Four-quadrant energy storage system optimal configuration method considering power grid adjustment demand
By establishing an optimized configuration model for multi-type adjustment requirements, optimizing the four-quadrant characteristics of the energy storage system, the problem of failure to fully utilize the potential of the energy storage system in the existing technology has been solved, the grid stability and investment returns have been improved, and the grid has been achieved. Effective consideration of type adjustment needs.
Patent Information
- Application Number
- CN202411189139.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-12-06
AI Technical Summary
The existing technology ignores the needs of multiple types of grid regulation in the optimized configuration of energy storage systems, and fails to fully tap the potential of energy storage systems, resulting in insufficient grid frequency stability and investment returns.
A four-quadrant energy storage system optimization configuration method considering the demand for power grid regulation is adopted. By establishing an optimized configuration model of multi-type adjustment requirements, combining the active-reactive adjustment characteristics, second-order cone relaxation, variable substitution, large M method and fraction The segment linearization method processes nonlinear terms and converts them into a mixed integer second-order cone planning model to optimize the single-day operating cost of the distribution network.
It improves the utilization rate of the power capacity of the energy storage system, reduces the peak shaving pressure of the power grid, enhances the stability of the power grid and investment returns, adapts to the focus of different grid regulation needs, and provides differentiated access solutions.
Smart Images

Figure CN119093428A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to an energy storage system optimization configuration method, and in particular to a four-quadrant energy storage system optimization configuration method considering power grid regulation requirements. Background Art
[0002] Wind power generation and photovoltaic power generation have achieved rapid development due to their clean, environmentally friendly and resource-rich characteristics. Due to the randomness and volatility of new energy output, the grid voltage fluctuates frequently, and the risk of over-limit increases. Its reverse peak-shaving characteristics also require more peak-shaving capacity in the power system to maintain the power balance of the system. The inertia of new energy units is lower than that of synchronous units, which makes the grid frequency stability face new challenges. The energy storage system has the characteristics of high energy density and rapid and independent absorption and release of active and reactive power. Reasonable configuration of energy storage systems in the power grid will effectively deal with the above problems.
[0003] Existing research on energy storage optimization configuration mostly focuses on a single grid regulation demand. However, further research has found that ignoring the other service values of energy storage systems will underestimate the benefits they bring. Energy storage optimization configuration methods that consider multiple types of grid regulation needs can fully tap the potential of energy storage, improve equipment utilization, and increase investment returns. At present, research on four-quadrant energy storage configuration that takes into account multiple types of grid regulation needs is still insufficient, and it is necessary to establish a unified analysis method that considers multiple types of regulation needs such as frequency regulation, peak regulation, and voltage regulation. In addition, in the issue of energy storage system access, a flexible energy storage optimization configuration method is needed according to the different emphasis of grid regulation needs. Summary of the invention
[0004] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide a method for optimizing the configuration of a four-quadrant energy storage system that can fully utilize the four-quadrant characteristics of the energy storage system, tap the potential of the energy storage system, and take into account the grid regulation needs.
[0005] The technical solution adopted by the present invention is: a four-quadrant energy storage system optimization configuration method considering the grid regulation demand, comprising the following steps:
[0006] 1) According to the active-reactive regulation characteristics of the four-quadrant energy storage system, the four-quadrant power characteristics of the energy storage system and the way the energy storage system participates in the grid regulation are controlled;
[0007] 2) Considering the regulation mechanism of the four-quadrant energy storage system in the scenarios of peak load regulation, frequency regulation and voltage regulation of the power grid, a four-quadrant energy storage optimization configuration model considering the multi-type regulation needs of the power grid is established. The objective function of the model is the daily operation cost of the distribution network, taking into account the multi-type regulation needs of the power grid. On the basis of the average daily investment and operation and maintenance cost, the power purchase cost of the distribution network and the network loss cost, the peak load penalty item that comprehensively considers the net load variance and the feeder power variance, the inertia penalty item that considers the system frequency regulation inertia demand, and the weighted node voltage deviation penalty item are added to the objective function; the constraints of the model are: considering the site selection and capacity constraints of the four-quadrant energy storage system, the constraints of the four-quadrant energy storage system, the constraints of the wind power and photovoltaic power generation system, the network flow constraints, the reactive equipment constraints, the virtual inertia constraints, and the safe operation constraints;
[0008] 3) Use second-order cone relaxation, variable substitution, big M method and piecewise linearization method to deal with nonlinear terms and non-cone functions in the model, and then transform the difficult-to-solve model into a mixed integer second-order cone programming model.
[0009] The present invention provides a four-quadrant energy storage system optimization configuration method considering the grid regulation requirements, which has the following advantages:
[0010] 1. The present invention analyzes the four-quadrant characteristics of energy storage and constructs a four-quadrant energy storage configuration mathematical model containing wind power active and reactive output, photovoltaic active and reactive output, and reactive equipment output;
[0011] 2. The present invention can optimize the distribution network cost and the total energy storage configuration cost while better taking into account the grid regulation needs of frequency regulation, voltage regulation and peak regulation. It can fully tap the potential of the four quadrants of energy storage, especially improve the utilization rate of the power capacity of the configured energy storage system, and increase the investment and construction benefits of energy storage;
[0012] 3. The model of the present invention comprehensively considers the regulation requirements of power grid peak regulation, frequency regulation and voltage regulation, integrates the net load variance and feeder power variance, and reflects the peak-valley characteristics of the system at both ends; determines the minimum inertia of the system based on the maximum frequency conversion rate of the system, thereby obtaining the inertia shortage of the system; and uses the node load proportion to correct the average voltage deviation of the system;
[0013] 4. The present invention can formulate differentiated access solutions for energy storage according to the different grid regulation requirements in different scenarios / complex scenarios, and provide guidance and reference for grid energy storage configuration. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Figure 1 It is a schematic diagram of the four-quadrant operation of the energy storage system;
[0015] Figure 2 It is the load and wind and solar maximum output curve used in the example;
[0016] Figure 3It is the distribution network power purchase curve and time-of-use electricity price in scenario 3 of the example;
[0017] Figure 4a is the BESS active power and power configured by node 18 in scenario 3 in the example;
[0018] Figure 4b is the BESS active power and power configured by node 30 in scenario 3 in the example;
[0019] Figure 5 is the average node voltage for scenarios 1-3 in the example. DETAILED DESCRIPTION
[0020] The following is a detailed description of a four-quadrant energy storage system optimization configuration method considering grid regulation requirements of the present invention in conjunction with the embodiments and drawings.
