Low-voltage power distribution network hybrid energy storage optimization configuration strategy considering incentive-based demand response

Abstract

The present invention relates to a low-voltage power distribution network hybrid energy storage optimization configuration strategy considering an incentive-based demand response. The method comprises the following steps: step 1: an optical storage system considering an incentive-based demand side response; step 2: a hybrid energy storage system configuration strategy considering the incentive-based demand side response; step 3: a hybrid energy storage system optimization configuration model considering the incentive-based demand side response; and step 4: exemplary application analysis. The present invention has the advantages of fully using the complementation characteristic of resources, maximally exerting the adjustment capability of flexible resources, having a good effect of stabilizing the fluctuation of a net load curve, and being fast in solving speed and high in calculation precision.

Description

Optimal configuration strategy of hybrid energy storage in low-voltage distribution network considering incentive-based demand response Technical Field

[0001] The present invention belongs to the technical field of low-voltage distribution network optimization configuration with high-proportion distributed photovoltaic grid connection and large-scale load growth, and specifically relates to a low-voltage distribution network hybrid energy storage optimization configuration strategy considering incentive-type demand response. Background Art

[0002] Since the structure of low-voltage distribution network is more complex than that of medium-voltage distribution network, the current planning issues of low-voltage distribution network are mostly concentrated on the site selection and sizing of flexible interconnected equipment, the site selection and sizing of energy storage equipment, and the reasonable allocation of demand-side response resources. However, simply configuring energy storage or relying solely on demand-side resource response is not enough to manage the fluctuation problem caused by the large-scale access of distributed power sources to the distribution network. Existing studies have proposed the following problems in the optimal configuration of hybrid energy storage: ① There may be modal aliasing between the sub-curves after EMD decomposition, which in turn affects the final configuration result of the hybrid energy storage system. ② The original power of wind power is first processed by the median average method, and then the EMD decomposition algorithm is applied to further improve the decomposition accuracy, but the median average method result has a certain time delay; ③ VMD requires pre-setting the decomposition mode order, and has poor adaptability. It is necessary to optimize the algorithm to obtain the optimal modal decomposition number; Existing studies study the configuration problem of hybrid energy storage system from the perspective of decomposition algorithm, aiming to smooth the volatility of distributed power sources through the characteristics of energy storage devices at different time scales, while there are still many resources to be awakened on the low-voltage user side, and the current user load demand Demand-side response technology is relatively mature, and it is of great practical significance to consider the impact of demand-side response in power system planning. In addition, in response to the voltage fluctuation problem caused by the increasing number of distributed power sources connected to the low-voltage side, existing technologies have used demand-side response technology to improve the voltage fluctuation of the distribution network. Although existing technologies have conducted detailed research on demand-side response from different angles, if the configuration of demand response and energy storage systems can be considered at the same time, it may further improve the acceptance rate of distributed power sources, reduce the grid interaction cost, reduce the daily load peak-to-valley difference, and reduce the system planning and operation cost. Therefore, the synergistic effect of the two still needs to be studied. Therefore, it is very necessary to provide a low-voltage distribution network hybrid energy storage optimization configuration strategy that fully utilizes the resource complementary characteristics, maximizes the flexibility of resource regulation capabilities, has a good effect on smoothing net load curve fluctuations, has a fast solution speed, and has high calculation accuracy, taking into account incentive-type demand response.

[0003] Summary of the Invention

[0004] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a low-voltage distribution network hybrid energy storage optimization configuration strategy that takes into account incentive-type demand response, which fully utilizes the resource complementarity characteristics, maximizes the flexibility of resource regulation capabilities, has a good effect on smoothing net load curve fluctuations, has a fast solution speed and high calculation accuracy.

[0005] The object of the present invention is achieved by: considering the incentive-type demand response low-voltage distribution network hybrid energy storage optimization configuration strategy, the method comprising the following steps:

[0006] Step 1: Considering the incentive-based demand-side response solar-storage system;

[0007] Step 2: Hybrid energy storage system configuration strategy considering incentive-based demand-side response;

[0008] Step 3: Optimal configuration model of hybrid energy storage system considering incentive-based demand-side response;

[0009] Step 4: Demonstration application analysis.

[0010] The hybrid energy storage system configuration strategy considering the incentive-based demand-side response in step 2 is specifically as follows: the configuration strategy of the hybrid energy storage system (HESS) includes both its location configuration and the configuration of its rated capacity and rated power; in order to reasonably allocate the lithium iron phosphate batteries and supercapacitors in the HESS, the system net load curve is first decomposed and recombined and reconstructed, and the high-frequency fluctuations are smoothed by supercapacitors and the directly controlled loads in the incentive-based demand-side response (IDR), while the low-frequency fluctuations are smoothed by the mains, lithium iron phosphate batteries, and the non-directly controlled loads in the IDR. Specifically, the following steps are included:

[0011] Step 2.1: Improved variational mode algorithm;

[0012] Step 2.2: Hybrid energy storage system power allocation;

[0013] Step 2.3: Hybrid energy storage system capacity allocation;

[0014] Step 2.4: Hybrid energy storage system installation location.

[0015] The improved variational modal algorithm in step 2.1 decomposes and transforms the net load curve and then recombines and reconstructs it, including the following steps:

[0016] Step 2.11: Use the Grey Wolf algorithm to optimize the optimal decomposition scale K. Decompose the reference power signal x(t) of the net power curve to obtain the modal components. The calculation formula is: Where x(t) is the power signal to be decomposed; u k (t) is the corresponding modal component; K is the number of decomposed modes;

[0017] Step 2.12: Use Hilbert transform to transform each modal component u k (t) is converted into its instantaneous amplitude and frequency phase, and the calculation formula is:

[0018] Step 2.13: Using the values ​​obtained in step 2.12, construct each modal component u k The analytical signal in polar coordinate form of (t) is calculated as follows: Where z k (t) is u k (t) in polar coordinate form; α k (t) is the amplitude of the kth mode function; φ k (t) is the center frequency of the kth mode function; u of each modal component k The instantaneous frequency w of (t) k (t) is: And the corresponding Hilbert spectrum is obtained as follows:

[0019] Step 2.14: The Hilbert amplitude spectrum is composed of the real part of the analytical signal. The Hilbert marginal spectrum can be obtained by integrating the amplitude spectrum over time:

[0020] Step 2.15: The Hilbert marginal spectrum can integrate the amplitude distribution of each modal component at different frequency scales. Finally, the high- and low-frequency boundary points are determined by the marginal spectrum. Then, the high- and low-frequency modal components are reconstructed and smoothed using HESS, IDR, and mains power respectively: P net (t) = P load (t)-P pv (t) = P high (t)+P low (t)(8), Where, P net (t) is the net load power at time t; P load (t) is the load power at time t; P pv (t) is the distributed photovoltaic output at time t; P high (t), P low (t) are the high-frequency power and low-frequency power at time t; P sc (t), P b (t) are the power of supercapacitor and lithium iron phosphate battery at time t; P DCL (t), P NDCL (t) are the actual response quantities of direct control load and non-direct control load at time t respectively; P buy (t) is the power provided by the mains at time t; a is the number of high-frequency modes in the system; K is the total number of modes; uk (t) is the corresponding modal component.

