as industrial control, telemedicine, and autonomous driving that have extremely high requirements for latency and reliability; massive machine-type communication (mMTC) is mainly aimed at application needs such as smart cities, smart homes, and environmental monitoring that target sensing and data collection. As a new type of mobile communication network, 5G not only solves the communication between people and provides users with more immersive and ultimate service experience such as augmented reality, virtual reality, and ultra-high-definition (3D) video, but also solves the communication problems between people and things, and between things and things, and meets the needs of Internet of Things applications such as mobile medical care, Internet of Vehicles, smart home, industrial control, and environmental monitoring. Ultimately, 5G will penetrate into all industries and fields of the economy and society, and become a key new infrastructure to support the digitalization, networking, and intelligent transformation of the economy and society.

[0004] 5G is developing rapidly, with high transmission rate, high bandwidth, high reliability, low latency and other performance. The power consumption of base station equipment has also increased significantly, which puts forward new requirements for the power supply system of base stations, especially how to meet the basic backup power demand, and it is necessary to carry out capacity expansion calculation for battery packs. At present, most 5G base stations uniformly use lithium batteries as base station batteries to participate in grid peak regulation. The participation of 5G base station energy storage in grid peak regulation, investment cost and output economy also need further study.

[0005] Through literature research, it is found that the SoC limit is a constant within a day, which is not conducive to energy storage operation, and most of the literature does not take the degradation cost of energy storage into account in the microgrid operation model. To this end, a demand response optimization method considering the participation of 5G base station energy storage in grid interaction is proposed to minimize the operation cost and improve the operation reliability. In the first stage, the day-ahead electricity price of PBDR is determined to further guide users to shift load demand to time periods with low electricity prices. At the same time, the hourly 5G energy storage SoC limit is optimized considering the degradation cost to solve the problem of limited time scale of uncertainty prediction. In the second stage, the energy storage is dispatched within the day-ahead optimized SoC limit to track uncertainty realization and compensate for the PBDR decision deviation in the first stage.

[0006] The "A method, device, electronic device and storage medium for day-ahead scheduling of an active distribution network" disclosed in Chinese patent literature has a publication number of CN113541198A. A method, device, electronic device and storage medium for day-ahead scheduling of an active distribution network belongs to the technical field of economic scheduling of active distribution networks. The method comprises: clustering the flexible loads on each bus of the active distribution network; according to the clustering results, establishing an active distribution network day-ahead scheduling optimization model based on the flexible load cluster, solving to obtain the pre-optimization result of the day-ahead scheduling, and obtaining the reference power of each cluster; decomposing and optimizing the reference power of each cluster to obtain the actual power demand of each flexible load in the cluster; using the actual power demand, completing the final optimization of the day-ahead scheduling of the active distribution network based on the flexible load cluster, and obtaining the final day-ahead scheduling optimization plan. The present invention can realize the coordinated scheduling of massive flexible loads in the active distribution network, and has the advantages of high computational efficiency, wide compatibility and strong practicality. However, it does not involve scheduling or optimization plans for other stages other than the day-ahead stage. Summary of the invention

[0007] The present invention solves the above problems and proposes a demand response optimization method that considers the participation of 5G base station energy storage in grid interaction. It is based on a two-stage coordination method of PBDR and energy storage to minimize operating costs and improve operational reliability. In the first stage, the day-ahead electricity price of PBDR is determined to further guide users to transfer load demand to time periods with low electricity prices. At the same time, the hourly 5G energy storage SoC limit is optimized under the consideration of degradation costs to solve the problem of limited time scale of uncertainty prediction. In the second stage, energy storage is scheduled within the day-ahead optimized SoC limit to track uncertainty realization and compensate for the PBDR decision deviation in the first stage. The purpose of the present invention is to minimize operating costs and improve operational reliability under the uncertainty of wind power, photovoltaics and loads.

[0008] In order to achieve the above object, the present invention adopts the following technical solution: a demand response optimization method considering 5G base station energy storage participating in grid interaction, comprising the following steps:

[0009] S1, establish a two-stage interval optimization model based on the first stage day-ahead stage and the second stage intraday stage, and propose an objective function;

[0010] S2, establish price-based demand response constraints;

[0011] S3, establish 5G base station energy storage constraints;

[0012] S4, establish other microgrid operation constraints;