[0021] 1. A method for optimizing the configuration of a four-quadrant energy storage system considering grid regulation requirements, characterized in that it comprises the following steps:
[0022] 1) According to the active-reactive regulation characteristics of the four-quadrant energy storage system, the four-quadrant power characteristics of the four-quadrant energy storage system and the way the energy storage system participates in the grid regulation are controlled; the four-quadrant operating energy storage system with flexible bidirectional regulation capability is composed of an energy storage system and a power conversion system. The output voltage and phase of the four-quadrant operating energy storage system are adjusted by controlling whether the device switch is turned on or off, thereby realizing active-reactive decoupling control and four-quadrant fast and smooth adjustment.
[0023] The schematic diagram of the four-quadrant operation of the energy storage system is as follows: Figure 1 As shown, when the power grid needs active and reactive power, the four-quadrant energy storage system discharges in the first quadrant and provides reactive power; when the power grid has abundant active power and reactive power, the four-quadrant energy storage system charges in the third quadrant and absorbs reactive power; when the power grid has abundant active power but lacks reactive power, the four-quadrant energy storage system charges in the second quadrant and provides reactive power; when the power grid lacks active power but has abundant reactive power, the four-quadrant energy storage system discharges in the fourth quadrant and absorbs reactive power.
[0024] The power load and renewable energy output fluctuate significantly in different time periods, resulting in large differences in power generation and load peaks and valleys. The four-quadrant energy storage system stores electricity during low periods and discharges during peak periods, smoothing the load curve, reducing the start and stop of generator sets, and reducing peak load pressure. The frequency of the power system is related to the balance of active power, and traditional frequency regulation methods have slow response and low efficiency. The four-quadrant energy storage system can respond to frequency deviations in milliseconds, maintain frequency stability through two-way regulation, improve frequency regulation accuracy and efficiency, and reduce wear and maintenance costs of generator sets. In addition, the four-quadrant energy storage system can adjust the inverter control strategy, provide or absorb reactive power, respond quickly, and dynamically adjust the voltage to ensure a reasonable voltage level.
[0025] In summary, the four-quadrant energy storage system adjusts active and reactive power, works in coordination with other power equipment, and uses advanced control algorithms to achieve multi-objective optimization control. Its multifunctional regulation capability significantly improves the reliability, flexibility, and stability of the power system, and has broad application prospects.
[0026] 2) Considering the regulation mechanism of the four-quadrant energy storage system in complex scenarios such as peak regulation, frequency regulation, and voltage regulation of the power grid, a four-quadrant energy storage optimization configuration model considering the multi-type regulation needs of the power grid is established. The objective function of the model is the daily operation cost of the distribution network, taking into account the multi-type regulation needs of the power grid. On the basis of the average daily investment and operation and maintenance cost, the distribution network power purchase cost, and the network loss cost, the peak regulation penalty item that comprehensively considers the net load variance and the feeder power variance, the inertia penalty item that considers the system frequency regulation inertia demand, and the weighted node voltage deviation penalty item are added to the objective function; the constraints of the model are: considering the site selection and capacity constraints of the four-quadrant energy storage system, the constraints of the four-quadrant energy storage system, the constraints of the wind power and photovoltaic power generation system, the network flow constraints, the reactive equipment constraints, the virtual inertia constraints, and the safe operation constraints;
[0027] 1. The objective function for establishing the four-quadrant energy storage optimization configuration considering the multi-type regulation requirements of the power grid is as follows:
[0028] min F=F BESS +(F elec +F loss )+F p +F H +F v (1)
[0029] Where, F is the daily operation cost of the distribution network; F BESS is the daily investment and operation and maintenance cost of the four-quadrant energy storage system; F elec F is the cost of purchasing electricity from the distribution network; loss is the network loss cost; F p is the peak load penalty term that comprehensively considers the net load variance and feeder power variance; F H F is the inertia penalty term considering the system frequency regulation inertia requirement; v is the weighted node voltage deviation penalty term; where,
[0030] (1.1) Daily investment and operation and maintenance cost of four-quadrant energy storage system F BESS :
[0031]
[0032] In the formula, f inv is the investment and construction cost of the four-quadrant energy storage system; f mis the annual operation and maintenance cost of the four-quadrant energy storage system; C is the annual coefficient; the annual investment cost of energy storage is converted to daily calculation, so the denominator is the number of days in a year; where:
[0033]
[0034] In the formula, and are the rated power and rated capacity of the four-quadrant energy storage system at node i; k S and k E are the unit rated power price and unit rated capacity price of the four-quadrant energy storage system respectively; k m is the operation and maintenance cost rate of the four-quadrant energy storage system; r is the discount rate, which is 0.1; y is the life of the energy storage system in years;
[0035] (1.2) Distribution network power purchase cost F elec :
[0036] The distribution network needs to obtain electric energy from the upper power grid through the root node, which is the main source of electric energy for the distribution network. The electricity price varies at different times, which is conducive to the four-quadrant energy storage system to achieve peak-valley arbitrage based on time-of-use electricity prices and improve the peak-valley characteristics of the power grid.