[0021] The power allocation of the hybrid energy storage system in step 2.2 is specifically as follows: to determine the rated power and rated capacity of each device in the HESS and the power of the direct-controlled and indirect-controlled loads in the IDR, the output ratio coefficients α and β are set respectively, and α+β=1 is satisfied. The rated power of each device is shown as follows: Where α is the proportional coefficient of the output required by the HESS; β is the proportional coefficient of the output required by the IDR part; the meanings of the other variables are the same as those in formulas (8)-(10); in addition, since the energy storage system has certain losses during the charging and discharging process, the charging and discharging efficiency of the energy storage is further considered in the rated power calculation. Finally, the actual rated powers of the lithium iron phosphate battery and supercapacitor are: Where P′ SC , P′ b are the actual rated power of supercapacitor and lithium iron phosphate battery respectively; are respectively the charging and discharging power signals of the supercapacitor at time t; are respectively the charging and discharging power signals of the lithium iron phosphate battery at time t; η c,SC ,η d,SC is the charge and discharge efficiency of the supercapacitor; η c,b ,η d,b is the charge and discharge efficiency of lithium iron phosphate battery.

[0022] The capacity allocation of the hybrid energy storage system in step 2.3 is specifically as follows: based on the actual charge and discharge conditions of the supercapacitor and the lithium iron phosphate battery, the cumulative capacity of the supercapacitor and the lithium iron phosphate battery in the time period T is integrated to obtain the cumulative capacity of the supercapacitor and the lithium iron phosphate battery in the time period: Where, E SC0 (t), E b0 (t) are the cumulative capacities of the supercapacitor and lithium iron phosphate battery during the cycle. Considering that the energy storage element cannot be overcharged or over-discharged and has SOC state limitations, the maximum and minimum values ​​of the cumulative capacity are taken to correct the above formula. Finally, the rated capacities of the supercapacitor and lithium iron phosphate battery are: Where, SOC up,SC , SOC down,SC SOC is the upper and lower limit constraint of supercapacitor state of charge; up,b , SOC down,b These are the upper and lower limit constraints for the state of charge of the lithium iron phosphate battery.

[0023] The installation location of the hybrid energy storage system in step 2.4 is specifically as follows: there are two location configuration methods for energy storage on the low-voltage side of the traditional distribution network, namely, distributed layout and centralized layout; distributed layout is to configure energy storage on-site near the distributed photovoltaic system according to the installation location of the user's rooftop photovoltaic system; centralized layout is to configure the energy storage equipment at the public connection point, and configure appropriate capacity according to the total amount of distributed photovoltaic system; the energy storage equipment distributed on the user side has the advantages of flexible access location and fewer restrictions on the location of the access site and environmental natural conditions, but because the distributed layout is close to the user side, the controllability is relatively weak and the operation and maintenance are more difficult; the centralized energy storage has the advantage of good aggregation effect, but the corresponding line loss will also be larger; taking into account the above two layout methods, as well as the control method, operation and maintenance difficulty, and economic benefits of the actual operation of distributed photovoltaics, a distributed layout method of energy storage system is adopted for the low-voltage distribution network with a radial form.

[0024] The hybrid energy storage system optimization configuration model considering incentive-type demand-side response in step 3 is specifically as follows: with the advancement of rooftop photovoltaic policies, more and more distributed photovoltaics are connected to the low-voltage distribution network nearby. The randomness and volatility of their output have aggravated the volatility of the net load curve, seriously affecting the normal operation of the low-voltage power grid; in order to smooth the fluctuation of the net load curve, a hybrid energy storage system optimization configuration model considering incentive-type demand-side response is constructed; since demand response resources naturally exist on the user side of the system, their utilization is only considered at the operation level, and it is impossible to increase the available demand response resources in the system by increasing investment at the planning level; the incentive-type demand response resource type considers interruptible loads, specifically including the following steps:

[0025] Step 3.1: Objective function;

[0026] Step 3.2: Constraints;

[0027] Step 3.3: Improve the whale algorithm.

[0028] The objective function in step 3.1 is specifically as follows: the objective function is to minimize the total system cost and active power fluctuation value, and the decision variables are the power and capacity of the supercapacitor and lithium iron phosphate battery and the actual response active power of the direct-controlled load and non-direct-controlled load participating in the demand-side response; the objective function calculation formula is: min F = λ1F1 + λ2F2 (19), min F1 = C SC +C b +C IDR +C buy (20), Where F1 is the total system cost, which is composed of the investment cost C of the energy storage equipment. SC , operation and maintenance costs C b , compensation cost of incentive-based demand-side response C IDRand the power purchase cost C of the upper power grid buy composition; F2 is the system active power fluctuation value; λ1 and λ2 are the weights of the two sub-objective functions F1 and F2, which can be obtained through the hierarchical analysis method. Considering the different dimensions of the two sub-objectives, normalization is required during the calculation process; K is the number of all load nodes; T is 24 hours a day; α is the discount rate; T SC 、T b k is the service life of the supercapacitor and lithium iron phosphate battery equipment; SC1 、k SC2 、k SC3 and k b1 、k b2 、k b3 are the unit power cost, unit capacity cost and unit capacity operation and maintenance cost of supercapacitor and lithium iron phosphate battery respectively; P SC 、P b are the power of supercapacitor and lithium iron phosphate battery respectively; E SC 、E b are the capacities of supercapacitor and lithium iron phosphate battery respectively; N is the number of users participating in demand-side response; P i,t and P i,t-1 are the active power values ​​at time t and time t-1 respectively; c DCL 、c NDCL are the unit compensation electricity prices for direct-controlled load and non-direct-controlled load respectively; c is the time-of-use electricity price; P DCL,i (t), P NDCL,i (t) are the actual response quantities of the direct-controlled load and the non-direct-controlled load at node i at time t; P t It is the power purchased by the low-voltage distribution network from the upper power grid at time t.

[0029] The constraints in step 3.2 are as follows: in addition to the most basic power flow constraints, they also need to satisfy distributed photovoltaic grid-connected constraints, hybrid energy storage charging and discharging constraints, hybrid energy storage SOC constraints, power balance constraints, demand-side response capability constraints, and power constraints with the superior contact point. ① The distributed photovoltaic grid-connected constraint is: combined with the restrictions on renewable energy grid-connected fluctuations at different time scales, for any time period t, the distributed photovoltaic volatility constraint is: under short-time-scale high-frequency fluctuations: Under long-term low-frequency fluctuations: Where, are the maximum fluctuation power at short time scale and long time scale respectively; P cap (t) is the installed capacity of distributed photovoltaic power generation; ② The hybrid energy storage charge and discharge constraints are: when the remaining capacity of the lithium battery and supercapacitor cannot meet the charge and discharge power at the next moment, the charge and discharge power of the lithium iron phosphate battery and supercapacitor should be adjusted: Where ηb is the charge and discharge efficiency of lithium iron phosphate battery; η SC is the charge and discharge efficiency of the supercapacitor; ③ The SOC constraint of the hybrid energy storage is: SOC b,min ≤SOC b (t)≤SOC b,max (26), SOC SC,min ≤SOC SC (t)≤SOC SC,max (27), where SOC b (t), SOC SC (t) is the state of charge of the lithium iron phosphate battery and supercapacitor at time t; SOC b,min , SOC b,max SOC is the lower and upper limits of the state of charge of lithium iron phosphate batteries; SC,min , SOC SC,max are the lower and upper limits of the supercapacitor state of charge respectively; ④The power balance constraint is: The meanings of the variables in the formula are the same as those in formula (22); ⑤ The demand-side response capacity constraint is that the response capacity of the load participating in the demand response during operation also has certain constraints. The maximum response capacity constraint of DCL and NDCL is: 0≤P DCL,i (t)≤M DCL,i (t)(29), 0≤P NDCL,i (t)≤M NDCL,i (t)(30), where M DCL,i 、M NDCL,i is the maximum response of DCL and NDCL at node i in time period t; ⑥ The power constraint with the upper contact point is: 0≤P(t)≤P max (31), where P(t) is the real-time power between the distribution transformer area and the upper power grid connection point; P max is the upper limit of the interaction power of the contact point.