[0013] S5, solving the objective function. In the present invention, based on the coordinated optimization of PBDR and BESS, the first stage belongs to the day-ahead stage, determines the optimal PBDR plan and the optimal SoC limit of BESS per hour, and in the first stage, the scheduling time scale is 24 hours and the time step is 1 hour. The second stage belongs to the intraday stage, scheduling the energy storage plan within the SoC optimal limit to accurately track uncertainty and compensate for the decision deviation of the first stage, and minimize the total operating cost of the microgrid under the premise of meeting the operating constraints. The purpose of the coordinated optimization method of the present invention is to minimize the operating cost and improve the operating reliability under the uncertainty of wind power, photovoltaics and load. The two-stage coordinated optimization method of the price-based demand response and the battery energy storage system of the 5G base station proposed in the present invention aims to minimize the total operating cost and improve the operating reliability. In the first stage, the PBDR scheme of the day-ahead is optimized to improve the utilization efficiency of renewable energy through load transfer. When scheduling the battery energy storage system of the 5G base station within the day, it is necessary to take into account the deviation of uncertain variables (renewable energy and load). Therefore, the energy storage hourly state of charge (SoC) limit needs to be optimized in the day-ahead stage. In the second stage, the dispatch decision of BESS is made within the SoC limit of the day-ahead optimization to track the uncertainty and compensate for the PBDR decision deviation in the first stage.

[0014] Preferably, the objective function is expressed as:

[0015]

[0016] in,

[0017]

[0018]

[0019]

[0020]

[0021] Among them, the first two mins represent minimizing the total operating cost by optimizing x and y respectively, the max and min in brackets [] represent the pessimistic and optimistic situations achieved by the uncertainty variables, x represents the variable of the first-stage price-based demand response, y represents the scheduling optimization variable of the second stage, u represents the uncertainty variable, τ represents the length of the unit time period, C WT represents the operating cost of the fan, represents the unit operation and maintenance cost of the wind turbine, represents the output power of the i-th wind turbine at time t, C PV represents the operating cost of photovoltaics, represents the unit operation and maintenance cost of photovoltaics, represents the output power of the ith photovoltaic at time t, C BESS represents the operating cost of 5G energy storage, represents the degradation cost of the i-th 5G energy storage at time t, C grid Represents the power interaction cost with the main network, represents the price of electricity purchased from the main grid at time t, represents the price of electricity sold to the main grid at time t, P t def represents the power deficit of the microgrid at time t, P t sur represents the excess power of the microgrid at time t, C rev represents the demand response benefit (i.e., the electricity fee paid by the user after implementing demand response), Pr j represents the demand response electricity price at price level j, represents the ideal load response active power without implementing demand response at time t, α j,t A binary variable indicating whether to respond at price level j, L j Represents the demand response rate under price level j. The goal of the model in the present invention is to minimize the total operating cost in pessimistic and optimistic cases respectively. After solving the model, the result of x is retained as the optimization strategy of the first stage, and the variable y can be adjusted in the second stage optimization scheduling according to the actual situation of the uncertainty variable. In addition, the upper and lower limits of the SoC of the BESS are also determined by the pessimistic and optimistic cases of uncertainty obtained together with the first stage decision x. In the second stage, the distributed renewable energy output and load are used as parameters, and the deterministic model is used to optimize the second stage operation.

[0022] Preferably, step S2 is specifically as follows: in price-based demand response, users should adjust their load demand according to the characteristics of their own price-elastic load and the electricity price, and the load demand response P LD The relationship with the changing electricity price Pr can be expressed as:

[0023] P LD =KP r ε

[0024] Among them, K is a constant, ε is the price elasticity coefficient, which can be obtained through historical data; and the price-based demand response is implemented by the method of stepwise approximation, and several price levels are designed. The ideal load response power corresponding to each price level can be obtained from the above formula. In the present invention, the hourly electricity price for the next day needs to be set for the microgrid users one day in advance.

[0025] As a preferred embodiment, the actual power consumption and reactive power consumption of the load after implementing demand response can be expressed as:

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032] in, represents the ideal load response active power without implementing demand response, represents the ideal load response reactive power without implementing demand response, It represents the actual load response active power after implementing demand response. represents the actual load response reactive power after implementing demand response, C rev represents the demand response benefit, Pr0 represents the price level without implementing demand response, α j,t A binary variable indicating whether to respond at price level j, L j represents the demand response rate at price level j. In the present invention, the fourth formula above ensures that only one PBDR price level is valid every hour, the fifth formula ensures that the electricity fee paid by the user after the implementation of demand response cannot exceed the original electricity fee, and the last formula indicates that the total electricity consumption of the user will not decrease after the implementation of demand response. This constraint model is applicable to all microgrid loads, not a single load. In addition, the model also takes into account the uncertainty of the load.