[0037]
[0038] In the formula, c pr,t is the electricity price at time t; P 0,t is the power purchased by the root node at time t; T is the number of hours in a day;
[0039] (1.3) Network loss cost F loss :
[0040]
[0041] In the formula, r ij is the resistance of branch ij; B is the set of nodes i and j that can form a branch; is the square of the current in branch ij at time t;
[0042] (1.4) Peak load penalty term F p :
[0043] The peak load penalty term includes the net load variance D l and root node feeder power variance D im ,in,
[0044] Net load variance D l , can be applied in the distribution network to reflect the peak and valley characteristics:
[0045]
[0046] Where P load,t is the grid load at time t; P t Wind is the wind power output at time t; P t PV is the photovoltaic power output at time t; P t BESS is the output of the four-quadrant energy storage system at time t, where discharge is positive and charge is negative; P l,av is the average net load value;
[0047] Root node feeder power variance D im , which can reflect the power supply requirements of the distribution network to the upper power grid at the source:
[0048]
[0049] Where P im,t is the root node feeder power at time t; P im,av is the average feeder power;
[0050] Taking into account the net load variance and feeder power variance, the peak load penalty term is defined as follows:
[0051] F p =α l D l +α im D im (8)
[0052] Among them, α l With α im is the peak load penalty coefficient;
[0053] (1.5) Inertia penalty term F H :
[0054] The system needs sufficient inertia support to ensure frequency stability, and when the inertia support is insufficient, the lower-level power grid needs to make inertia demands to the upper-level power grid;
[0055] F H =C H S g (9)
[0056] In the formula, C H is the inertia penalty coefficient, S g is the required capacity of the upper synchronous unit;
[0057] (1.6) Weighted node voltage deviation penalty term F v :
[0058] The calculation formula of node voltage deviation DI is as follows:
[0059]
[0060] Where U i,t is the voltage value of node i at time t; U e is the rated voltage; N is the total number of nodes;
[0061] The influence of node voltage deviation on different node loads is different. The larger the load of the node, the more important its voltage stability is. Therefore, the node voltage deviation is corrected according to the load weight, and the load weight coefficient χ at node i at time t is defined as i,t as follows:
[0062]
[0063] Where P load,i,t is the load value of node i at time t; is the average load of all nodes at time t;
[0064] Then we can obtain the weighted node voltage deviation penalty term F v :
[0065]
[0066] Where μ is the proportional conversion coefficient of active power loss and node voltage deviation.
[0067] 2. The constraints are as follows:
[0068] (2.1) Four-quadrant energy storage system site selection and capacity constraints
[0069] Due to limited investment costs, the total number of installation locations and the maximum installable rated power and rated capacity need to be limited;
[0070]
[0071] Where N is the total number of nodes; is a 0-1 variable for the site selection decision of the four-quadrant energy storage system at node i, 1 represents that node i is configured with energy storage, and 0 represents that node i is not configured with energy storage; U is the maximum number of nodes that can be configured with a four-quadrant energy storage system in the distribution network; Configure the rated power of the four-quadrant energy storage system for node i; and Configure the upper and lower limits of the rated power of the four-quadrant energy storage system for each i-node; Configure the rated capacity of the four-quadrant energy storage system for node i; Configure the upper and lower limits of the rated capacity of the four-quadrant energy storage system for node i respectively;
[0072] (2.2) Constraints of four-quadrant energy storage system:
[0073] Introducing 0-1 variables and Represents the charging and discharging state of the four-quadrant energy storage system at node i at time t, and satisfies:
[0074]
[0075] is 1, When it is 0, the four-quadrant energy storage system is charged; when is 0, When it is 1, the four-quadrant energy storage system discharges; when and When it is 0 at the same time, it means that node i is not configured with a four-quadrant energy storage system or is neither charging nor discharging at time t;
[0076] The power of the four-quadrant energy storage system should meet the following requirements:
[0077]
[0078] In the formula, are the computational intermediate quantities of the charging power and discharging power of the four-quadrant energy storage system at node i at time t; are the reactive power output and residual power of the four-quadrant energy storage system at node i at time t; min and λ max are the minimum and maximum state of charge coefficients, respectively;
[0079] The relationship between the battery's output power and capacity is as follows:
[0080]
[0081] In the formula, are the power of the four-quadrant energy storage system at node i at time t+1 and time t respectively; σ is the charging and discharging efficiency of the four-quadrant energy storage system; Δt is an operating period;
[0082] In order to ensure the continuity of energy storage, the daily initial power of the four-quadrant energy storage system is specified. End of day power The same, that is:
[0083]
[0084] (2.3) Wind and photovoltaic power generation system constraints:
[0085] When the capacity of wind power generation system and photovoltaic power generation system is given, the actual output of each wind power and photovoltaic power source is affected by external factors such as wind speed, temperature, and light. The maximum output fluctuates over time. Considering the reactive output of wind power and photovoltaic, the power factor of the wind power and photovoltaic power generation system is fixed at 0.95.