[0030] The improved whale algorithm in step 3.3 specifically includes the following steps:

[0031] Step 3.31: Input the photovoltaic and load timing output curve parameters and algorithm-related parameters;

[0032] Step 3.32: Generate the initial whale population using the cubic chaos mapping method and calculate the objective function value.

[0033] Step 3.33: Perform a fast non-dominated ranking of the position of each individual whale in the algorithm, i.e., the power and capacity of the supercapacitor and lithium iron phosphate battery, and the actual response size of the directly controlled load and the non-directly controlled load participating in the demand-side response;

[0034] Step 3.34: Each whale's individual position is optimized through three steps: surround search, bubble attack, and random mutation, and the position of the whale individual is updated;

[0035] Step 3.35: Determine whether the maximum number of iteration terminations has been reached. If not, return to step 3.32; if reached, output the final optimization solution.

[0036] Beneficial effects of the invention: The invention is a hybrid energy storage optimization configuration strategy for low-voltage distribution network considering incentive-type demand response. Taking into account that distributed photovoltaic with randomness and volatility characteristics will also be connected to the low-voltage distribution network on a large scale, with the growth of electricity load, the low-voltage distribution network in the future will also show significant "source-load double-side randomness" characteristics; in order to solve the random volatility brought by the two to the low-voltage distribution network and improve the power quality of the low-voltage distribution network, it is urgent to optimize and integrate various flexibility resource elements; low-voltage side flexibility resources include energy storage systems, demand-side response, flexible interconnection equipment, etc.; different types of flexibility resources have their own advantages and disadvantages in the distribution network, such as storage It can be used to smooth out peaks and fill valleys on the user load side, but the equipment investment cost is high, and economic efficiency has become the main factor limiting its large-scale development; demand-side response resources have low costs and fast responses, but are greatly affected by the subjective wishes of users and require cooperation with a sound market mechanism; considering the limitations of the above-mentioned single flexibility resource, the above-mentioned resource characteristics should be coordinated in the low-voltage distribution network, and the complementary characteristics between resources should be fully utilized to maximize the regulation capacity of flexibility resources, achieve a high proportion of distributed photovoltaics and load fluctuation smoothing and system economic operation; the present invention proposes a low-voltage distribution network hybrid energy storage optimization configuration model taking into account incentive-type demand-side response. According to the characteristics of energy storage equipment and demand-side response resources at different time scales, the fluctuations caused by distributed photovoltaics and loads are jointly smoothed; first, the improved VMD algorithm is used to perform multi-scale decomposition and combined reconstruction of the net load curve; then an objective function is established with the minimum sum of the total system cost and the active power fluctuation value; then the improved whale optimization algorithm is used to solve the optimization configuration model; the demonstration example shows that compared with the configuration of a single energy storage, this scheme has the best effect on smoothing the fluctuations of the net load curve; the present invention has the advantages of fully utilizing the complementary characteristics of resources, maximizing the flexibility of resource regulation capabilities, having a good effect on smoothing the fluctuations of the net load curve, fast solution speed, and high calculation accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] FIG1 is a topological diagram of a low-voltage distribution network in which hybrid energy storage and demand-side response are involved in the present invention.

[0038] FIG2 is a flow chart of the improved VMD algorithm of the present invention.

[0039] FIG3 is a flow chart of the improved whale algorithm of the present invention.

[0040] FIG4 is an actual topological diagram of a certain area in northern China according to the method of the present invention.

[0041] FIG5 is a diagram showing the decomposition results of the improved VMD algorithm of the present invention.

[0042] FIG6 is a graph showing the reconstructed Plow curve and Phigh curve of the present invention.

[0043] FIG7 is a schematic diagram of the iterative process of the grey wolf algorithm and the genetic algorithm of the present invention.

[0044] FIG8 is a schematic diagram showing a comparison of active power fluctuations of the method of the present invention.

[0045] FIG9 is a load and photovoltaic output per unit value curve diagram of the present invention. DETAILED DESCRIPTION

[0046] The present invention will be further described below with reference to the accompanying drawings.

[0047] Example 1

[0048] As shown in Figure 1-9, the hybrid energy storage optimization configuration strategy for a low-voltage distribution network considering incentive-based demand response includes the following steps:

[0049] Step 1: Considering the incentive-based demand-side response solar-storage system;

[0050] Step 2: Hybrid energy storage system configuration strategy considering incentive-based demand-side response;

[0051] Step 3: Optimal configuration model of hybrid energy storage system considering incentive-based demand-side response;

[0052] Step 4: Demonstration application analysis.

[0053] In this embodiment, the photovoltaic storage system considering the incentive-type demand-side response is specifically as follows: the present invention defines the net load curve as the curve obtained by subtracting the distributed photovoltaic output from the load curve, so the fluctuation of the net load curve is composed of the fluctuations of the two; the volatility of the load is mostly caused by its temporal and spatial changes; and the volatility of the distributed photovoltaic output is greatly affected by the environment, and existing studies have shown that photovoltaic fluctuations at different time scales have different impacts on the distribution network, among which photovoltaic fluctuations at the min~h level mainly affect the system's backup and scheduling strategies, while photovoltaic fluctuations at the s~min level mainly affect the system's frequency regulation strategy; therefore, using flexible resources in different states to balance net load fluctuations at different time scales will effectively smooth the level of net load curve fluctuations in the low-voltage distribution network; the present invention uses a hybrid energy storage system (HESS) and incentive-type demand-side response (IDR) to achieve this purpose;

[0054] The architecture of a low-voltage distribution network system equipped with hybrid energy storage and incentive-based demand-side response is shown in Figure 1. The hybrid energy storage system consists of lithium iron phosphate batteries and supercapacitors, which respectively smooth power signals of different frequencies. The lithium iron phosphate batteries utilize the high energy density and long response time to compensate for the large fluctuation amplitude and low frequency components in the power signal. The supercapacitors utilize the high power density and short response time to compensate for the small fluctuation amplitude and high frequency components in the power signal. The demand-side response of the present invention selects incentive-based demand-side response with fast adjustment speed, flexible adjustment mode, and large adjustment potential. Based on the load response time and response method, incentive-based demand-side response can be divided into direct-controlled load (DCL) and non-direct-controlled load (NDCL). DCL directly participates in the response by controlling the load disconnection and disconnection by the dispatching center, thus having the characteristics of fast response time and balancing the fluctuations in the high-frequency part. NDCL is a spontaneous response participated by users, and its response time is longer than DCL, often ranging from 10 minutes to hours, so it can be used to balance the fluctuations in the low-frequency part. In summary, through the coordination of hybrid energy storage systems and incentive-based demand-side response at different time scales, the goal of smoothing the fluctuations of the net load curve can be achieved.

