Source network load storage clustering joint optimization method and device based on distributed collaboration

Abstract

The invention relates to the technical field of source network load storage, and discloses a source network load storage clustering joint optimization method and device based on distributed coordination, and the method comprises the steps: determining an active power demand and a reactive power demand at a current moment; obtaining the internal resistance of the energy storage unit, the change rate of the internal resistance to the temperature and the heat capacity; determining a positive feedback intensity core based on the internal resistance, the change rate of the internal resistance to the temperature and the heat capacity, and forming a distribution criterion by minimizing the maximum value of the positive feedback intensity core; generating a current instruction according to the distribution criterion meeting the active power demand; and generating a reactive current instruction according to the distribution criterion meeting the reactive power demand. The overall operation life of the energy storage cluster can be prolonged, and the economical efficiency and reliability of the system are remarkably improved on the premise that active and reactive power requirements and apparent power safety constraints are met.

Description

Technical Field

[0001] The present invention relates to the field of source-grid-load-storage technology, and more specifically, to a distributed collaborative source-grid-load-storage clustered joint optimization method and device. Background Art

[0002] With the rapid development of renewable energy power generation and energy storage systems, large-scale battery energy storage is increasingly being used for coordinated optimization of sources, grids, loads, and storage. Parallel battery modules or energy storage clusters, when operated in clusters, can rapidly respond to tasks such as frequency regulation, peak shaving, and voltage support. However, in practical applications, different battery cells often exhibit varying internal resistance and capacity levels due to differences in manufacturing processes, usage history, and heat dissipation environments. In parallel systems, these variations directly determine the bias in current distribution, thereby impacting system stability and lifespan.

[0003] However, in a parallel system, cells with low internal resistance are more likely to share more current, and the current passing through these cells generates greater heat, causing their temperature to rise further. This temperature increase in turn reduces the resistance of the cell, allowing it to continue to carry greater current in subsequent cycles. At the same time, energy storage systems often operate under conditions of frequent shallow cycles and narrow states of charge when frequency modulation or smoothing renewable energy output. This operating characteristic accelerates uneven current distribution. The above operating mechanism causes the degradation rates of different battery cells to gradually widen, resulting in inconsistent performance within the system.

[0004] During long-term operation, the interaction between uneven current distribution and temperature effects can lead to the paradoxical phenomenon of healthy cells failing first. Previously high-performing cells, continuously bearing excessive electrical and thermal stress, reach their lifespans prematurely, while weaker cells, passively carrying less current, experience delayed degradation, ultimately reversing the lifespan ranking of the cluster. This not only shortens the overall lifespan of the energy storage cluster but also reduces the reliability and economic efficiency of the system, becoming a key technical issue that urgently needs to be addressed in the current optimization of source-grid-load-storage integration. Summary of the Invention

[0005] The present invention provides a distributed collaborative source-grid-load-storage clustered joint optimization method and device, which solves the technical problems raised in the background technology.

[0006] In a first aspect, the present invention provides a distributed collaborative source-grid-load-storage clustered joint optimization method, comprising:

[0007] Determine the active power demand and reactive power demand at the current moment;

[0008] Obtain the internal resistance, rate of change of internal resistance with temperature, and heat capacity of the energy storage unit;

[0009] The positive feedback strength kernel is determined based on the internal resistance, the rate of change of the internal resistance to temperature and the heat capacity, and the allocation criterion is formed by minimizing the maximum value of the positive feedback strength kernel;

[0010] Generate current instructions according to the distribution criteria to meet the active power demand;

[0011] Generate reactive current instructions according to the distribution criteria to meet the reactive power demand.

[0012] Furthermore, the internal resistance, the rate of change of the internal resistance with respect to temperature, and the heat capacity of the energy storage unit are obtained, including:

[0013] The current of energy storage unit i is collected at fixed time intervals , terminal voltage and unit temperature ;

[0014] The process of obtaining the internal resistance of energy storage unit i is as follows:

[0015] Calculation of internal resistance by least squares straight line fitting , ;

[0016] in, represents the number of fitting samples, represents the average current of the fitted sample, represents the average terminal voltage of the fitted sample;

[0017] The process of obtaining the rate of change of the internal resistance of energy storage unit i with respect to temperature is as follows:

[0018] The current moment The corresponding internal resistance and unit temperature are combined to obtain: ;

[0019] Calculate the rate of change of internal resistance with temperature by least squares straight line fitting ,as follows:

[0020]

[0021] in, represents the average cell temperature of the fitted sample, represents the average internal resistance of the fitted samples;

[0022] The process of obtaining the heat capacity of energy storage unit i is as follows:

[0023] At the moment To the current moment Between , the current of energy storage unit i is raised to calculate the thermal capacity as follows:

[0024]

[0025]

[0026] in, represents the heat capacity of energy storage unit i, Indicates the heat generation power, Indicates a fixed time interval.

[0027] Furthermore, the positive feedback strength kernel is determined based on the internal resistance, the rate of change of the internal resistance with respect to temperature, and the heat capacity, including:

[0028]

[0029] in, represents the nuclear coefficient of energy storage unit i;

[0030] According to the current Calculates power mapping functions, including:

[0031]

[0032] The product of the power mapping function and the kernel coefficient is used as the positive feedback intensity kernel .