[0086]
[0087] In the formula, is the photovoltaic active power output of node i at time t; is the upper limit of photovoltaic active output at node i at time t; is the photovoltaic active power output of node i at time t; is the upper limit of photovoltaic active output at node i at time t; is the photovoltaic reactive power output of node i at time t; is the wind power reactive power output at node i at time t; is the photovoltaic reactive power output of node i at time t; is the wind power reactive power output of node i at time t;
[0088] (2.4) Network flow constraints:
[0089] The application line ij needs to satisfy Ohm's law, branch flow and power balance equations, that is, the DistFlow model is used to represent the network flow constraints:
[0090] U i -U j =z ij I ij (twenty one)
[0091]
[0092] Where i and j are node numbers; U i and U j are the voltages of nodes i and j respectively; I ij is the current flowing from node i to node j; z ij is the impedance of branch ij; S ij is the power flowing from node i to node j; is the total power flowing from upstream node i to node j; is the total power flowing from node j to downstream node k; For I ij The conjugate vector of j The power injected into node j;
[0093] (2.5) Reactive equipment constraints:
[0094] The power grid operates normally and is equipped with reactive equipment, including group switching capacitors and continuous reactive compensation devices:
[0095]
[0096] In the formula, is the reactive power output of the grouped switched capacitors at node i at time t; is the gear position of node i at time t; The gear compensation amount set for node i; N CB is the upper limit of the gear position; are the upper and lower limits of the reactive power output of the continuous reactive power compensation device of node i at time t respectively; is the reactive power output of the continuous reactive power compensation device of node i at time t;
[0097] (2.5) Virtual inertia constraint:
[0098] Minimum inertia of power system H min The actual inertia H that the power system can provide sys The comparison is an important way to measure the power system's ability to resist frequency fluctuations. Both wind power systems and energy storage systems can provide virtual inertia. For the stable operation of the power system, the total inertia of the power system must be greater than the minimum inertia of the power system.
[0099]
[0100] Where ΔP 1 is the disturbance power; f n is the power system frequency; RoCoF max is the maximum frequency change rate; S B is the power system capacity; H g is the inertia of the synchronous machine; S g is the synchronous machine capacity; Virtual inertia provided for the fan; S Wind is the fan capacity; Virtual inertia provided for the four-quadrant energy storage system; S BESS The total power of the four-quadrant energy storage system configured for the full power system; The maximum virtual inertia provided for the fan; ω r0 is the initial speed of the fan rotor; H w is the inherent inertia time constant of the fan;
[0101] (2.7) Safety operation constraints:
[0102]
[0103] Where U i,t is the voltage value of node i at time t; Ui+1,t is the voltage value of node i at time t+1; U min and U max are the lower and upper limits of the node voltage respectively; U e is the rated voltage; I ij,t is the current value of branch ij at time t; I max The maximum current carrying capacity of the line.
[0104] 3) Use second-order cone relaxation, variable substitution, big M method and piecewise linearization method to deal with nonlinear terms and non-cone functions in the model, and then transform the difficult-to-solve model into a mixed integer second-order cone programming model; including:
[0105] Substituting formula (22) into formula (21), we obtain:
[0106]
[0107] Where: i, j are node numbers; U i and U j are the voltages at nodes i and j respectively; For U i The conjugate vector of ij is the impedance of branch ij; For S ij The conjugate vector of ij is the power flowing from node i to node j;
[0108] Multiply both sides of equation (27) by their respective conjugates, and let J ij =|I ij 2 , V i =U i 2 , V j =U j 2 get:
[0109]
[0110] In the formula, For z ij The conjugate vector of ;
[0111] Expanding formula (28), we get:
[0112]
[0113] Where P ij is the active power flowing from node i to node j; Q ij is the reactive power flowing from node i to node j; r ij is the resistance of branch ij; x ijis the reactance of branch ij; I ij is the current flowing from node i to node j;
[0114] After separating the active and reactive power balance equations in (23), the line ij satisfies the following constraints:
[0115]
[0116] is the total active power flowing from node j to downstream node k; is the total reactive power flowing from node j to downstream node k; is the total active power flowing from upstream node i to node j; is the total reactive power flowing from upstream node i to node j; P j is the active power injected into node j; Q j is the reactive power injected into node j;
[0117] Since equation (32) is non-convex, convex relaxation is performed to obtain:
[0118]
[0119] Therefore, the DistFlow model is relaxed into a second-order cone programming model, and the detailed network flow constraints are obtained as follows:
[0120]
[0121] Where P ij,t , Q ij,t are the active power and reactive power flowing from node i to node j at time t respectively; is the total active power flowing from upstream node i to node j at time t; is the total reactive power flowing from upstream node i to node j at time t; is the total active power flowing from node j to downstream node k at time t; is the total reactive power flowing from node j to downstream node k at time t; and are the active and reactive power of the load at node j at time t respectively; is a 0-1 variable, indicating the charging state of the four-quadrant energy storage system at node j at time t; is a 0-1 variable, indicating the discharge state of the four-quadrant energy storage system at node j at time t; They are the computational intermediate quantities of the charging power and discharging power of the four-quadrant energy storage system at node j at time t; is the reactive power of the four-quadrant energy storage system at node j at time t; are the active output and reactive output of the photovoltaic power generation system at node j at time t respectively; are respectively the active output and reactive output of the wind power generation system at node j at time t; are the reactive power outputs of the continuous reactive power compensation device and the group switching capacitor at node j at time t; J ij,t is the square of the absolute value of the current flowing through branch ij at time t; V i,t 、V j,t are the squares of the absolute values of the voltages at nodes i and j at time t respectively;
[0122] Node voltage deviation calculation:
[0123] The second-order cone programming method uses the square of the voltage as the decision variable, and the voltage deviation in the objective function is the difference of the first-order voltage value, which needs to be transformed. It is known that:
[0124] V i,t -V e =U i,t 2 -U e 2 (38)
[0125] |U i,t | 2 -|U e | 2 =(|U i,t |-|U e |)(|U i,t |+|U e |) (39)
[0126] Where U e is the rated voltage; V e is the square of the absolute value of the rated voltage; U i,t is the voltage value of node i at time t;