[0033] Preferably, the 5G base station energy storage constraint model is as follows:

[0034]

[0035]

[0036]

[0037]

[0038]

[0039]

[0040]

[0041]

[0042] in, represents the charging binary decision variable of the i-th energy storage at time t, represents the discharge binary decision variable of the i-th energy storage at time t, represents the charging power of the i-th energy storage at time t, represents the discharge power of the i-th energy storage at time t, represents the minimum charging power of the i-th energy storage, represents the maximum charging power of the i-th energy storage, represents the minimum discharge power of the i-th energy storage, represents the maximum discharge power of the i-th energy storage, SoC min Indicates the minimum SoC value, SoC max Indicates the maximum value of SoC, represents the SoC lower limit of the optimization of the i-th energy storage at time t, represents the SoC upper limit value optimized for the i-th energy storage at time t, represents the storage capacity of the i-th energy storage at time t, represents the internal charging and discharging power of the i-th energy storage at time t, represents the capacity of the i-th energy storage, η ch Represents the charging efficiency of energy storage; η dis represents the discharge efficiency of energy storage, represents the initial storage energy of the i-th energy storage, Indicates the storage capacity of the last time interval. In the present invention, the first and second equations indicate that the energy storage charging and discharging power is limited within the minimum and maximum power ranges, the third equation ensures that the charging and discharging processes of the energy storage cannot be carried out simultaneously, the fourth equation describes the net charging power of the energy storage considering the charging and discharging efficiency, the fifth equation indicates the storage capacity after the charging and discharging is completed, the sixth equation indicates limiting the SoC of the energy storage per hour within the optimized SoC limit, in addition, the seventh equation indicates limiting the SoC of the energy storage within the maximum or minimum SoC range to prevent overcharging / discharging, and the last equation indicates ensuring that the storage capacity of the energy storage is equal at the beginning and end of a time period.

[0043] Preferably, the other microgrid operation constraints are as follows:

[0044]

[0045]

[0046]

[0047]

[0048]

[0049]

[0050] Among them, P i,t represents the active power flowing from node i to the trunk branch at time t, Q i,t represents the reactive power flowing from node i to the main branch at time t, P 1,t represents the power flowing from the root node to the main network, represents the active power flowing from node i to the side branch, represents the reactive power flowing from node i to the side branch, R i represents the line resistance of the branch outflowing from node i, X i represents the line reactance of the branch flowing out from node i, V i,t represents the node voltage, V0 represents the reference voltage of the substation, V min Indicates the minimum voltage limit, V max Indicates the maximum voltage limit, S iIt represents the apparent power of the branch. In the present invention, the first to third equations give the linearized distribution network flow, in which the line loss is not considered because its influence is small compared with other cost items in the objective function. The fourth equation represents the exchange power between the microgrid and the main grid. The fifth equation ensures that the node voltage varies within the allowed range. The last equation represents the restriction of the flow of each branch. The feasible domain formed is the circle of real power and reactive power, which can be linearized by the dodecagonal approximation method.

[0051] Preferably, the objective function of the solution includes the following steps: simplifying the model into a main problem and a sub-problem, and performing iterative solution. The main problem is expressed as:

[0052]

[0053]

[0054] The goal of the main problem is to minimize the total system operating cost under pessimistic uncertainty by optimizing x and y simultaneously. When solving the main problem, uncertainty is a known quantity and its pessimistic situation is From the subproblems, the solution to the main problem is obtained and Represents and passes it to the sub-problem;

[0055] The main problem is expressed as:

[0056]

[0057]

[0058]

[0059] Decision variables obtained based on the main problem and It is treated as a constant in the sub-problem solving process, aiming to find the optimistic situation of uncertainty from the interval prediction. and pessimistic situation These two situations correspond to the current decision and The maximum and minimum operating costs under and Will The main problem passed to the next iteration;

[0060] The main problem and sub-problems are solved iteratively until the gap between the main problem and the sub-problem is less than the preset termination threshold δ in the pessimistic case. The subscript k represents the number of iterations. After k iterations, the final first-stage decision And the corresponding optimistic situation and pessimistic situation It is used to determine the upper and lower limits of SoC. In the present invention, in general, a smaller δ will improve the accuracy of the result, but increase the solution time; a larger δ will reduce the accuracy, but reduce the solution time. Considering the accuracy of the result and the calculation speed, this paper adopts δ = 0.001.

[0061] Preferably, the SoC upper and lower limits are determined by the following model:

[0062]

[0063]

[0064]

[0065] In the present invention, the SoC upper limit and lower limit are obtained through a series of calculations.

[0066] The beneficial effects of the present invention are:

[0067] 1. Considering the SoC energy management of 5G energy storage, a two-stage operation method of microgrid coordinated by PBDR and BESS is proposed to minimize the system operation cost;

[0068] 2. An energy storage operation model that takes into account the degradation cost of 5G energy storage is constructed and linearized to improve the calculation efficiency;

[0069] 3. Aiming at the optimization and coordination problem under uncertainty, a two-stage interval optimization method is proposed and a corresponding solution algorithm is proposed. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0071] Example:

[0072] This embodiment proposes a demand response optimization method that considers 5G base station energy storage participating in grid interaction. Figure 1 , mainly includes the following steps: Step S1, establish a two-stage interval optimization model based on the first stage day-ahead stage and the second stage intraday stage, and propose an objective function.