[0033] Furthermore, the allocation criteria are formed based on the maximum and minimum values ​​of the positive feedback intensity kernel, including:

[0034] The assignment criteria expressed in terms of infinity norm include: ;

[0035] in, represents the allocation criterion function, represents the positive feedback intensity kernel of the i-th energy storage unit, represents the intensity kernel vector formed by the positive feedback intensity kernel of N energy storage units, represents the infinity norm.

[0036] Furthermore, generating a current instruction to meet the active power demand according to the allocation criteria includes:

[0037] Synchronously measure the current increment and power increment of energy storage unit i; the current increment and power increment represent the current moment and time The current difference and power difference;

[0038] The ratio of power increment to current increment is used as the active conversion coefficient of energy storage unit i;

[0039] The product of the core coefficient, internal resistance and fixed time interval of energy storage unit i is used as the first parameter ;

[0040] Use the active power conversion factor as the second parameter .

[0041] Furthermore, generating a current instruction to meet the active power demand according to the allocation criterion also includes:

[0042] The first optimization model is constructed based on the first parameter and the second parameter as follows:

[0043]

[0044] in, Indicates the active power demand, N indicates the number of energy storage units, Represents auxiliary variables;

[0045] Obtain the closed-form solution of the first optimization model, including:

[0046]

[0047] in, represents the i-th component of the optimal current vector obtained by solving the optimization model;

[0048] and will As the current instruction of energy storage unit i .

[0049] Furthermore, generating a reactive current instruction to meet the reactive power demand according to the allocation criterion includes:

[0050] Synchronously measure the reactive current increment and reactive power increment of energy storage unit i; wherein, the reactive current increment and reactive power increment represent the current moment and time The reactive current difference and reactive power difference;

[0051] The ratio of reactive power increment to reactive current increment is used as reactive conversion factor ;

[0052] Read the AC terminal voltage of energy storage unit i and the corresponding apparent power upper limit , to construct apparent power constraints, including:

[0053]

[0054] in, represents the reactive current to be solved on the AC side of energy storage unit i.

[0055] Furthermore, generating a reactive current instruction to meet the reactive power demand according to the allocation criterion also includes:

[0056] At the moment To the current moment Between, the current of energy storage unit i is raised and the AC / DC equivalent coefficient is calculated. , ;in, represents the perturbation difference operator, represents the DC side current of the i-th energy storage unit, represents the active component current on the AC side of the i-th energy storage unit;

[0057] The product of the AC / DC equivalent coefficient and the first parameter is used as the AC side core constant ;

[0058] The second optimization model is constructed based on the AC side core constant and reactive power conversion coefficient as follows:

[0059]

[0060] in, represents the reactive power demand, represents the upper bound variable;

[0061] If the apparent power constraint of energy storage unit i is not triggered, the reactive current to be solved is calculated, including:

[0062]

[0063] and will As the current instruction of energy storage unit i ;

[0064] If the apparent power constraint of energy storage unit i is triggered, the cutoff value of reactive current calculation is to be solved. ; Determine the residual reactive power demand based on the cutoff value , repeatedly solve the remaining reactive power demand on the untriggered set, and finally get the reactive current instruction ; Among them, the remaining reactive power demand The calculation formula is as follows:

[0065]

[0066] in, Indicates the set of energy storage units that trigger the apparent power constraint.

[0067] In a second aspect, a distributed collaborative source-grid-load-storage clustered joint optimization device is applied to any of the distributed collaborative source-grid-load-storage clustered joint optimization methods described above, including:

[0068] Data acquisition module, used to determine the current active power demand and reactive power demand;

[0069] A data acquisition module is used to obtain the internal resistance of the energy storage unit, the rate of change of the internal resistance with respect to temperature, and the thermal capacity;

[0070] An allocation criterion module is used to determine a positive feedback strength kernel based on internal resistance, a rate of change of internal resistance with respect to temperature, and heat capacity, and to form an allocation criterion by minimizing the maximum value of the positive feedback strength kernel;

[0071] Active current module, used to generate current instructions according to the distribution criteria to meet the active power demand;

[0072] The reactive current module is used to generate reactive current instructions according to the distribution criteria to meet the reactive power demand.

[0073] The beneficial effects of this invention include: By constructing a positive feedback strength core based on internal resistance, temperature sensitivity, and heat capacity, a unified measure of the tendency of energy storage units to change in admittance due to temperature rise under current carrying is established; and further establishing distribution criteria. This prevents healthy units from being continuously overloaded due to low resistance, thus preventing their accelerated degradation, and achieving balanced load distribution and synchronized degradation across units. The ultimate effect is to reliably extend the overall operating life of the energy storage cluster, while significantly improving the system's economic efficiency and reliability while meeting active and reactive power requirements and apparent power safety constraints. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 is a flow chart of the method of the present invention;

[0075] Figure 2 It is a device module diagram of the present invention. DETAILED DESCRIPTION

[0076] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed solely to enable those skilled in the art to better understand and implement the subject matter described herein, and that the functions and arrangements of the elements discussed may be varied without departing from the scope of this specification. Various examples may omit, substitute, or add various processes or components as needed. In addition, features described with respect to some examples may also be combined in other examples.