[0055] The hybrid energy storage system configuration strategy considering incentivized demand-side response in step 2 is as follows: the configuration strategy of the hybrid energy storage system (HESS) includes both its location configuration and the configuration of its rated capacity and rated power. To rationally allocate the lithium iron phosphate batteries and supercapacitors in the HESS, the system net load curve is first decomposed and recombined. High-frequency fluctuations are smoothed using supercapacitors and the directly controlled loads in incentivized demand-side response (IDR), while low-frequency fluctuations are smoothed using the mains, lithium iron phosphate batteries, and the non-directly controlled loads in the IDR. The specific steps include:

[0056] Step 2.1: Improved variational mode algorithm;

[0057] Step 2.2: Hybrid energy storage system power allocation;

[0058] Step 2.3: Hybrid energy storage system capacity allocation;

[0059] Step 2.4: Hybrid energy storage system installation location.

[0060] In this embodiment, based on the low-voltage distribution network architecture involving hybrid energy storage and demand-side response proposed in step 1, the present invention uses an improved VMD algorithm to decompose, transform, and reconstruct the system net load curve to obtain the initial power allocation of each control resource, so as to subsequently carry out targeted net load curve volatility smoothing. The decomposition process is shown in Figure 2; the VMD algorithm is an algorithm that can better realize signal frequency decomposition, and can decompose the signal into sub-signals of multiple frequency components at one time, thereby avoiding the generation of false components in multiple iterative decomposition processes, which affects the occurrence of problems in the final decomposition results. However, the selection of the decomposition mode number K value of this method is relatively subjective. When the K value does not match the number of useful components in the processed signal, insufficient data decomposition, increased error or over-decomposition, and the generation of false signal components will occur; therefore, the present invention uses the gray wolf algorithm to determine the decomposition mode number K value in the VMD algorithm, so that it has an adaptive decomposition mode number, and the sub-signal sequence after decomposition by the VMD algorithm is less affected by noise.

[0061] The improved variational modal algorithm in step 2.1 decomposes and transforms the net load curve and then recombines and reconstructs it, including the following steps:

[0062] Step 2.11: Use the Grey Wolf algorithm to optimize the optimal decomposition scale K. Decompose the reference power signal x(t) of the net power curve to obtain the modal components. The calculation formula is: Where x(t) is the power signal to be decomposed; u k (t) is the corresponding modal component; K is the number of decomposed modes;

[0063] Step 2.12: Use Hilbert transform to transform each modal component u k (t) is converted into its instantaneous amplitude and frequency phase, and the calculation formula is:

[0064] Step 2.13: Using the values ​​obtained in step 2.12, construct each modal component u k The analytical signal in polar coordinate form of (t) is calculated as follows: Where z k (t) is u k (t) in polar coordinate form; α k (t) is the amplitude of the kth mode function; φ k (t) is the center frequency of the kth mode function; u of each modal component k The instantaneous frequency w of (t) k (t) is: And the corresponding Hilbert spectrum is obtained as follows:

[0065] Step 2.14: The Hilbert amplitude spectrum is composed of the real part of the analytical signal. The Hilbert marginal spectrum can be obtained by integrating the amplitude spectrum over time:

[0066] Step 2.15: The Hilbert marginal spectrum can integrate the amplitude distribution of each modal component at different frequency scales. Finally, the high- and low-frequency boundary points are determined by the marginal spectrum. Then, the high- and low-frequency modal components are reconstructed and smoothed using HESS, IDR, and mains power respectively: P net (t) = P load (t)-P pv (t) = P high (t)+P low (t)(8), Where, P net (t) is the net load power at time t; P load (t) is the load power at time t; P pv (t) is the distributed photovoltaic output at time t; P high (t), P low (t) are the high-frequency power and low-frequency power at time t; P sc (t), P b (t) are the power of supercapacitor and lithium iron phosphate battery at time t; P DCL (t), P NDCL (t) are the actual response quantities of direct control load and non-direct control load at time t respectively; P buy (t) is the power provided by the mains at time t; a is the number of high-frequency modes in the system; K is the total number of modes; u k (t) is the corresponding modal component.

[0067] The power allocation of the hybrid energy storage system in step 2.2 is specifically as follows: to determine the rated power and rated capacity of each device in the HESS and the power of the direct-controlled and indirect-controlled loads in the IDR, the output ratio coefficients α and β are set respectively, and α+β=1 is satisfied. The rated power of each device is shown as follows: Where α is the proportional coefficient of the HESS output required; β is the proportional coefficient of the IDR output required; the other variables have the same meanings as formulas (8)-(10); when the energy storage and demand-side response load ratios are equal, the maximum range of wind and solar fluctuations can be absorbed, that is, α = β = 0.5, so the initial values ​​of the two are also set to 0.5, but during operation, the ratio of the two should be appropriately adjusted according to factors such as economy; the above formula is only a rough calculation of the rated power of lithium iron phosphate batteries and supercapacitors. Considering the continuous operation of the energy storage system, its rated power should be greater than or equal to the absolute value of the charging and discharging power it bears at any time; in addition, since the energy storage system has certain losses during the charging and discharging process, the charging and discharging efficiency of the energy storage is further considered in the rated power calculation. Finally, the actual rated power of the lithium iron phosphate battery and supercapacitor are: Where P′ SC , P′ b are the actual rated power of supercapacitor and lithium iron phosphate battery respectively; are respectively the charging and discharging power signals of the supercapacitor at time t; are respectively the charging and discharging power signals of the lithium iron phosphate battery at time t; η c,SC ,η d,SC is the charge and discharge efficiency of the supercapacitor; η c,b ,η d,b is the charge and discharge efficiency of lithium iron phosphate battery.

[0068] The capacity allocation of the hybrid energy storage system in step 2.3 is specifically as follows: based on the actual charge and discharge conditions of the supercapacitor and the lithium iron phosphate battery, the integration within a time period T (T is taken as 2h) is performed to obtain the cumulative capacity of the two within the period: Where, E SC0 (t), E b0 (t) are the cumulative capacities of the supercapacitor and lithium iron phosphate battery during the cycle. Considering that the energy storage element cannot be overcharged or over-discharged and has SOC state limitations, the maximum and minimum values ​​of the cumulative capacity are taken to correct the above formula. Finally, the rated capacities of the supercapacitor and lithium iron phosphate battery are: Where, SOC up,SC , SOC down,SC SOC is the upper and lower limit constraint of supercapacitor state of charge; up,b , SOC down,b These are the upper and lower limit constraints for the state of charge of the lithium iron phosphate battery.