[0073] The objective function is specifically expressed as follows:

[0074]

[0075] in,

[0076]

[0077]

[0078]

[0079]

[0080] Among them, the first two mins represent minimizing the total operating cost by optimizing x and y respectively, the max and min in brackets [] represent the pessimistic and optimistic situations achieved by the uncertainty variables, x represents the variable of the first-stage price-based demand response, y represents the scheduling optimization variable of the second stage, u represents the uncertainty variable, τ represents the length of the unit time period, C WT represents the operating cost of the fan, represents the unit operation and maintenance cost of the wind turbine, represents the output power of the i-th wind turbine at time t, C PV represents the operating cost of photovoltaics, represents the unit operation and maintenance cost of photovoltaics, represents the output power of the ith photovoltaic at time t, C BESS represents the operating cost of 5G energy storage, represents the degradation cost of the i-th 5G energy storage at time t, C grid Represents the power interaction cost with the main network, represents the price of electricity purchased from the main grid at time t, represents the price of electricity sold to the main grid at time t, P t def represents the power deficit of the microgrid at time t, P t sur represents the excess power of the microgrid at time t, C rev represents the demand response benefit (i.e., the electricity fee paid by the user after implementing demand response), Pr j represents the demand response electricity price at price level j, represents the ideal load response active power without implementing demand response at time t, α j,t A binary variable indicating whether to respond at price level j, L j represents the demand response rate at price level j.

[0081] In addition, for the first-stage day-ahead planning model, it uses the day-ahead market electricity price, the 24-hour forecast interval of distributed renewable energy output, and load demand to determine the hourly electricity price provided to microgrid users the next day (price-based demand response guides users' load behavior through electricity prices). The hourly electricity price will be sent to users one day in advance, giving them enough time to react and decide on their energy use behavior. In other words, users' loads will be transferred according to different hourly electricity prices, thereby improving the utilization efficiency of renewable energy. It should be pointed out that the actual renewable energy output and load demand can be any value in the forecast interval. In addition, in order to reduce the impact of uncertainty and improve the reliability of system operation, the hourly SoC limit will also be optimized. The optimized SoC limit can ensure that energy storage can adjust power within this range under uncertainty scenarios. Therefore, the optimization variables in the first stage include the electricity price and SoC limit of price-based demand response.

[0082] In the second-stage intraday dispatch model, as the accuracy of hourly forecasts of renewable energy output and load demand increases, the charging and discharging power of energy storage is optimized within the SoC limit. Unlike the one-time 24-hour dispatch the day before, the intraday stage is a rolling optimization process, and the rolling time window in this article is 4 hours. At the current moment, the latest forecast information of the uncertain quantity is used to optimize the next 4 hours, but the actual decision only executes the optimization results of the first hour, and the rest will be discarded until the next update. Therefore, the optimization variable in the second stage is the charging and discharging power of energy storage.

[0083] The above optimization model is established on the basis of the general form, and the general form of the model is as follows:

[0084]

[0085] stg(x,y,u)=0,h(x,y,u)≤0

[0086] in, F(x ,y,u) represents the objective function, g(x ,y,u)=0 and h(x,y,u)≤0 represent the corresponding equality constraints and inequality constraints respectively.

[0087] The model includes three types of variables:

[0088] 1. x represents the variable of the first-stage price-based demand response, α j,t ;

[0089] 2. y represents the second-stage scheduling optimization variables, including and related variables;

[0090] 3. u represents the uncertainty variable, including and

[0091] For the general form of the model, the first two mins represent minimizing the total operating cost by optimizing x and y respectively, and the max and min in the brackets [] represent the pessimistic and optimistic cases achieved by the uncertainty variables. Therefore, the goal of the model is to minimize the total operating cost in the pessimistic and optimistic cases respectively. After solving the model, the result of x is retained as the optimization strategy for the first stage, while the variable y can be adjusted in the second stage optimization scheduling according to the actual situation of the uncertainty variable. In addition, the upper and lower limits of the SoC of the BESS are also determined by the pessimistic and optimistic cases of the uncertainty obtained together with the first stage decision x. In the second stage, the distributed renewable energy output and load are used as parameters, and the deterministic model is used to optimize the second stage operation.