[0077] Example 1:

[0078] like Figure 1 As shown in the figure, the distributed collaborative source-grid-load-storage clustered joint optimization method includes:

[0079] Determine the active power demand and reactive power demand at the current moment;

[0080] Obtain the internal resistance, rate of change of internal resistance with temperature, and heat capacity of the energy storage unit;

[0081] The positive feedback strength kernel is determined based on the internal resistance, the rate of change of the internal resistance to temperature and the heat capacity, and the allocation criterion is formed by minimizing the maximum value of the positive feedback strength kernel;

[0082] Generate current instructions according to the distribution criteria to meet the active power demand;

[0083] Generate reactive current instructions according to the distribution criteria to meet the reactive power demand.

[0084] It should be noted that active power demand and reactive power demand are obtained based on a prediction model. For example, through an LSTM neural network model, time series active power demand and reactive power demand are obtained as training data, thereby obtaining an LSTM neural network model capable of predicting active power demand and reactive power demand.

[0085] It should be noted that thermal resistance shunting refers to the dynamic coupling process in which the current distribution of parallel cells in an energy storage cluster is uneven due to differences in resistance. This uneven distribution is further amplified by temperature fluctuations. Low-resistance cells naturally carry more current, generating more heat and causing the temperature to rise. If the resistance decreases with rising temperature, the resistance of the energy storage cell decreases further, allowing it to receive more current in the next cycle. This creates a positive feedback loop: increased current → increased heat generation → increased temperature → decreased resistance → increased current. This positive feedback loop can cause initially healthy energy storage cells to prematurely age due to continuous overload.

[0086] In one embodiment of the present invention, obtaining the internal resistance, the rate of change of the internal resistance with respect to temperature, and the heat capacity of the energy storage unit includes:

[0087] The current of energy storage unit i is collected at fixed time intervals , terminal voltage and unit temperature ;

[0088] The process of obtaining the internal resistance of energy storage unit i is as follows:

[0089] Calculation of internal resistance by least squares straight line fitting , ;

[0090] in, represents the number of fitting samples, represents the average current of the fitted sample, represents the average terminal voltage of the fitted sample;

[0091] Specifically, based on a linear circuit model, the terminal voltage and current satisfy a linear relationship, where is the open-circuit voltage and is the internal resistance. Least squares fitting minimizes fitting error and improves resistance estimation accuracy by calculating the sum of the products of the current and voltage deviations for a set of sample data and dividing this sum by the sum of the squares of the current deviations. The number of fitting samples selected balances computational efficiency and accuracy. Average current and average terminal voltage are used as intermediate quantities to eliminate the effects of DC bias in the data, ensuring that the fitting results more closely resemble the true resistance characteristics.

[0092] The process of obtaining the rate of change of the internal resistance of energy storage unit i with respect to temperature is as follows:

[0093] The current moment The corresponding internal resistance and unit temperature are combined to obtain: ;

[0094] Calculate the rate of change of internal resistance with temperature by least squares straight line fitting ,as follows:

[0095]

[0096] in, represents the average cell temperature of the fitted sample, represents the average internal resistance of the fitted samples;

[0097] Specifically, within the local temperature range, resistance and temperature approximately follow a linear relationship, where the intercept is the rate of change of resistance with temperature. During the fitting process, the resistance variation with temperature is quantified by calculating the product of the temperature deviation and the resistance deviation, and dividing it by the sum of the squares of the temperature deviations. The average cell temperature and average internal resistance are used to smooth out data fluctuations, making the rate of change calculation more accurate.

[0098] The process of obtaining the heat capacity of energy storage unit i is as follows:

[0099] At the moment To the current moment Between , the current of energy storage unit i is raised to calculate the thermal capacity as follows:

[0100]

[0101]

[0102] in, represents the heat capacity of energy storage unit i, Indicates the heat generation power, Indicates a fixed time interval.

[0103] Specifically, the current is increased to enhance the thermal signal and make the temperature change more significant, allowing for accurate calculation of the thermal capacitance.

[0104] Detailed, at the moment To the current moment By increasing the current between the resistors, the unit generates a measurable heat change, which is then used to calculate the heat capacity. The calculation of heat generation power is based on Joule's law, which states that the power generated by a current passing through a resistor is proportional to the square of the current and the resistance. The heat capacity calculation formula is based on the principle of thermal equilibrium, which states that the ratio of heat generation power to the rate of temperature change reflects the unit's ability to absorb heat. The fixed time interval ensures a consistent time base for the temperature change calculation, avoiding heat capacity errors caused by different time scales.

[0105] It should be noted that the rate of change of the internal resistance with respect to temperature is used to reflect the trend of the internal resistance of the energy storage unit changing with temperature, and to reflect the mapping relationship between temperature change and resistance change.