[0127] Because the operation uses per-unit values, U e =1, combined with the constraints of safe operation regulations, the voltage deviation is less than 10%, and |U i,t |+|U e |=2, then (|U i,t |-|U e |) is converted to
[0128] Thus, the node voltage deviation penalty term F v Translates to:
[0129]
[0130] Where N is the total number of nodes; T is the number of hours in a day; μ is the conversion coefficient of the ratio of active power loss to node voltage deviation; χ i,tis the load weight coefficient of node i at time t; c pr,t is the electricity price at time t;
[0131] Big M method transformation of the product of continuous variables and 0-1 variables:
[0132] Formula (17), (18) and (34) contain the form of multiplication of 0-1 variables and continuous variables. The Big M method is used for transformation and a new continuous variable is introduced. and represents the actual charging power and discharging power of the four-quadrant energy storage system at node i at time t, and adds additional constraints as follows:
[0133]
[0134] In the above formula, It is a 0-1 variable, indicating the charging state of the four-quadrant energy storage system; It is a 0-1 variable, indicating the discharge state of the four-quadrant energy storage system; are the computational intermediate quantities of the charging power and discharging power of the four-quadrant energy storage system at node i at time t; are the charging power and discharging power of the four-quadrant energy storage system at node i at time t respectively; M is taken as the upper limit of the rated power of the configured four-quadrant energy storage system;
[0135] Therefore, the nonlinearity in equations (17) and (18) and Replace with continuous variable and Similarly, the nonlinear and Replace with continuous variable and as follows:
[0136]
[0137] In the formula, Configure the rated power of the four-quadrant energy storage system for node i; Configure the rated capacity of the four-quadrant energy storage system for node i; are the reactive power output and residual power of the four-quadrant energy storage system at node i at time t; min and λ max are the minimum and maximum state of charge coefficients, respectively; are the power of the four-quadrant energy storage system at node i at time t+1 and time t respectively; σ is the charging and discharging efficiency of the four-quadrant energy storage system; Δt is an operating period;
[0138] Circular Constrained Linearization:
[0139] Combine the first two equations of equation (45) into:
[0140]
[0141] In the formula, are the charging power, discharging power and reactive output of the four-quadrant energy storage system at node i at time t; Configure the rated power of the four-quadrant energy storage system for node i;
[0142] Furthermore, the circular constraint in equation (48) is linearized into a regular dodecagonal constraint:
[0143]
[0144] In the formula, a h ,b h ,c h is the parameter vector in the expression of twelve straight lines, where h = 1…12, representing twelve curves.
[0145] Here are some specific examples:
[0146] The present invention provides a four-quadrant energy storage system optimization configuration method that takes into account the grid regulation requirements, and uses the improved IEEE 33-node radial distribution network model as an example for simulation verification. The total system load is (3.715+j2.3)MVA, the rated voltage is 12.66kV, and the upper and lower limits of the node voltage are set to ±5%. Nodes 6, 14, and 19 are each equipped with a photovoltaic power generation system with a capacity of 300kW, and node 3 is equipped with a wind power generation system with a capacity of 500kW. Set the initial rotor speed of the wind turbine to 0.8, the inherent inertia time constant of the wind turbine to 4, and the system capacity to 4MVA;
[0147] Due to the similarity of the regions, each photovoltaic system also uses the same photovoltaic output prediction curve. The typical output curves of load, wind power and photovoltaic power stations in a certain area of Shandong Province in summer are selected as the load curve and the maximum output curve of wind power and photovoltaic power in the system, such as Figure 2 As shown;
[0148] The battery energy storage system (BESS) has high energy density and fast charging and discharging capabilities, meeting the energy storage needs of different scales. The present invention uses lithium battery BESS as the example, and the converter uses the current domestic market price. The specific parameters are: k S 1000 yuan / kVA, k E 850 yuan / (kW·h), k m is 0.04, r is 0.1. Set the maximum number of BESSs that can be installed U to 2, and the maximum power limit of BESS at each node S iBESS,max All are 0.1 (per unit value), the maximum rated capacity limit is E i BESS,ma All are 0.3 (standard per unit value), the life y is 10 years, and the efficiency is 90%. The electricity price uses the peak and valley time-of-use electricity price for urban and rural residents of State Grid Shandong Electric Power Company from September to November 2023, as shown in Table 1;
[0149] Table 1 Time-of-use electricity price
[0150]
[0151] In order to reflect the role of configuring four-quadrant energy storage on the distribution network, three scenarios are set: no BESS; BESS is configured but only its active power is considered; BESS is configured and the four-quadrant capacity is considered. After calculation, the BESS configuration results are shown in Table 2 and the configuration effect is shown in Table 3.
[0152] Table 2 Configuration results
[0153]
[0154] Table 3 Configuration effects of scenarios 1-3
[0155]
[0156] By comparison, it can be seen that the BESS configuration considering the four-quadrant capacity of energy storage, that is, scenario 3, has the lowest comprehensive cost. Although the BESS configuration cost of scenario 3 is higher, its comprehensive cost is slightly lower than that of scenario 2 by low charging and high discharging, reducing voltage deviation and reducing network loss. Figure 3 As shown. It can be seen that after the distribution network is equipped with a four-quadrant energy storage system, when the demand is met, a large amount of electricity is purchased for storage during the period of low electricity prices, and the purchase of electricity is reduced accordingly during the period of relatively high electricity prices. At the same time, the stored electricity is released, thereby reducing the cost of purchasing electricity and improving economic efficiency.
[0157] The output and power of the BESS configured in scenario 3 are as follows: Figure 4a , Figure 4b As shown, its charging and discharging trend is the same as the electricity purchasing trend in Figure 4. The charging period is mostly concentrated in the low-price electricity purchasing period, while the discharging period is mostly concentrated in the high-price electricity purchasing period.
[0158] Compared with Scenario 1, Scenario 2 and Scenario 3, active power loss is significantly reduced, because BESS stores excess wind and solar power output during low power consumption and releases it during peak power consumption, especially avoiding reliance on root node feeders to remotely supply power to the end of the distribution network, which results in large network losses. The advantages of Scenario 3 over Scenario 2 are mainly reflected in improving reactive power flow and increasing the starting voltage value of the branch, so it has lower network losses than Scenario 2.