[0069] The installation location of the hybrid energy storage system in step 2.4 is specifically as follows: there are two location configuration methods for energy storage on the low-voltage side of the traditional distribution network, namely, distributed layout and centralized layout; distributed layout is to configure energy storage on-site near the distributed photovoltaic system according to the installation location of the user's rooftop photovoltaic system; centralized layout is to configure the energy storage equipment at the public connection point, and configure appropriate capacity according to the total amount of distributed photovoltaic system; the energy storage equipment distributed on the user side has the advantages of flexible access location and fewer restrictions on the location of the access site and environmental natural conditions, but because the distributed layout is close to the user side, the controllability is relatively weak and the operation and maintenance are more difficult; the centralized energy storage has the advantage of good aggregation effect, but the corresponding line loss will also be larger; taking into account the above two layout methods, as well as the control method, operation and maintenance difficulty, and economic benefits of the actual operation of distributed photovoltaics, a distributed layout method of energy storage system is adopted for the low-voltage distribution network with a radial form.

[0070] The hybrid energy storage system optimization configuration model considering incentive-type demand-side response in step 3 is specifically as follows: with the advancement of rooftop photovoltaic policies, more and more distributed photovoltaics are connected to the low-voltage distribution network nearby. The randomness and volatility of their output have aggravated the volatility of the net load curve, seriously affecting the normal operation of the low-voltage power grid; in order to smooth the fluctuation of the net load curve, a hybrid energy storage system optimization configuration model considering incentive-type demand-side response is constructed; since demand response resources naturally exist on the user side of the system, their utilization is only considered at the operation level, and it is impossible to increase the available demand response resources in the system by increasing investment at the planning level; the incentive-type demand response resource type considers interruptible loads, specifically including the following steps:

[0071] Step 3.1: Objective function;

[0072] Step 3.2: Constraints;

[0073] Step 3.3: Improve the whale algorithm.

[0074] The objective function in step 3.1 is specifically as follows: the objective function is to minimize the total system cost and active power fluctuation value, and the decision variables are the power and capacity of the supercapacitor and lithium iron phosphate battery and the actual response active power of the direct-controlled load and non-direct-controlled load participating in the demand-side response; the objective function calculation formula is: min F = λ1F1 + λ2F2 (19), min F1 = C SC +C b +C IDR +C buy (20), Where F1 is the total system cost, which is composed of the investment cost C of the energy storage equipment. SC , operation and maintenance costs C b , compensation cost of incentive-based demand-side response C IDRand the power purchase cost C of the upper power grid buy composition; F2 is the system active power fluctuation value; λ1 and λ2 are the weights of the two sub-objective functions F1 and F2, which can be obtained through the hierarchical analysis method. Considering the different dimensions of the two sub-objectives, normalization is required during the calculation process; K is the number of all load nodes; T is 24 hours a day; α is the discount rate; T SC 、T b k is the service life of the supercapacitor and lithium iron phosphate battery equipment; SC1 、k SC2 、k SC3 and k b1 、k b2 、k b3 are the unit power cost, unit capacity cost and unit capacity operation and maintenance cost of supercapacitor and lithium iron phosphate battery respectively; P SC 、P b are the power of supercapacitor and lithium iron phosphate battery respectively; E SC 、E b are the capacities of supercapacitor and lithium iron phosphate battery respectively; N is the number of users participating in demand-side response; P i,t and P i,t-1 are the active power values ​​at time t and time t-1 respectively; c DCL 、c NDCL are the unit compensation electricity prices for direct-controlled load and non-direct-controlled load respectively; c is the time-of-use electricity price; P DCL,i (t), P NDCL,i (t) are the actual response quantities of the direct-controlled load and the non-direct-controlled load at node i at time t; P t It is the power purchased by the low-voltage distribution network from the upper power grid at time t.

[0075] The constraints in step 3.2 are as follows: in addition to the most basic power flow constraints, the distributed photovoltaic grid-connected constraints, hybrid energy storage charging and discharging constraints, hybrid energy storage SOC constraints, power balance constraints, demand-side response capability constraints, and power constraints with the superior contact point must also be met. The first distributed photovoltaic grid-connected constraint is: based on the restrictions on renewable energy grid-connected fluctuations at different time scales in my country's "Q / GDW 1480-2015 Technical Regulations for Distributed Power Sources Access to the Grid," for any time period t, the distributed photovoltaic volatility constraint is: under short-time scale (1 minute) high-frequency fluctuations: Under long-term (10min) low-frequency fluctuations: Where, are the maximum fluctuation power at short time scale and long time scale respectively; P cap(t) is the installed capacity of distributed photovoltaic power generation; ② The hybrid energy storage charge and discharge constraints are: when the remaining capacity of the lithium battery and supercapacitor cannot meet the charge and discharge power at the next moment, the charge and discharge power of the lithium iron phosphate battery and supercapacitor should be adjusted: Where η b is the charge and discharge efficiency of lithium iron phosphate battery; η SC is the charge and discharge efficiency of the supercapacitor; ③ The SOC constraint of the hybrid energy storage is: SOC b,min ≤SOC b (t)≤SOC b,max (26), SOC SC,min ≤SOC SC (t)≤SOC SC,max (27), where SOC b (t), SOC SC (t) is the state of charge of the lithium iron phosphate battery and supercapacitor at time t; SOC b,min , SOC b,max The lower and upper limits of the state of charge of lithium iron phosphate batteries are usually taken as 0.2 and 0.8; SOC SC,min , SOC SC,max are the lower and upper limits of the supercapacitor state of charge, respectively, usually taken as 0.2 and 0.9; ④ The power balance constraint is: The meanings of the variables in the formula are the same as those in formula (22); ⑤ The demand-side response capacity constraint is that the response capacity of the load participating in the demand response during operation also has certain constraints. The maximum response capacity constraint of DCL and NDCL is: 0≤P DCL,i (t)≤M DCL,i (t)(29), 0≤P NDCL,i (t)≤M NDCL,i (t)(30), where M DCL,i 、M NDCL,i is the maximum response of DCL and NDCL at node i in time period t; ⑥ The power constraint with the upper contact point is: 0≤P(t)≤P max (31), where P(t) is the real-time power between the distribution transformer area and the upper power grid connection point; P max is the upper limit of the interaction power of the contact point.

[0076] The improved whale algorithm in step 3.3 specifically includes the following steps:

[0077] Step 3.31: Input the photovoltaic and load timing output curve parameters and algorithm-related parameters;

[0078] Step 3.32: Generate the initial whale population using the cubic chaos mapping method and calculate the objective function value.

[0079] Step 3.33: Perform a fast non-dominated ranking of the position of each individual whale in the algorithm, i.e., the power and capacity of the supercapacitor and lithium iron phosphate battery, and the actual response size of the directly controlled load and the non-directly controlled load participating in the demand-side response;

[0080] Step 3.34: Each whale's individual position is optimized through three steps: surround search, bubble attack, and random mutation, and the position of the whale individual is updated;

[0081] Step 3.35: Determine whether the maximum number of iteration terminations has been reached. If not, return to step 3.32; if reached, output the final optimization solution.