[0106] In one embodiment of the present invention, determining the positive feedback strength kernel based on the internal resistance, the rate of change of the internal resistance with respect to temperature, and the heat capacity includes:

[0107]

[0108] in, represents the nuclear coefficient of energy storage unit i;

[0109] In detail, the calculation formula of the kernel coefficient is By integrating the internal resistance of the energy storage unit, the rate of change of the internal resistance to temperature and the heat capacity, a positive feedback potential is formed. The rate of change of the internal resistance to temperature represents the effect of temperature on resistance: when When it is less than 0, the temperature rise will lead to a decrease in resistance, which is the core driving factor of the positive feedback of thermal resistance shunt. The negative sign is introduced in the formula to When it is less than 0, the kernel coefficient is positive, thus representing a positive feedback trend.

[0110] Specifically, internal resistance is the basis for heat generation by current. The greater the resistance, the more heat generated at the same current. Heat capacity reflects the ease with which a unit's temperature changes after absorbing heat. The smaller the heat capacity, the greater the temperature rise for the same amount of heat, and the greater the likelihood of exacerbating positive feedback. The internal resistance, its rate of change with temperature, and heat capacity are coupled together to make the core coefficient a quantitative indicator of the admittance amplification capability per unit of heat injection.

[0111] According to the current Calculates power mapping functions, including:

[0112]

[0113] In detail, the Joule heat generated by current passing through a resistor follows the Joule heat formula, where is the heat production per unit time, multiplied by the fixed time interval back, To reflect the current moment The total heat generated by internal current. Since the core of positive feedback is the coupling of current, thermal capacitance, and internal resistance, the impact of temperature changes on resistance cannot be correlated through current or power alone. However, after conversion to energy, it can be correlated with the core coefficient. The electric energy mapping function is used to convert current into heat energy.

[0114] The product of the power mapping function and the kernel coefficient is used as the positive feedback intensity kernel .

[0115] In detail, the positive feedback strength kernel is used to quantify the real-time strength of positive feedback.

[0116] When the unit is Temperature zone less than 0 (most operating scenarios), For positive, is a non-negative value (current heat generation is always positive), so the positive feedback strength core is positive. The larger the positive feedback strength core, the stronger the positive feedback cycle of the energy storage unit under the current: temperature rises due to heat generation → resistance decreases → shunt increases. Greater than or equal to 0 (resistance remains unchanged or increases when temperature rises), The positive feedback intensity kernel is non-positive, indicating that the positive feedback is weakened or there is no significant positive feedback.

[0117] In one embodiment of the present invention, the allocation criterion is formed by minimizing the maximum value of the positive feedback intensity kernel, including:

[0118] The assignment criteria expressed in terms of infinity norm include: ;

[0119] in, represents the allocation criterion function, represents the positive feedback intensity kernel of the i-th energy storage unit, represents the intensity kernel vector formed by the positive feedback intensity kernel of N energy storage units, represents the infinity norm.

[0120] Specifically, the maximum and minimum values ​​of the positive feedback intensity kernel are used to form the allocation criterion, thereby constructing an index for quantifying the positive feedback intensity, and optimizing the current allocation with this index as the target. Its mathematical expression is: Positive feedback strength kernel By the kernel coefficient and power mapping function The kernel coefficient integrates the unit's resistance temperature sensitivity, inherent resistance, and thermal capacitance characteristics, reflecting the unit's inherent positive feedback potential. The energy mapping function correlates current magnitude with duration, quantifying the degree of thermal disturbance to the unit under current conditions. The combination of the kernel coefficient and the energy mapping function describes the strength of the positive feedback loop corresponding to current generating heat, resistance change, and increased current diversion.

[0121] Detailed, intensity kernel vector It is composed of the positive feedback strength cores of all energy storage units, and includes the positive feedback characteristics of each unit under the current. The role of the infinity norm is that the positive feedback of the thermal resistance shunt has a nonlinear amplification characteristic, that is, the larger the positive feedback strength core of a certain energy storage unit, the more obvious the resistance reduction under the same current increment, the more shunts, and the stronger the positive feedback loop formed. Since the harm of positive feedback is mainly determined by the most sensitive unit, that is, if the positive feedback of a certain energy storage unit is too strong, it will first enter the vicious cycle of thermal resistance shunt, and eventually cause the energy storage unit to fail prematurely. Therefore, taking the maximum value as the optimization target can directly control the most dangerous link. The allocation criterion is used for the most significant positive feedback intensity of all energy storage units under the current current distribution. By optimizing Minimize the positive feedback so that the strongest positive feedback is as weak as possible, thereby balancing the burden of each energy storage unit and avoiding the accelerated degradation of a single energy storage unit due to excessive positive feedback.

[0122] Specifically, the allocation criterion provides a clear goal for current distribution: while meeting the total power demand, adjust the current of each unit to minimize the allocation criterion. Specifically, when the positive feedback strength kernel of a storage unit is large, its current is reduced, thereby lowering the power mapping function (because the power mapping function is proportional to the square of the current), thereby reducing the positive feedback strength kernel. For storage units with smaller positive feedback strength kernels, the current can be appropriately increased to ensure that the total power demand is met. Ultimately, the positive feedback strength kernels of all storage units are balanced, and the maximum value is minimized.