[0159] Configuring BESS can provide additional virtual inertia for the system and improve the system's ability to resist power disturbances. According to formula (25), the distribution network is analyzed. If the system is required to face a disturbance power of 0.1 (pu), the maximum frequency change rate does not exceed 1Hz / s. At this time, the minimum inertia requirement of the system is 2.5s. However, the wind turbine in scenario 1 can provide a maximum of 2s of virtual inertia, which cannot meet the system requirements. The distribution network is difficult to complete frequency regulation independently and must rely on the inertia of the upper power grid synchronous unit, which makes it pay an additional frequency regulation inertia penalty. For scenarios 2 and 3, the wind turbine can also provide 2s of virtual inertia. In addition to relying on the inertia of the upper power grid synchronous unit, BESS can also be configured to make up for the inertia gap of 0.5s. It can be seen from the formula that the system requires at least 0.5MW or 0.5MVA of BESS rated power. According to the configuration results, the BESS rated power configuration result of scenario 2 is greatly affected by the virtual inertia constraint, and the total configuration rated power of BESS is exactly 0.5MW. The BESS rated power configuration result of scenario 3 is less affected by the virtual inertia constraint. It can be seen that the BESS considering the four-quadrant power capability can better meet the economic and other grid requirements while exceeding the frequency regulation requirements.
[0160] Regarding the impact of BESS on the distribution network voltage, it can be seen in Table 3 that the average voltage deviation of scenario 1 and scenario 2 The difference is not big. In fact, in order to reduce the power purchase cost, scenario 2 slightly sacrifices the voltage quality, and has a slightly higher average voltage deviation than scenario 1. In scenario 3, with lower power purchase cost and comprehensive cost, the system average voltage deviation is reduced by 19.88% compared with scenario 1. It is not intuitive to only use the system average voltage deviation. Draw the average voltage radar chart of each node in the 33-node system as shown below: Figure 5 shown.
[0161] Correspondingly, at nodes 28-33, scenario 3 has a voltage condition that is significantly better than the other two scenarios, which means that the configured energy storage system makes good use of its four-quadrant power advantage, while taking into account other effects and providing reactive support for the distribution network, reducing the voltage drop caused by reactive power flow.
[0162] In summary, through the optimization results of the examples of the present invention, it is shown that the four-quadrant energy storage system optimization configuration method of the present invention that takes into account the grid regulation needs can effectively take into account the grid economy and multiple types of regulation needs, fully tap the potential of the four quadrants of energy storage, especially improve the utilization rate of the power capacity of the configured energy storage system, and increase the investment and construction returns of energy storage.
[0163] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A four-quadrant energy storage system optimization configuration method considering grid regulation requirements, It is characterized in that The steps include: 1) According to the active-reactive regulation characteristics of the four-quadrant energy storage system, the four-quadrant power characteristics of the energy storage system and the way the energy storage system participates in the grid regulation are controlled; 2) Considering the regulation mechanism of the four-quadrant energy storage system in the scenarios of peak load regulation, frequency regulation and voltage regulation of the power grid, a four-quadrant energy storage optimization configuration model considering the multi-type regulation needs of the power grid is established. The objective function of the model is the daily operation cost of the distribution network, taking into account the multi-type regulation needs of the power grid. On the basis of the average daily investment and operation and maintenance cost, the power purchase cost of the distribution network and the network loss cost, the peak load penalty item that comprehensively considers the net load variance and the feeder power variance, the inertia penalty item that considers the system frequency regulation inertia demand, and the weighted node voltage deviation penalty item are added to the objective function; the constraints of the model are: considering the site selection and capacity constraints of the four-quadrant energy storage system, the constraints of the four-quadrant energy storage system, the constraints of the wind power and photovoltaic power generation system, the network flow constraints, the reactive equipment constraints, the virtual inertia constraints, and the safe operation constraints; 3) Use second-order cone relaxation, variable substitution, big M method and piecewise linearization method to deal with nonlinear terms and non-cone functions in the model, and then transform the difficult-to-solve model into a mixed integer second-order cone programming model.
2. According to claim 1, a four-quadrant energy storage system optimization configuration method considering grid regulation requirements, It is characterized in that The objective function of establishing the four-quadrant energy storage optimization configuration considering the multi-type regulation requirements of the power grid described in step 2) is as follows: min F=F BESS +(F elec +F loss )+F p +F H +F v (1) Where, F is the daily operation cost of the distribution network; F BESS is the daily investment and operation and maintenance cost of the four-quadrant energy storage system; F elec F is the cost of purchasing electricity from the distribution network; loss is the network loss cost; F p is the peak load penalty term that comprehensively considers the net load variance and feeder power variance; F H F is the inertia penalty term considering the system frequency regulation inertia requirement; v is the weighted node voltage deviation penalty term; where, Daily investment and operation and maintenance cost of four-quadrant energy storage system F BESS : In the formula, f inv is the investment and construction cost of the four-quadrant energy storage system; f m is the annual operation and maintenance cost of the four-quadrant energy storage system; C is the annual coefficient; the annual investment cost of energy storage is converted to daily calculation, so the denominator is the number of days in a year; where: In the formula, and are the rated power and rated capacity of the four-quadrant energy storage system at node i; k S and k E are the unit rated power price and unit rated capacity price of the four-quadrant energy storage system respectively; k m is the operation and maintenance cost rate of the four-quadrant energy storage system; r is the discount rate, which is 0.1; y is the life of the energy storage system in years; Distribution network power purchase cost F elec : In the formula, c pr,t is the electricity price at time t; P 0,t is the power purchased by the root node at time t; T is the number of hours in a day; Network loss cost F loss : In the formula, r ij is the resistance of branch ij; B is the set of nodes i and j that can form a branch; is the square of the current in branch ij at time t; Peak load penalty term F p : The peak load penalty term includes the net load variance D l and root node feeder power variance D im ,in, Net load variance D l : Where P load,t is the grid load at time t; is the wind power output at time t; is the photovoltaic output at time t; is the output of the four-quadrant energy storage system at time t, where discharge is positive and charge is negative; P l,av is the average value of net load; Root node feeder power variance D im : Where P im,t is the root node feeder power at time t; P im,av is the average feeder power; Taking into account the net load variance and feeder power variance, the peak load penalty term is defined as follows: F p =a l D l +a im D im (8) Among them, α l With α im is the peak load penalty coefficient; Inertia penalty term F H : The system needs sufficient inertia support to ensure frequency stability, and when the inertia support is insufficient, the lower-level power grid needs to make inertia demands to the upper-level power grid; F H =C H S g (9) In the formula, C H is the inertia penalty coefficient, S g is the required capacity of the upper synchronous unit; Weighted node voltage deviation penalty term F v : The calculation formula of node voltage deviation DI is as follows: Where U i,t is the voltage value of node i at time t; U e is the rated voltage; N is the total number of nodes; The influence of node voltage deviation on different node loads is different. The larger the load of the node, the more important the voltage stability is. Therefore, the node voltage deviation is corrected according to the load weight, and the load weight coefficient χ at node i at time t is defined as i,t as follows: Where P load,i,t is the load value of node i at time t; is the average load of all nodes at time t; Then we can obtain the weighted node voltage deviation penalty term F v : Where μ is the proportional conversion coefficient of active power loss and node voltage deviation.