[0092] Step S2, establishing price-based demand response constraints;

[0093] Specifically, the hourly electricity price for the next day is set for microgrid users one day in advance. In price-based demand response, users should adjust their load demand according to the electricity price based on the characteristics of their own price-elastic load. Load demand response P LD The relationship with the changing electricity price Pr can be expressed as:

[0094] P LD =KP r ε

[0095] Among them, K is a constant, ε is the price elasticity coefficient, which can be obtained through historical data; and the method of stepwise approximation is used to implement price-based demand response, and several price levels are designed. The ideal load response power corresponding to each price level can be obtained from the above formula. Based on this, the actual power consumption and reactive power consumption of the load after implementing demand response can be expressed as:

[0096]

[0097]

[0098]

[0099]

[0100]

[0101]

[0102] in, represents the ideal load response active power without implementing demand response, represents the ideal load response reactive power without implementing demand response, It represents the actual load response active power after implementing demand response. represents the actual load response reactive power after implementing demand response, C rev represents the demand response benefit, Pr0 represents the price level without implementing demand response, α j,t A binary variable indicating whether to respond at price level j, L j represents the demand response rate at price level j.

[0103] also, represents an uncertainty variable. The fourth formula above ensures that only one PBDR price level is valid every hour, the fifth formula ensures that the electricity bill paid by the user after the implementation of demand response cannot exceed the original electricity bill, and the last formula indicates that the total electricity consumption of the user will not decrease after the implementation of demand response. This constraint model is applicable to all microgrid loads, not a single load. In addition, the model also takes into account the uncertainty of the load.

[0104] Step S3, establishing 5G base station energy storage constraints;

[0105] Specifically, in this step, with the widespread application of energy storage technology, the impact of battery degradation on energy storage economic dispatch cannot be ignored. Two technical terms are usually used to evaluate the state of energy storage batteries: energy storage state SoC and discharge depth DoD. SoC is defined as the ratio of the remaining energy of the battery to the full capacity, while DoD refers to the ratio of the charge and discharge energy to the battery capacity. Therefore, if the energy storage is charged at power during the τ time interval If charging and discharging are performed, the DoD in this time interval is expressed as:

[0106]

[0107] in, represents the internal charging and discharging power of the i-th energy storage at time t, Represents the capacity of the i-th energy storage.

[0108] The relationship between energy storage cycle life cycle and DoD can be expressed as:

[0109] L(DoD)=A·DoD -B · e -CDoD ,A,B,C>0

[0110] Among them, L(DoD) is the function of DoD used to calculate the BESS cycle life cycle, A, B, and C are function coefficients. By adjusting the coefficients A, B, and C, the model can be applied to different types of batteries, such as nickel-cadmium batteries, lithium-ion batteries, and lead-acid batteries.

[0111] Although other factors also affect the life of energy storage, this paper takes DoD as the main determinant of BESS life degradation. Therefore, the degradation cost of energy storage as a function of DoD can be expressed as:

[0112]

[0113] Among them, C deg (DoD) is a function of DoD, which is used to calculate the degradation cost of energy storage; represents the replacement cost of the i-th energy storage; η ch Represents the charging efficiency of energy storage; η dis It represents the discharge efficiency of energy storage. The above equation is a nonlinear equation, which can be linearized using piecewise linear approximation. It is worth noting that frequent charging and discharging may cause battery capacity decay, but this effect is small in a short time. Therefore, this article assumes that the battery capacity does not change over time, but in fact the battery capacity is updated regularly.

[0114] Then, a BESS operation model is given, which considers the impact of continuous charging and discharging in adjacent time intervals on battery cost. In this model, a charge / discharge event is counted once until the BESS changes its operating state (from discharging to charging or vice versa). Therefore, the DoD value of continuous charging and discharging between two adjacent time intervals needs to be calculated;

[0115] The 5G base station energy storage constraint model is as follows:

[0116]

[0117]

[0118]

[0119]

[0120]

[0121]

[0122]

[0123]

[0124] in, represents the charging binary decision variable of the i-th energy storage at time t, represents the discharge binary decision variable of the i-th energy storage at time t, represents the charging power of the i-th energy storage at time t, represents the discharge power of the i-th energy storage at time t, represents the minimum charging power of the i-th energy storage, represents the maximum charging power of the i-th energy storage, represents the minimum discharge power of the i-th energy storage, represents the maximum discharge power of the i-th energy storage, SoC min Indicates the minimum SoC value, SoC max Indicates the maximum value of SoC, represents the SoC lower limit of the optimization of the i-th energy storage at time t, represents the SoC upper limit value optimized for the i-th energy storage at time t, represents the storage capacity of the i-th energy storage at time t, represents the internal charging and discharging power of the i-th energy storage at time t, represents the capacity of the i-th energy storage, η ch Represents the charging efficiency of energy storage; η dis represents the discharge efficiency of energy storage, represents the initial storage energy of the i-th energy storage, Indicates the storage capacity of the last time interval. In the present invention, the first and second equations indicate that the energy storage charging and discharging power is limited within the minimum and maximum power ranges, the third equation ensures that the charging and discharging processes of the energy storage cannot be carried out simultaneously, the fourth equation describes the net charging power of the energy storage considering the charging and discharging efficiency, the fifth equation indicates the storage capacity after the charging and discharging is completed, the sixth equation indicates limiting the SoC of the energy storage per hour within the optimized SoC limit, in addition, the seventh equation indicates limiting the SoC of the energy storage within the maximum or minimum SoC range to prevent overcharging / discharging, and the last equation indicates ensuring that the storage capacity of the energy storage is equal at the beginning and end of a time period.