[0123] In one embodiment of the present invention, generating a current instruction to meet active power demand according to an allocation criterion includes:

[0124] Synchronously measure the current increment and power increment of energy storage unit i; the current increment and power increment represent the current moment and time The current difference and power difference;

[0125] In detail, by synchronously measuring the current increment and the power increment, the mapping relationship between current and power is established by capturing the changes in electrical parameters within a fixed time.

[0126] The ratio of power increment to current increment is used as the active conversion coefficient of energy storage unit i;

[0127] Specifically, the active power conversion factor quantifies the change in power caused by a unit change in current, reflecting the energy storage unit's current-to-power conversion efficiency under current operating conditions. For example, if the power increment is 50 watts when the current increases by 1 ampere, the active power conversion factor is 50, indicating that the energy storage unit has a strong current-to-power drive capability.

[0128] The product of the core coefficient, internal resistance and fixed time interval of energy storage unit i is used as the first parameter ;

[0129] In detail, the calculation formula of the first parameter is the product of the nuclear coefficient, the internal resistance and the fixed time interval.

[0130] The first parameter combines the inherent characteristics of the energy storage unit (resistance, resistance temperature sensitivity, and heat capacity) with the time factor to obtain a comprehensive scale that reflects the ability of the admittance to change under the action of unit energy, thereby providing an energy dimension benchmark for subsequent current distribution and ensuring that the characteristics of different energy storage units are comparable.

[0131] Use the active power conversion factor as the second parameter .

[0132] In detail, the second parameter is the active power conversion factor, which relates current and power and provides conversion from current command to power output.

[0133] In one embodiment of the present invention, generating a current instruction according to a distribution criterion to satisfy an active power demand further includes:

[0134] The first optimization model is constructed based on the first parameter and the second parameter as follows:

[0135]

[0136] in, Indicates the active power demand, N indicates the number of energy storage units, Represents auxiliary variables;

[0137] In detail, the core of the first optimization model is to “minimize the maximum value of positive feedback intensity” as the goal, while meeting the total active power demand and balancing the positive feedback trends of each energy storage unit.

[0138] The objective function is , which represents minimizing the auxiliary variable ζ, where ζ reflects the maximum allowable value of the positive feedback intensity of all energy storage units.

[0139] :in is the first parameter, is the current flowing through energy storage unit i. This constraint ensures that the positive feedback strength of each energy storage unit at the current current (given by quantification) must not exceed ζ to ensure that no unit falls into a vicious cycle of thermal resistance shunting due to excessive positive feedback.

[0140] :in is the second parameter, is the total active power demand. This constraint ensures that the total active power output of all energy storage units meets the system demand, reflecting the combination of optimization objectives and actual power balance.

[0141] In detail, N is the number of energy storage units, and its value is determined by the cluster size; Integrates the unit's resistance temperature sensitivity, inherent resistance and thermal capacitance characteristics. The conversion relationship between current and active power is related. and Together they ensure that the first optimization model incorporates both the inherent characteristics of the unit and the operating conditions.

[0142] Obtain the closed-form solution of the first optimization model, including:

[0143]

[0144] in, represents the i-th component of the optimal current vector obtained by solving the optimization model;

[0145] The detailed closed-form solution is calculated as , its derivation logic is as follows:

[0146] 1. In the ideal case without hard constraints on the equipment (such as current limit and temperature limit), the optimal solution must satisfy the equal positive feedback strength of each unit, that is, ( is the optimal auxiliary variable). Due to the characteristic of the objective function of minimizing the maximum value, the maximum value can only be minimized when the positive feedback intensity of all energy storage units reaches the same level.

[0147] 2. Power balance constraint: Substitute the total power constraint , we can get . Solve after sorting ,get .

[0148] 3. Substitution , that is, to obtain the optimal current of each unit Thus, the optimal current of the energy storage unit i ensures the balance of positive feedback intensity and meets the total power demand.

[0149] and will As the current instruction of energy storage unit i .

[0150] In detail, the optimal current component Directly used as the current instruction of energy storage unit i .

[0151] Based on power requirements and unit inherent characteristics and The joint decision not only reflects the system's demand for active power, but also takes into account the unit's ability to suppress positive feedback, ensuring that the command matches the unit status.

[0152] Consistency with the goal of suppressing positive feedback: current command The size and Inversely proportional, that is The larger it is, the stronger the positive feedback potential of the energy storage unit is and the smaller the current command is, which reflects the optimization goal of suppressing strong positive feedback units and preventing such energy storage units from exacerbating the positive feedback loop due to excessive current.