[0082] In this embodiment, the present invention adopts an improved whale optimization algorithm to solve the established hybrid energy storage system optimization configuration model considering the incentive-type demand-side response; the whale optimization algorithm, like the particle swarm optimization algorithm, is also one of the group intelligence optimization algorithms that simulate animal behavior. This algorithm finds the best search result by simulating the spiral attack mode adopted in the whale hunting process. Its advantages are simple optimization mechanism, strong optimization ability and fewer parameters to be considered; but the iteration efficiency is often affected by the randomness of the initial population, so the improved whale optimization algorithm is adopted to generate dynamic and A uniform initial population; the decision variables of the objective function established by the present invention are the power and capacity of supercapacitors and lithium iron phosphate batteries and the actual response active power of direct-controlled loads and non-direct-controlled loads participating in the demand-side response, which corresponds to the position space of the whale group; the position initialization of the whale population corresponds to the power configuration of the hybrid energy storage system and the initialization of the power of the demand-side response; after initialization, the position is continuously updated through three steps of surround search, bubble attack, and random mutation to search for optimization, and the best position finally obtained by the whale group is the optimal configuration result of the scheme; the solution flow chart of the improved whale algorithm is shown in Figure 3.

[0083] The present invention is a hybrid energy storage optimization configuration strategy for low-voltage distribution network considering incentive-type demand response. During use, the present invention considers the characteristics of energy storage equipment and incentive-type demand-side response resources at different time scales. First, the improved VMD algorithm is used to decompose and reconstruct the system net load curve. The high-frequency fluctuations are smoothed by supercapacitors and directly controlled loads, and the low-frequency part is smoothed by mains power, lithium iron phosphate batteries and non-directly controlled loads. Then, an optimization configuration model of the hybrid energy storage system considering incentive-type demand-side response is established, and the improved whale optimization algorithm is used to solve the optimization configuration model. Finally, an actual low-voltage distribution network station in a certain place in northern my country is used for example verification. The results show that the model proposed by the present invention has the following advantages: ① Using the gray wolf algorithm The improved VMD algorithm has better solution speed and calculation accuracy than the VMD algorithm improved by the genetic algorithm, indicating that the proposed method has certain advantages. ② Compared with the solution of only configuring a hybrid energy storage system without the participation of demand-side response resources, the proposed solution saves 22.2% of the total cost and reduces the active power fluctuation value by 28.4%. That is, the participation of demand-side response resources effectively reduces the capacity and power configuration of the hybrid energy storage system. The synergistic effect of flexible resources in both load and storage enables the system to achieve optimal economic efficiency, which is suitable for the future large-scale access of distributed photovoltaic and load. ③ The total cost of the hybrid energy storage configuration solution using the adaptive VMD algorithm is reduced by 19.2% and 3.8% respectively compared with the solution of configuring a single supercapacitor energy storage solution and a single lithium iron phosphate battery energy storage solution. The fast response and high power density operating characteristics of the supercapacitor improve the operating environment of the lithium iron phosphate battery energy storage system, reduce the high-frequency power component borne by the lithium iron phosphate battery, and reduce the number of replacements, thereby achieving the purpose of delaying equipment investment. The present invention has the advantages of fully utilizing the complementary characteristics of resources, maximizing the adjustment capacity of flexible resources, effectively smoothing the fluctuation of the net load curve, fast solution speed, and high calculation accuracy.

[0084] Example 2

[0085] As shown in Figure 1-9, the hybrid energy storage optimization configuration strategy for a low-voltage distribution network considering incentive-based demand response includes the following steps:

[0086] Step 1: Considering the incentive-based demand-side response solar-storage system;

[0087] Step 2: Hybrid energy storage system configuration strategy considering incentive-based demand-side response;

[0088] Step 3: Optimal configuration model of hybrid energy storage system considering incentive-based demand-side response;

[0089] Step 4: Demonstration application analysis.

[0090] In this embodiment, the application analysis of the demonstration example is as follows: ① Demonstration example overview: Take the low-voltage distribution network in a certain place in northern my country as an example for calculation analysis, and the topology diagram is shown in Figure 4; there are 1 transformer node and 18 load nodes, whose sizes are shown in Appendix 1, the total load is 267.22kW, and the branch parameters between each node in the station are shown in Appendix 2; the rated capacity of the distribution transformer is 500kVA; the qualified voltage range of the station is rated voltage + 7% - 7%; since only the impact of demand-side response resources and hybrid energy storage systems on the net load curve fluctuation is considered, the light The volt-capacity is not optimized. An 8kW rooftop distributed photovoltaic system is installed at each load node, and it is assumed that the photovoltaic output of each node is the same. The load and photovoltaic output curves are shown in Figure 9. The basic parameters of the hybrid energy storage system are shown in Appendix 3. The installation locations are between 1 and 18 nodes, and only two energy storage systems are installed in the low-voltage area. The specific installation location will be configured according to different plans. The electricity price is the time-of-use electricity price in the area, as shown in Appendix 4. The discount rate is 10%, the maximum response capacity of DCL and NDCL is 12kW and 8kW respectively, and the load participating in demand-side response accounts for 5% of the total load. The maximum number of decomposition modes is set to 2.

[0091] ②Result analysis: 1) Decomposition results of the improved VMD algorithm: The net load curve is decomposed using the proposed improved VMD algorithm. The optimal decomposition mode number is 8, thus obtaining 8 sub-signal sequences IMF1-IMF8, as shown in Figure 5; IMF1-4 are sub-signals with lower frequency fluctuations, which are recombined and reconstructed into P low Signal, using lithium iron phosphate battery and non-direct control load to smooth fluctuations; IMF5-8 is a sub-signal with higher frequency fluctuations, which is recombined and reconstructed into P high The signal is smoothed using supercapacitors and direct-controlled loads. The reconstructed curve is shown in Figure 6. Through the above decomposition and reconstruction, the advantages of different flexibility resources at different time scales are fully utilized, and the frequent charging and discharging of the energy storage system is avoided to a certain extent.

[0092] In addition, in order to verify the advantages of using the gray wolf algorithm in the improved VMD algorithm of the present invention, the modal decomposition parameters of VMD are compared between the genetic algorithm and the gray wolf algorithm, and the iterative process is shown in Figure 7. As can be seen from Figure 7, the gray wolf algorithm converges to the minimum fitness function value after 8 iterations, while the genetic algorithm converges to the minimum fitness function value after 18 iterations. The comparison results show that the gray wolf algorithm can find the optimal solution more quickly, and its solution speed and calculation accuracy are better than the genetic algorithm. This is because the local search and global search of the gray wolf algorithm are carried out simultaneously, making it easier to jump out of the local optimum. The gray wolf algorithm is inspired by the predatory behavior of gray wolves and has better information sharing and transmission, thereby significantly improving the global search capability.

[0093] 2) Analysis of hybrid energy storage and demand-side response configuration schemes: To verify the effectiveness of the strategy proposed in this invention, the following three schemes are proposed to analyze and compare the energy storage capacity configuration results. The configuration results are shown in Table 1.

[0094] Option 1: Ignoring demand-side response, only consider installing energy storage equipment to smooth out fluctuations in the net load curve;

[0095] Option 2: Ignoring the impact of energy storage equipment, the fluctuation of the net load curve is smoothed only through demand-side response.