[0125] Considering that the operating state of the energy storage may change between two adjacent time intervals, the binary variable b is used i,t Indicates whether the BESS operating status has changed, as shown below:

[0126]

[0127] When the BESS operates in the same state in two adjacent time intervals, that is, continues to charge and discharge, the above formula is equal to 1. Note that the above formula contains the multiplication of two binary variables. By introducing two additional binary variables and Linearization is performed, respectively:

[0128]

[0129]

[0130] It represents the accumulated energy of the i-th energy storage before the operating state changes at time t, which can be expressed as:

[0131]

[0132] The above formula describes the cumulative calculation of charging and discharging energy if the energy storage continues to charge and discharge in two adjacent time intervals. However, the above formula is nonlinear. By introducing a new variable The above equation can be linearized as follows:

[0133]

[0134] Among them, L and U are The lower and upper bounds of Calculated;

[0135] Therefore, the accumulated degradation cost of BESS can be expressed as:

[0136]

[0137] Here, we take an example to explain the above formula. Assume that there is no energy storage and no charging or discharging before t1. If the BESS discharges 20% DoD in t1 and then continues to discharge 30% DoD in t2, that is and Therefore, the actual degradation costs at t1 and t2 are C deg (0.2) and C deg (0.5)-C deg (0.2). Since C deg The original expression of (DoD) is nonlinear, C deg (0.5)-C deg (0.2) is actually greater than C deg (0.3).

[0138] Step S4, establish other microgrid operation constraints; other microgrid operation constraints are as follows:

[0139]

[0140]

[0141]

[0142]

[0143]

[0144]

[0145] Among them, P i,t represents the active power flowing from node i to the trunk branch at time t, Q i,t represents the reactive power flowing from node i to the main branch at time t, P 1,t represents the power flowing from the root node to the main network, represents the active power flowing from node i to the side branch, represents the reactive power flowing from node i to the side branch, R i represents the line resistance of the branch outflowing from node i, X i represents the line reactance of the branch flowing out from node i, V i,t represents the node voltage, V0 represents the reference voltage of the substation, V min Indicates the minimum voltage limit, V max Indicates the maximum voltage limit, S i It represents the apparent power of the branch. In the present invention, the first to third equations give the linearized distribution network flow, in which the line loss is not considered because its influence is small compared with other cost items in the objective function. The fourth equation represents the exchange power between the microgrid and the main grid. The fifth equation ensures that the node voltage varies within the allowed range. The last equation represents the restriction of the flow of each branch. The feasible domain formed is the circle of real power and reactive power, which can be linearized by the dodecagonal approximation method.

[0146] Step S5, solving the objective function; mainly simplifying the model into a main problem and a sub-problem, and solving iteratively. The main problem is expressed as:

[0147]

[0148]

[0149] The goal of the main problem is to minimize the total system operating cost under pessimistic uncertainty by optimizing x and y simultaneously. When solving the main problem, uncertainty is a known quantity and its pessimistic situation is From the subproblems, the solution to the main problem is obtained and Represents and passes it to the sub-problem;

[0150] The main problem is expressed as:

[0151]

[0152]

[0153]

[0154] Decision variables obtained based on the main problem and It is treated as a constant in the sub-problem solving process, aiming to find the optimistic situation of uncertainty from the interval prediction. and pessimistic situation These two situations correspond to the current decision and The maximum and minimum operating costs under and Will Passed to the main problem for the next iteration.

[0155] The main problem and sub-problems are solved iteratively until the gap between the main problem and the sub-problem is less than the preset termination threshold δ in the pessimistic case. In general, a smaller δ will improve the accuracy of the result but increase the solution time; a larger δ will reduce the accuracy but reduce the solution time.

[0156] Considering the accuracy of the results and the speed of calculation, this paper adopts δ = 0.001. In this model, the subscript k represents the number of iterations. After k iterations, the final first-stage decision And the corresponding optimistic situation and pessimistic situation Used to determine the upper and lower SoC limits.