3. According to claim 1, a four-quadrant energy storage system optimization configuration method considering grid regulation requirements, It is characterized in that Step 2): Four-quadrant energy storage system site selection and capacity constraints: Where N is the total number of nodes; is a 0-1 variable for the site selection decision of the four-quadrant energy storage system at node i, 1 represents that node i is configured with energy storage, and 0 represents that node i is not configured with energy storage; U is the maximum number of nodes that can be configured with a four-quadrant energy storage system in the distribution network; Configure the rated power of the four-quadrant energy storage system for node i; and Configure the upper and lower limits of the rated power of the four-quadrant energy storage system for each i-node; Configure the rated capacity of the four-quadrant energy storage system for node i; Configure the upper and lower limits of the rated capacity of the four-quadrant energy storage system for node i respectively; Four-quadrant energy storage system constraints: Introducing 0-1 variables and Represents the charging and discharging state of the four-quadrant energy storage system at node i at time t, and satisfies: is 1, When it is 0, the four-quadrant energy storage system is charged; when is 0, When it is 1, the four-quadrant energy storage system discharges; when and When it is 0 at the same time, it means that node i is not configured with a four-quadrant energy storage system or is neither charging nor discharging at time t; The power of the four-quadrant energy storage system should meet the following requirements: In the formula, are the computational intermediate quantities of the charging power and discharging power of the four-quadrant energy storage system at node i at time t; are the reactive power output and residual power of the four-quadrant energy storage system at node i at time t; min and λ max are the minimum and maximum state of charge coefficients, respectively; The relationship between the battery's output power and capacity is as follows: In the formula, are the power of the four-quadrant energy storage system at node i at time t+1 and time t respectively; σ is the charging and discharging efficiency of the four-quadrant energy storage system; Δt is an operating period; In order to ensure the continuity of energy storage, the daily initial power of the four-quadrant energy storage system is specified. End of day power The same, that is: Wind power and photovoltaic power generation system constraints: When the capacity of wind power generation system and photovoltaic power generation system is given, the actual output of each wind power and photovoltaic power source is affected by external factors such as wind speed, temperature, and light. The maximum output fluctuates over time. Considering the reactive output of wind power and photovoltaic, the power factor of the wind power and photovoltaic power generation system is fixed at 0.
95. In the formula, is the photovoltaic active power output of node i at time t; is the upper limit of photovoltaic active output at node i at time t; is the photovoltaic active power output of node i at time t; is the upper limit of photovoltaic active output at node i at time t; is the photovoltaic reactive power output of node i at time t; is the wind power reactive power output at node i at time t; is the photovoltaic reactive power output of node i at time t; is the wind power reactive power output of node i at time t; Network flow constraints: The application line ij needs to satisfy Ohm's law, branch flow and power balance equations, that is, the DistFlow model is used to represent the network flow constraints: U i -U j =z ij I ij (21) Where i and j are node numbers; U i and U j are the voltages of nodes i and j respectively; I ij is the current flowing from node i to node j; z ij is the impedance of branch ij; S ij is the power flowing from node i to node j; is the total power flowing from upstream node i to node j; is the total power flowing from node j to downstream node k; For I ij The conjugate vector of j The power injected into node j; Reactive equipment constraints: The power grid operates normally and is equipped with reactive equipment, including group switching capacitors and continuous reactive compensation devices: In the formula, is the reactive power output of the grouped switched capacitors at node i at time t; is the gear position of node i at time t; The gear compensation amount set for node i; N CB is the upper limit of the gear position; are the upper and lower limits of the reactive power output of the continuous reactive power compensation device of node i at time t respectively; is the reactive power output of the continuous reactive power compensation device of node i at time t; Virtual inertia constraint: Minimum inertia of power system H min The actual inertia H that the power system can provide sys The comparison is an important way to measure the power system's ability to resist frequency fluctuations. Both wind power systems and energy storage systems can provide virtual inertia. For the stable operation of the power system, the total inertia of the power system must be greater than the minimum inertia of the power system. In the formula, ΔP 1 is the disturbance power; f n is the power system frequency; RoCoF max is the maximum frequency change rate; S B is the power system capacity; H g is the inertia of the synchronous machine; S g is the synchronous machine capacity; Virtual inertia provided for the fan; S Wind is the fan capacity; Virtual inertia provided for the four-quadrant energy storage system; S BESS The total power of the four-quadrant energy storage system configured for the full power system; The maximum virtual inertia provided for the fan; ω r0 is the initial speed of the fan rotor; H w is the inherent inertia time constant of the fan; Safe operation constraints: Where U i,t is the voltage value of node i at time t; U i+1,t is the voltage value of node i at time t+1; U min and U max are the lower and upper limits of the node voltage respectively; U e is the rated voltage; I ij,t is the current value of branch ij at time t; I max The maximum current carrying capacity of the line.