[0153] In one embodiment of the present invention, generating a reactive current instruction to meet reactive power demand according to an allocation criterion includes:

[0154] Synchronously measure the reactive current increment and reactive power increment of energy storage unit i; wherein, the reactive current increment and reactive power increment represent the current moment and time The reactive current difference and reactive power difference;

[0155] The ratio of reactive power increment to reactive current increment is used as reactive conversion factor ;

[0156] Specifically, the reactive power conversion factor is the ratio of the reactive power increment to the reactive current increment. It reflects the reactive power increment corresponding to a unit reactive current increment and is used to convert reactive current into reactive power. In AC circuits, the relationship between reactive power and reactive current can be approximately expressed as: reactive power is directly proportional to reactive current. The reactive power conversion factor can eliminate interference such as voltage fluctuations, resulting in a linearized conversion relationship.

[0157] Read the AC terminal voltage of energy storage unit i and the corresponding apparent power upper limit , to construct apparent power constraints, including:

[0158]

[0159] in, represents the reactive current to be solved on the AC side of energy storage unit i.

[0160] In detail, the apparent power upper limit is a device rating parameter that reflects the maximum capacity limit of the energy storage unit.

[0161] Specifically, the apparent power generated by the active and reactive currents must not exceed the upper apparent power limit to prevent equipment overload due to excessive total current. The squared term reflects the orthogonal relationship between active and reactive currents, which conforms to the physical laws of apparent power in AC circuits.

[0162] In one embodiment of the present invention, generating a reactive current instruction according to a distribution criterion to satisfy reactive power demand further includes:

[0163] At the moment To the current moment Between, the current of energy storage unit i is raised and the AC / DC equivalent coefficient is calculated. , ;in, represents the perturbation difference operator, represents the DC side current of the i-th energy storage unit, represents the active component current on the AC side of the i-th energy storage unit;

[0164] In detail, in order to equate the impact of the reactive current on the system positive feedback to the same measurement system as the active side, it is necessary to calculate the AC / DC equivalent coefficient . It is achieved by perturbed difference: at adjacent moments (from moment To the current moment ) After the current of energy storage unit i is disturbed by a lifting force, the ratio of the change in the square of the DC side current to the change in the sum of the squares of the AC side currents is quantified. The formula is: ; among them, among them, is the perturbation difference operator, is the DC side current of energy storage unit i, The AC-DC equivalent coefficient reflects the equivalent effect of changes in the AC-side current on positive feedback-related characteristics such as DC-side heating.

[0165] It should be noted that the process of obtaining the perturbation difference operator is as follows:

[0166] Select a base time , synchronously collect three types of current parameters of energy storage unit i:

[0167] DC side current: ;

[0168] AC side active current: ;

[0169] Reactive current on AC side: .

[0170] At the moment To the current moment In the example, a lifting disturbance is applied to the AC side current of the energy storage unit i, for example:

[0171] Only increase the active current: from Increase to ;

[0172] Only increase the reactive current: from Increase to ;

[0173] At the same time, the active and reactive currents are raised: Increase to 、 .

[0174] At the current moment , and synchronously collect the three types of current parameters of energy storage unit i again:

[0175] DC side current: ;

[0176] AC side active current: ;

[0177] Reactive current on AC side: .

[0178] The perturbation difference operator is used to quantify the change in the physical quantity before and after the disturbance, and needs to calculate:

[0179] 1. Change in the square of DC side current:

[0180] Positive feedback is directly related to the square of the current (corresponding to positive feedback drivers such as power and heat), so we focus on the change in the square of the DC side current: ;

[0181] 2. Change in the sum of squares of active and reactive currents on the AC side:

[0182] The total current effect on the AC side is determined by both active and reactive currents (according to the physical laws of apparent current). Therefore, we focus on the change in the sum of the squares of active and reactive currents on the AC side: ;

[0183] The disturbance difference operator Δ quantifies the influence of the AC side current change on the DC side positive feedback related characteristics as the AC-DC equivalent coefficient by calculating the ratio of the change amount. ), the formula is: .

[0184] The product of the AC / DC equivalent coefficient and the first parameter is used as the AC side core constant ;

[0185] Specifically, the AC / DC equivalent coefficient is multiplied by the first parameter to obtain the AC side core constant: The AC side core constant represents the positive feedback potential coefficient after integrating the AC and DC characteristics, so that the impact of reactive current on positive feedback can be quantified in a kernel form consistent with the active current, ensuring that reactive power distribution can continue to use the allocation criterion of minimizing the maximum positive feedback intensity.

[0186] The second optimization model is constructed based on the AC side core constant and reactive power conversion coefficient as follows:

[0187]

[0188] in, represents the reactive power demand, represents the upper bound variable;

[0189] In detail, based on the AC / DC equivalent coefficient and the reactive power conversion coefficient, a second optimization model is constructed with the goal of minimizing the maximum positive feedback intensity:

[0190]

[0191] Among them, the objective function middle, is the upper bound variable, which represents the maximum allowable value of the positive feedback intensity on the AC side of all energy storage units; , combining the fixed influence of the active current command with the variable influence of the reactive current to be calculated, quantifying the contribution of the total current on the AC side to the positive feedback, requiring the positive feedback intensity of all units to not exceed ;constraint , ensuring the total reactive power demand Be satisfied.