[0157] Then, the next step is to determine the SoC limit of the energy storage. When the renewable energy output is the lowest and the load is the highest, the constraint may be exceeded. In this extreme case, Used to improve operational reliability. With the decision results of the first stage, the following model is used to solve the decision of the second stage, including the energy storage discharge power in this extreme case It is expressed as follows:

[0158]

[0159]

[0160] In order to consider this extreme condition when determining the SoC limit, the solution is that if the discharge requirement in the extreme case can be met, then the operating constraints in other possible cases should also be met. Therefore, if discharge is required in this extreme case, the difference between two adjacent SoC lower limit values ​​should be large enough to allow the energy storage to discharge. Therefore, the following constraints are added:

[0161]

[0162] Among them, φ is the upper and lower limit interval coefficient, which represents the minimum reserved charging or discharging capacity, that is, the minimum distance between the upper and lower limits of SoC. A larger φ means that a larger capacity is reserved for energy storage, which can flexibly adjust the output power, but it will increase the operating cost of BESS.

[0163] Finally, the upper and lower SoC limits can be determined by the following model:

[0164]

[0165]

[0166]

[0167] In the present invention, based on the coordinated optimization of PBDR and BESS, the first stage belongs to the day-ahead stage, which determines the optimal PBDR plan and the optimal SoC limit of BESS per hour. In the first stage, the scheduling time scale is 24 hours and the time step is 1 hour. The second stage belongs to the intraday stage, which schedules the energy storage plan within the optimal SoC limit to accurately track uncertainty and compensate for the decision deviation of the first stage, and minimize the total operating cost of the microgrid under the premise of meeting the operating constraints. The purpose of the coordinated optimization method of the present invention is to minimize the operating cost and improve the operational reliability under the uncertainty of wind power, photovoltaics and load.

[0168] In the first stage, the day-ahead stage, the optimization variables include the price-based demand response scheme and the SoC limit of 5G base station energy storage. In the second stage, the intraday stage, the hourly charging and discharging power of the energy storage is optimized to ensure that it is scheduled within the day-ahead optimized SoC limit to track uncertainty and compensate for the scheduling decisions of the first stage. On this basis, an energy storage operation model that considers the impact of continuous charging and discharging in adjacent time intervals on the battery degradation cost is established and linearized. In addition, a solution algorithm for the two-stage interval optimization model is established. The proposed coordination method can minimize the total operating cost of the microgrid under uncertainty conditions and improve the operational reliability of the microgrid.

[0169] The two-stage coordinated optimization method of price-based demand response and battery energy storage system of 5G base stations proposed in the present invention aims to minimize the total operating cost and improve operational reliability. In the first stage, the PBDR scheme of the day-ahead is optimized to improve the utilization efficiency of renewable energy through load transfer. When scheduling the battery energy storage system of the 5G base station during the day, it is necessary to take into account the deviation of uncertain variables (renewable energy and load). Therefore, the hourly state of charge (SoC) limit of the energy storage needs to be optimized in the day-ahead stage. In the second stage, the dispatch decision of the BESS is made within the SoC limit optimized day-ahead to track uncertainty and compensate for the PBDR decision deviation in the first stage.

[0170] Since the above embodiments are all described by citing and combining with other embodiments, different embodiments have the same parts, and the same and similar parts between the embodiments in this specification can be referred to each other. No further detailed description is given here.

[0171] It should be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a circuit structure, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such circuit structure, article or device. In the absence of further restrictions, the existence of an element defined by the sentence "includes a..." does not exclude the existence of other identical elements in the circuit structure, article or device including the element.

[0172] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the disclosure of the invention herein. The present application is intended to cover any modification, use or adaptation of the present invention, which follows the general principles of the present application and includes common knowledge or customary technical means in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the content of the claims. The above-described embodiments of the present application do not constitute a limitation on the scope of protection of the present application.

Claims

1. A demand response optimization method considering 5G base station energy storage participating in grid interaction, characterized in that: The following steps are involved: S1, establish a two-stage interval optimization model based on the first stage day-ahead stage and the second stage intraday stage, and propose an objective function; S2, establish price-based demand response constraints; specifically: in price-based demand response, users should adjust their load demand according to the characteristics of their own price-elastic load and the electricity price. The load demand response P LD The relationship with the changing electricity price Pr is expressed as: P LD =KPr ε K is a constant, ε is the price elasticity coefficient, which can be obtained through historical data; S3, establish 5G base station energy storage constraints; S4, establish other microgrid operation constraints; S5, solve the objective function.