[0096] Solution 3: The method of the present invention comprehensively considers the joint configuration of demand-side response and energy storage equipment;

[0097] Table 1 Comparison of three schemes of hybrid energy storage and demand-side response configuration

[0098] As can be seen from Table 1, the proposed solution minimizes total system cost and minimizes active power fluctuations after stabilization. Comparing Solution 1 and Solution 3, Solution 3 (the proposed method) reduces total costs by 22.2% and active power fluctuations by 28.4% compared to Solution 1. The participation of demand-side response resources effectively reduces the capacity and power configuration of the hybrid energy storage system. In other words, the synergistic effect of flexible load and storage resources optimizes the system's economic efficiency, making it suitable for future large-scale access to distributed photovoltaics. Although Solution 2 offers the best economic performance, due to the strong subjectivity of user-side demand response, uncertainty becomes a key factor affecting the effectiveness of distributed photovoltaic fluctuation stabilization. Therefore, the market mechanism for demand-side response still needs to be improved.

[0099] 3) Analysis of hybrid and single energy storage configuration schemes: To explore the superiority of the proposed hybrid energy storage configuration scheme considering incentive-based demand-side response over the single energy storage scheme considering incentive-based demand-side response, the following three schemes are proposed to analyze and compare the energy storage capacity configuration results;

[0100] Option 4: Considering the impact of incentive-based demand-side response, the energy storage device is equipped with only supercapacitors;

[0101] Option 5: While considering the impact of incentive-based demand-side response, the energy storage equipment is equipped with only lithium iron phosphate batteries;

[0102] Scheme 6: The scheme proposed by the present invention;

[0103] The energy storage equipment configurations of the three schemes are shown in Table 2. The comparison of active power fluctuation values ​​is shown in Figure 8;

[0104] Table 2 Energy storage equipment configuration of three schemes: hybrid energy storage and single energy storage configuration

[0105] As can be seen from Figure 8, Scheme 6, the hybrid energy storage scheme proposed in the present invention, significantly reduces the system active power fluctuation value, and is also significantly improved compared to the schemes 4 and 5 that configure energy storage separately. Comparing the hybrid energy storage configuration schemes and the single energy storage configuration schemes in Schemes 4, 5, and 6, it can be seen from Table 2 that the total cost of the hybrid energy storage configuration scheme (Scheme 6) using the adaptive VMD algorithm is reduced by 19.2% and 3.8% respectively compared to the single energy storage schemes (Schemes 4 and 5). This is because the system using a single energy storage device needs to bear all power fluctuations in order to smooth the fluctuations of the net load curve, which aggravates the single energy storage. The number of charge and discharge cycles and the depth of discharge of the equipment increase the loss of the equipment's service life and increase the cost of replacing the energy storage equipment. However, in the hybrid energy storage capacity configuration using the adaptive VMD algorithm, the fast response and high power density operating characteristics of supercapacitors improve the operating environment of the lithium iron phosphate battery energy storage system, reduce the high-frequency power component borne by the lithium iron phosphate battery, reduce the number of replacements, and increase its operating life. Although the configuration of supercapacitors will increase the investment and economic cost of the energy storage system, its investment and economic cost is far less than the equipment renewal cost increased by the frequent replacement of lithium iron phosphate battery energy storage.

[0106] appendix

[0107] Appendix 1 Node power size

[0108] Appendix 2 Branch impedance parameters

[0109] Appendix 3 Energy storage system related parameters

[0110] Appendix 4: Power Purchase Price and Incentive Demand Response Compensation Price

[0111] The present invention is a hybrid energy storage optimization configuration strategy for low-voltage distribution network considering incentive-type demand response. During use, the present invention considers the characteristics of energy storage equipment and incentive-type demand-side response resources at different time scales. First, the improved VMD algorithm is used to decompose and reconstruct the system net load curve. The high-frequency fluctuations are smoothed by supercapacitors and directly controlled loads, and the low-frequency part is smoothed by mains power, lithium iron phosphate batteries and non-directly controlled loads. Then, an optimization configuration model of the hybrid energy storage system considering incentive-type demand-side response is established, and the improved whale optimization algorithm is used to solve the optimization configuration model. Finally, an actual low-voltage distribution network station in a certain place in northern my country is used for example verification. The results show that the model proposed by the present invention has the following advantages: ① Using the gray wolf algorithm The improved VMD algorithm has better solution speed and calculation accuracy than the VMD algorithm improved by the genetic algorithm, indicating that the proposed method has certain advantages. ② Compared with the solution of only configuring a hybrid energy storage system without the participation of demand-side response resources, the proposed solution saves 22.2% of the total cost and reduces the active power fluctuation value by 28.4%. That is, the participation of demand-side response resources effectively reduces the capacity and power configuration of the hybrid energy storage system. The synergistic effect of flexible resources in both load and storage enables the system to achieve optimal economic efficiency, which is suitable for the future large-scale access of distributed photovoltaic and load. ③ The total cost of the hybrid energy storage configuration solution using the adaptive VMD algorithm is reduced by 19.2% and 3.8% respectively compared with the solution of configuring a single supercapacitor energy storage solution and a single lithium iron phosphate battery energy storage solution. The fast response and high power density operating characteristics of the supercapacitor improve the operating environment of the lithium iron phosphate battery energy storage system, reduce the high-frequency power component borne by the lithium iron phosphate battery, and reduce the number of replacements, thereby achieving the purpose of delaying equipment investment. The present invention has the advantages of fully utilizing the complementary characteristics of resources, maximizing the adjustment capacity of flexible resources, effectively smoothing the fluctuation of the net load curve, fast solution speed, and high calculation accuracy.

Claims

1. A hybrid energy storage optimization configuration strategy for low-voltage distribution network considering incentive-based demand response, characterized by: The method comprises the following steps: Step 1: Consider the PV-storage system with incentive-based demand-side response; Step 2: Considering the hybrid energy storage system configuration strategy of incentive-based demand-side response; Step 3: Optimal configuration model of hybrid energy storage system considering incentive-based demand-side response; Step 4: Demonstration application analysis.

2. The low-voltage distribution network hybrid energy storage optimization configuration strategy considering incentive-based demand response as claimed in claim 1, characterized in that: The configuration strategy of the hybrid energy storage system considering the incentive-type demand-side response in step 2 is specifically as follows: the configuration strategy of the hybrid energy storage system, i.e., HESS, includes both its location configuration and its rated capacity and rated power configuration; in order to reasonably allocate the lithium iron phosphate batteries and supercapacitors in the HESS, the net load curve of the system is firstly decomposed and reconstructed, and the high-frequency fluctuations are smoothed by supercapacitors and the direct-controlled loads in the incentive-type demand-side response, i.e., IDR, and the low-frequency part is smoothed by the mains, lithium iron phosphate batteries, and the non-direct-controlled loads in the IDR, specifically including the following steps: Step 2.1: Improved variational mode algorithm; Step 2.2: Hybrid energy storage system power allocation; Step 2.3: Hybrid energy storage system capacity allocation; Step 2.4: Hybrid energy storage system installation location.