[0192] If the apparent power constraint of energy storage unit i is not triggered, the reactive current to be solved is calculated, including:

[0193]

[0194] and will As the current instruction of energy storage unit i ;

[0195] In detail, if the apparent power constraint of energy storage unit i is (in is the AC terminal voltage, is the apparent power upper limit) is not triggered (i.e., the optimization solution will not cause equipment overload), then the optimal solution satisfies the equal AC side positive feedback intensity of all units (optimality condition of minimizing the maximum value), that is:

[0196] in, is the optimal upper bound;

[0197] Combined with total reactive power constraint , the combined deduction can be obtained:

[0198]

[0199] At this time, It is directly used as the reactive current instruction of energy storage unit i, which not only meets the total reactive power demand, but also ensures that the positive feedback intensity of each energy storage unit on the AC side is balanced and minimized.

[0200] Specifically, if the apparent power constraint of a certain energy storage unit i is triggered (i.e. ), its reactive current needs to be cut off and the cut-off value needs to be calculated. (Usually the apparent power constraint equation is established, that is, , making sure not to exceed the limit).

[0201] Then, define the set of energy storage units that trigger the apparent power constraint ), calculate the sum of the reactive powers provided by these units through the cutoff value , and deducted from the total reactive power demand to obtain the residual reactive power demand: .

[0202] If the apparent power constraint of energy storage unit i is triggered, the cutoff value of reactive current calculation is to be solved. ; Determine the residual reactive power demand based on the cutoff value , repeatedly solve the remaining reactive power demand on the untriggered set, and finally get the reactive current instruction ; Among them, the remaining reactive power demand The calculation formula is as follows:

[0203]

[0204] in, Indicates the set of energy storage units that trigger the apparent power constraint.

[0205] In detail, in the set of cells that do not trigger the constraint (non part), with For the new reactive power demand, repeat the above second optimization model solving process until all units meet the apparent power constraint, and finally obtain the reactive current instruction of each energy storage unit (The cells that trigger the constraint take the cutoff value , the untriggered ones use the original values ​​from the re-solve).

[0206] It should be noted that the cutoff value The acquisition process is as follows:

[0207] When the apparent power constraint of energy storage unit i is triggered (i.e. ,in is the AC terminal voltage, is the active current instruction, is the original reactive current, When is the upper limit of apparent power, the cutoff value needs to be solved by making the apparent power constraint equation hold true to ensure that the device does not exceed the limit.

[0208] The apparent power constraint is expressed in equation form as: ;

[0209] Solve the transformation of this equation :

[0210] Divide both sides by : ;

[0211] Shift the terms and take the positive square root (the current is non-negative): ;

[0212] Cutoff value It is the maximum reactive current that the unit can provide while meeting the apparent power constraint to ensure that the equipment is not overloaded.

[0213] Example 2:

[0214] like Figure 2 As shown, the distributed collaborative source-grid-load-storage clustered joint optimization device is applied to any of the distributed collaborative source-grid-load-storage clustered joint optimization methods described above, including:

[0215] Data acquisition module, used to determine the current active power demand and reactive power demand;

[0216] A data acquisition module is used to obtain the internal resistance of the energy storage unit, the rate of change of the internal resistance with respect to temperature, and the thermal capacity;

[0217] An allocation criterion module is used to determine a positive feedback strength kernel based on internal resistance, a rate of change of internal resistance with respect to temperature, and heat capacity, and to form an allocation criterion by minimizing the maximum value of the positive feedback strength kernel;

[0218] Active current module, used to generate current instructions according to the distribution criteria to meet the active power demand;

[0219] The reactive current module is used to generate reactive current instructions according to the distribution criteria to meet the reactive power demand.

[0220] The above describes the embodiments of this embodiment, but this embodiment is not limited to the above specific implementation methods. The above specific implementation methods are merely illustrative and not restrictive. Ordinary technicians in this field can also make many forms based on the inspiration of this embodiment, all of which are protected by this embodiment.

Claims

1. A distributed collaborative source-grid-load-storage clustered joint optimization method, characterized by: include: Determine the active power demand and reactive power demand at the current moment; Obtain the internal resistance, rate of change of internal resistance with temperature, and heat capacity of the energy storage unit; The positive feedback strength kernel is determined based on the internal resistance, the rate of change of the internal resistance to temperature and the heat capacity, and the allocation criterion is formed by minimizing the maximum value of the positive feedback strength kernel; Generate current instructions according to the distribution criteria to meet the active power demand; Generate reactive current instructions according to the distribution criteria to meet the reactive power demand.

2. The distributed collaborative source-grid-load-storage clustered joint optimization method according to claim 1 is characterized in that: Obtain the internal resistance, rate of change of internal resistance with temperature, and thermal capacity of the energy storage unit, including: The current of energy storage unit i is collected at fixed time intervals , terminal voltage and unit temperature ; The process of obtaining the internal resistance of energy storage unit i is as follows: Calculation of internal resistance by least squares straight line fitting , ; in, represents the number of fitting samples, represents the average current of the fitted sample, represents the average terminal voltage of the fitted sample; The process of obtaining the rate of change of the internal resistance of energy storage unit i with respect to temperature is as follows: The current moment The corresponding internal resistance and unit temperature are combined to obtain: ; Calculate the rate of change of internal resistance with temperature by least squares straight line fitting ,as follows: ； in, represents the average cell temperature of the fitted sample, represents the average internal resistance of the fitted samples; The process of obtaining the heat capacity of energy storage unit i is as follows: At the moment To the current moment Between , the current of energy storage unit i is raised to calculate the thermal capacity as follows: ； ； in, represents the heat capacity of energy storage unit i, Indicates the heat generation power, Indicates a fixed time interval.