2. According to claim 1, a demand response optimization method considering 5G base station energy storage participating in grid interaction is characterized in that: The objective function is expressed as: in, Among them, the first two mins represent minimizing the total operating cost by optimizing x and y respectively, the max and min in brackets [] represent the pessimistic and optimistic situations achieved by the uncertainty variables, x represents the variable of the first-stage price-based demand response, y represents the scheduling optimization variable of the second stage, u represents the uncertainty variable, τ represents the length of the unit time period, C WT represents the operating cost of the fan, represents the unit operation and maintenance cost of the wind turbine, represents the output power of the i-th wind turbine at time t, C PV represents the operating cost of photovoltaics, represents the unit operation and maintenance cost of photovoltaics, represents the output power of the ith photovoltaic at time t, C BESS represents the operating cost of 5G energy storage, represents the degradation cost of the i-th 5G energy storage at time t, C grid Represents the power interaction cost with the main network, represents the price of electricity purchased from the main grid at time t, represents the price of electricity sold to the main grid at time t, P t def represents the power deficit of the microgrid at time t, P t sur represents the excess power of the microgrid at time t, C rev represents the demand response benefit (i.e., the electricity fee paid by the user after implementing demand response), Pr j represents the demand response electricity price at price level j, represents the ideal load response active power without implementing demand response at time t, α j,t A binary variable indicating whether to respond at price level j, L j represents the demand response rate at price level j.

3. A demand response optimization method considering 5G base station energy storage participating in grid interaction according to claim 1 or 2, characterized in that: The step S2 also includes: using a stepwise approximation method to implement price-based demand response, designing several price levels, and the ideal load response power corresponding to each price level can be obtained from the above formula.

4. A demand response optimization method considering 5G base station energy storage participating in grid interaction according to claim 3, characterized in that: The actual power consumption and reactive power consumption of the load after implementing demand response can be expressed as: in, represents the ideal load response active power without implementing demand response, represents the ideal load response reactive power without implementing demand response, It represents the actual load response active power after implementing demand response. represents the actual load response reactive power after implementing demand response, C rev represents the demand response benefit, Pr0 represents the price level without implementing demand response, α j,t A binary variable indicating whether to respond at price level j, L j represents the demand response rate at price level j.

5. According to claim 1, a demand response optimization method considering 5G base station energy storage participating in grid interaction is characterized in that: The 5G base station energy storage constraint model is as follows: in, represents the charging binary decision variable of the i-th energy storage at time t, represents the discharge binary decision variable of the i-th energy storage at time t, represents the charging power of the i-th energy storage at time t, represents the discharge power of the i-th energy storage at time t, represents the minimum charging power of the i-th energy storage, represents the maximum charging power of the i-th energy storage, represents the minimum discharge power of the i-th energy storage, represents the maximum discharge power of the i-th energy storage, SoC min Indicates the minimum SoC value, SoC max Indicates the maximum value of SoC, represents the SoC lower limit of the optimization of the i-th energy storage at time t, represents the SoC upper limit value optimized for the i-th energy storage at time t, represents the storage capacity of the i-th energy storage at time t, represents the internal charging and discharging power of the i-th energy storage at time t, represents the capacity of the i-th energy storage, η ch Represents the charging efficiency of energy storage; η dis represents the discharge efficiency of energy storage, represents the initial storage energy of the i-th energy storage, Indicates the storage capacity of the last time interval.

6. A demand response optimization method considering 5G base station energy storage participating in grid interaction according to claim 1, characterized in that: The other microgrid operation constraints are as follows: Among them, P i,t represents the active power flowing from node i to the trunk branch at time t, Q i,t represents the reactive power flowing from node i to the main branch at time t, P 1,t represents the power flowing from the root node to the main network, represents the active power flowing from node i to the side branch, represents the reactive power flowing from node i to the side branch, R i represents the line resistance of the branch outflowing from node i, X i represents the line reactance of the branch flowing out from node i, V i,t represents the node voltage, V0 represents the reference voltage of the substation, V min Indicates the minimum voltage limit, V max Indicates the maximum voltage limit, S i It represents the apparent power of the branch.

7. A demand response optimization method considering 5G base station energy storage participating in grid interaction according to claim 1, characterized in that: The objective function of the solution includes the following steps: simplifying the model into a main problem and a sub-problem, and performing iterative solution. The main problem is expressed as: The goal of the main problem is to minimize the total system operating cost under pessimistic uncertainty by optimizing x and y simultaneously. When solving the main problem, the uncertainty is a known quantity and its pessimistic situation is From the subproblems, the solution to the main problem is obtained and Represents and passes it to the sub-problem; The main problem is expressed as: Decision variables obtained based on the main problem and It is treated as a constant in the sub-problem solving process, aiming to find the optimistic situation of uncertainty from the interval prediction. and pessimistic situation These two situations correspond to the current decision and The maximum and minimum operating costs under and Will The main problem passed to the next iteration; The main problem and sub-problems are solved iteratively until the gap between the main problem and the sub-problem is less than the preset termination threshold δ in the pessimistic case. The subscript k represents the number of iterations. After k iterations, the final first-stage decision And the corresponding optimistic situation and pessimistic situation Used to determine the upper and lower SoC limits.

8. A demand response optimization method considering 5G base station energy storage participating in grid interaction according to claim 7, characterized in that: The SoC upper and lower limits are determined by the following model:

Images