4. According to claim 1, a four-quadrant energy storage system optimization configuration method considering grid regulation requirements, It is characterized in that Step 3) includes: Substituting formula (22) into formula (21), we obtain: Where: i, j are node numbers; U i and U j are the voltages at nodes i and j respectively; For U i The conjugate vector of ij is the impedance of branch ij; For S ij The conjugate vector of ij is the power flowing from node i to node j; Multiply both sides of equation (27) by their respective conjugates, and let J ij =|I ij 2 , V i =U i 2 , V j =U j 2 get: In the formula, For z ij The conjugate vector of ; Expanding formula (28), we get: Where P ij is the active power flowing from node i to node j; Q ij is the reactive power flowing from node i to node j; r ij is the resistance of branch ij; x ij is the reactance of branch ij; I ij is the current flowing from node i to node j; After separating the active and reactive power balance equations in (23), the line ij satisfies the following constraints: is the total active power flowing from node j to downstream node k; is the total reactive power flowing from node j to downstream node k; is the total active power flowing from upstream node i to node j; is the total reactive power flowing from upstream node i to node j; P j is the active power injected into node j; Q j is the reactive power injected into node j; Since equation (32) is non-convex, convex relaxation is performed to obtain: Therefore, the DistFlow model is relaxed into a second-order cone programming model, and the detailed network flow constraints are obtained as follows: Where P ij,t , Q ij,t are the active power and reactive power flowing from node i to node j at time t respectively; is the total active power flowing from upstream node i to node j at time t; is the total reactive power flowing from upstream node i to node j at time t; is the total active power flowing from node j to downstream node k at time t; is the total reactive power flowing from node j to downstream node k at time t; and are the active and reactive power of the load at node j at time t respectively; is a 0-1 variable, indicating the charging state of the four-quadrant energy storage system at node j at time t; is a 0-1 variable, indicating the discharge state of the four-quadrant energy storage system at node j at time t; They are the computational intermediate quantities of the charging power and discharging power of the four-quadrant energy storage system at node j at time t; is the reactive power of the four-quadrant energy storage system at node j at time t; are the active output and reactive output of the photovoltaic power generation system at node j at time t respectively; are respectively the active output and reactive output of the wind power generation system at node j at time t; are the reactive power outputs of the continuous reactive power compensation device and the group switching capacitor at node j at time t; J ij,t is the square of the absolute value of the current flowing through branch ij at time t; V i,t 、V j,t are the squares of the absolute values of the voltages at nodes i and j at time t respectively; Node voltage deviation calculation: The second-order cone programming method uses the square of the voltage as the decision variable, and the voltage deviation in the objective function is the difference of the first-order voltage value, which needs to be transformed. It is known that: V i,t -V e =|U i,t | 2 -|U e | 2 (38) |In i,t | 2 -|In e | 2 =(|U i,t |-|U e |)(|U i,t |+|U e |) (39) Where U e is the rated voltage; V e is the square of the absolute value of the rated voltage; U i,t is the voltage value of node i at time t; Because the operation uses per-unit values, U e =1, combined with the constraints of safe operation regulations, the voltage deviation is less than 10%, and |U i,t |+|U e |=2, then (|U i,t |-|U e |) is converted to Thus, the node voltage deviation penalty term F v Translates to: Where N is the total number of nodes; T is the number of hours in a day; μ is the conversion coefficient of the ratio of active power loss to node voltage deviation; χ i,t is the load weight coefficient of node i at time t; c pr,t is the electricity price at time t; Big M method transformation of the product of continuous variables and 0-1 variables: Formula (17), (18) and (34) contain the form of multiplication of 0-1 variables and continuous variables. The Big M method is used for transformation and a new continuous variable is introduced. and represents the actual charging power and discharging power of the four-quadrant energy storage system at node i at time t, and adds additional constraints as follows: In the above formula, It is a 0-1 variable, indicating the charging state of the four-quadrant energy storage system; It is a 0-1 variable, indicating the discharge state of the four-quadrant energy storage system; are the computational intermediate quantities of the charging power and discharging power of the four-quadrant energy storage system at node i at time t; are the charging power and discharging power of the four-quadrant energy storage system at node i at time t respectively; M is taken as the upper limit of the rated power of the configured four-quadrant energy storage system; Therefore, the nonlinearity in equations (17) and (18) and Replace with continuous variable and Similarly, the nonlinear and Replace with continuous variable and as follows: In the formula, Configure the rated power of the four-quadrant energy storage system for node i; Configure the rated capacity of the four-quadrant energy storage system for node i; are the reactive power output and residual power of the four-quadrant energy storage system at node i at time t; min and λ max are the minimum and maximum state of charge coefficients, respectively; are the power of the four-quadrant energy storage system at node i at time t+1 and time t respectively; σ is the charging and discharging efficiency of the four-quadrant energy storage system; Δt is an operating period; Circular Constrained Linearization: Combine the first two equations of equation (45) into: In the formula, are the charging power, discharging power and reactive output of the four-quadrant energy storage system at node i at time t; Configure the rated power of the four-quadrant energy storage system for node i; Furthermore, the circular constraint in equation (48) is linearized into a regular dodecagonal constraint: In the formula, a h ,b h ,c h is the parameter vector in the expression of twelve straight lines, where h = 1…12, representing twelve curves.
Citation Information
Cited By
Power distribution network energy storage optimization configuration method considering dynamic uncertainty and multi-target cooperation
CN121036044A