3. The distributed collaborative source-grid-load-storage clustered joint optimization method according to claim 2 is characterized in that: The positive feedback strength kernel is determined based on the internal resistance, the rate of change of the internal resistance with respect to temperature, and the heat capacity, including: ； in, represents the nuclear coefficient of energy storage unit i; According to the current Calculates power mapping functions, including: ； The product of the power mapping function and the kernel coefficient is used as the positive feedback intensity kernel .

4. The distributed collaborative source-grid-load-storage clustered joint optimization method according to claim 3 is characterized in that: The allocation criteria are formed by maximizing and minimizing the positive feedback intensity kernel, including: The assignment criteria expressed in terms of infinity norm include: ; in, represents the allocation criterion function, represents the positive feedback intensity kernel of the i-th energy storage unit, represents the intensity kernel vector formed by the positive feedback intensity kernel of N energy storage units, represents the infinity norm.

5. The distributed collaborative source-grid-load-storage clustered joint optimization method according to claim 4 is characterized in that: Generate current commands based on the active power demand according to the allocation criteria, including: Synchronously measure the current increment and power increment of energy storage unit i; the current increment and power increment represent the current moment and time The current difference and power difference; The ratio of power increment to current increment is used as the active conversion coefficient of energy storage unit i; The product of the core coefficient, internal resistance and fixed time interval of energy storage unit i is used as the first parameter ; Use the active power conversion factor as the second parameter .

6. The distributed collaborative source-grid-load-storage clustered joint optimization method according to claim 5 is characterized in that: Generate current instructions according to the distribution criteria to meet the active power demand, including: The first optimization model is constructed based on the first parameter and the second parameter as follows: ； in, Indicates the active power demand, N indicates the number of energy storage units, Represents auxiliary variables; Obtain the closed-form solution of the first optimization model, including: ； in, represents the i-th component of the optimal current vector obtained by solving the optimization model; and will As the current instruction of energy storage unit i .

7. The distributed collaborative source-grid-load-storage clustered joint optimization method according to claim 6 is characterized in that: Generate reactive current instructions according to the distribution criteria to meet reactive power demand, including: Synchronously measure the reactive current increment and reactive power increment of energy storage unit i; the reactive current increment and reactive power increment represent the current moment and time The reactive current difference and reactive power difference; The ratio of reactive power increment to reactive current increment is used as reactive conversion factor ; Read the AC terminal voltage of energy storage unit i and the corresponding apparent power upper limit , to construct apparent power constraints, including: ； in, represents the reactive current to be solved on the AC side of energy storage unit i.

8. The distributed collaborative source-grid-load-storage clustered joint optimization method according to claim 7 is characterized in that: Generate reactive current instructions according to the distribution criteria to meet reactive power demand, including: At the moment To the current moment Between, the current of energy storage unit i is raised and the AC / DC equivalent coefficient is calculated. , ;in, represents the perturbation difference operator, represents the DC side current of the i-th energy storage unit, represents the active component current on the AC side of the i-th energy storage unit; The product of the AC / DC equivalent coefficient and the first parameter is used as the AC side core constant ; The second optimization model is constructed based on the AC side core constant and reactive power conversion coefficient as follows: ； in, represents the reactive power demand, represents the upper bound variable; If the apparent power constraint of energy storage unit i is not triggered, the reactive current to be solved is calculated, including: ； and will As the current instruction of energy storage unit i ; If the apparent power constraint of energy storage unit i is triggered, the cutoff value of reactive current calculation is to be solved. ; Determine the residual reactive power demand based on the cutoff value , repeatedly solve the remaining reactive power demand on the untriggered set, and finally get the reactive current instruction ; Among them, the remaining reactive power demand The calculation formula is as follows: ； in, Indicates the set of energy storage units that trigger the apparent power constraint.

9. A distributed collaborative source-grid-load-storage clustered joint optimization device, applied to the distributed collaborative source-grid-load-storage clustered joint optimization method according to any one of claims 1 to 8, characterized in that: include: Data acquisition module, used to determine the current active power demand and reactive power demand; A data acquisition module is used to obtain the internal resistance of the energy storage unit, the rate of change of the internal resistance with respect to temperature, and the thermal capacity; An allocation criterion module is used to determine a positive feedback strength kernel based on internal resistance, a rate of change of internal resistance with respect to temperature, and heat capacity, and to form an allocation criterion by minimizing the maximum value of the positive feedback strength kernel; Active current module, used to generate current instructions according to the distribution criteria to meet the active power demand; The reactive current module is used to generate reactive current instructions according to the distribution criteria to meet the reactive power demand.

Images