3. The low-voltage distribution network hybrid energy storage optimization configuration strategy considering incentive-based demand response as claimed in claim 2, characterized in that: The improved variational modal algorithm in step 2.1 decomposes and transforms the net load curve and then recombines and reconstructs it, including the following steps: Step 2.11: The optimal decomposition scale K is obtained by optimizing the Grey Wolf Algorithm, and the reference power signal x(t) of the net power curve is decomposed to obtain the modal components. The calculation formula is: Where x(t) is the power signal to be decomposed; u k (t) is the corresponding modal component; K is the number of decomposed modes; Step 2.12: Use Hilbert transform to transform each modal component u k (t) is converted into its instantaneous amplitude and frequency phase, and the calculation formula is: Step 2.13: Using the values ​​obtained in step 2.12, construct each modal component u k The analytical signal in polar coordinate form of (t) is calculated as: In the formula, z k (t) is u k (t) in polar coordinate form; α k (t) is the amplitude of the kth mode function; φ k (t) is the center frequency of the kth mode function; u of each modal component k (t) is the instantaneous frequency w k (t) is: The corresponding Hilbert spectrum is obtained as follows: Step 2.14: The Hilbert amplitude spectrum is composed of the real part of the analytical signal. The Hilbert marginal spectrum can be obtained by integrating the amplitude spectrum over time: Step 2.15: The Hilbert marginal spectrum can integrate the amplitude distribution of each modal component at different frequency scales. Finally, the high- and low-frequency boundary points are determined by the marginal spectrum. Then, the high- and low-frequency modal components are reconstructed and smoothed by HESS, IDR and mains respectively: P net (t) = P load (t)-P pv (t) = P high (t)+P low (t)(8), Where P net (t) is the net load power at time t; P load (t) is the load power at time t; P pv (t) is the distributed photovoltaic output at time t; P high (t), P low (t) are the high-frequency power and low-frequency power at time t respectively; P sc (t), P b (t) are the power of supercapacitor and lithium iron phosphate battery at time t; P DCL (t), P NDCL (t) are the actual response quantities of direct-controlled load and non-direct-controlled load at time t respectively; P buy (t) is the power provided by the mains at time t; a is the number of high-frequency modes in the system; K is the total number of modes; u k (t) is the corresponding modal component.

4. The low-voltage distribution network hybrid energy storage optimization configuration strategy considering incentive-based demand response as claimed in claim 2, characterized in that: The power allocation of the hybrid energy storage system in step 2.2 is specifically as follows: to determine the rated power and rated capacity of each device in the HESS and the power of the direct-controlled and non-direct-controlled loads in the IDR, the output ratio coefficients α and β are set respectively, and α+β=1 is satisfied, then the rated power of each device is as follows: Where α is the proportional coefficient of the output required by the HESS; β is the proportional coefficient of the output required by the IDR part; the meanings of the other variables are the same as those of formulas (8)-(10); in addition, since the energy storage system has certain losses during the charging and discharging process, the charging and discharging efficiency of the energy storage is further considered in the rated power calculation. Finally, the actual rated powers of the lithium iron phosphate battery and the supercapacitor are: In the formula, P′ SC , P′ b are the actual rated powers of supercapacitor and lithium iron phosphate battery respectively; They are respectively the charging and discharging power signals of the supercapacitor at time t; are respectively the charging and discharging power signals of the lithium iron phosphate battery at time t; η c,SC , η d,SC is the charge and discharge efficiency of the supercapacitor; η c,b , η d,b is the charge and discharge efficiency of lithium iron phosphate battery.

5. The low-voltage distribution network hybrid energy storage optimization configuration strategy considering incentive-based demand response as claimed in claim 2, characterized in that: The capacity allocation of the hybrid energy storage system in step 2.3 is specifically as follows: according to the actual charging and discharging conditions of the supercapacitor and the lithium iron phosphate battery, the integration is performed within a time period T to obtain the cumulative capacity of the two within the period: In the formula, E SC0 (t), E b0 (t) are the cumulative capacities of the supercapacitor and lithium iron phosphate battery in the cycle; considering that the energy storage element cannot be overcharged or over-discharged and has SOC state restrictions, the maximum and minimum values ​​of the cumulative capacity are taken to correct the above formula. Finally, the rated capacities of the supercapacitor and lithium iron phosphate battery are: In the formula, SOC up,SC , SOC down,SC SOC is the upper and lower limit constraints of the supercapacitor state of charge; up,b , SOC down,b These are the upper and lower limit constraints for the state of charge of the lithium iron phosphate battery.

6. The low-voltage distribution network hybrid energy storage optimization configuration strategy considering incentive-based demand response as claimed in claim 2, characterized in that: The installation location of the hybrid energy storage system in step 2.4 is specifically as follows: There are two location configuration modes for energy storage on the low-voltage side of the traditional distribution network, namely, decentralized layout and centralized layout; decentralized layout is to configure energy storage on-site near the distributed photovoltaics according to the installation location of the user's rooftop photovoltaics; centralized layout is to configure the energy storage equipment at the public connection point and configure appropriate capacity according to the total amount of distributed photovoltaics; The energy storage equipment distributed on the user side has the advantages of flexible access location and fewer restrictions on the location of the access point and environmental natural conditions. However, due to the distributed arrangement close to the user side, the controllability is relatively weak and the operation and maintenance is more difficult. The centralized energy storage has the advantage of good aggregation effect, but the corresponding line loss will also be larger. Taking into account the above two arrangements, as well as the control methods, operation and maintenance difficulties, and economic benefits of distributed photovoltaics in actual operation, a distributed arrangement of energy storage systems is adopted for low-voltage distribution networks that present a radial form.

7. The low-voltage distribution network hybrid energy storage optimization configuration strategy considering incentive-based demand response as claimed in claim 1, characterized in that: The hybrid energy storage system optimization configuration model considering incentive-type demand-side response in step 3 is specifically as follows: with the advancement of rooftop photovoltaic policies, more and more distributed photovoltaics are connected to the low-voltage distribution network nearby. The randomness and volatility of their output have aggravated the volatility of the net load curve, seriously affecting the normal operation of the low-voltage power grid; in order to smooth the fluctuation of the net load curve, a hybrid energy storage system optimization configuration model considering incentive-type demand-side response is constructed; since demand response resources naturally exist on the user side of the system, their utilization is only considered at the operation level, and at the planning level, it is impossible to increase the available demand response resources in the system by increasing investment; the incentive-type demand response resource type considers interruptible loads, which specifically includes the following steps: Step 3.1: Objective function; Step 3.2: Constraints; Step 3.3: Improve the whale algorithm.

8. The low-voltage distribution network hybrid energy storage optimization configuration strategy considering incentive-based demand response as claimed in claim 7, characterized in that: The objective function in step 3.1 is specifically as follows: the objective function is to minimize the total system cost and active power fluctuation value, and the decision variables are the power and capacity of the supercapacitor and lithium iron phosphate battery and the actual response active power of the direct-controlled load and the non-direct-controlled load participating in the demand-side response; The objective function calculation formula is: min F = λ1F1 + λ2F2 (19), min F1 = C SC +C b +C IDR +C buy (20) Where F1 is the total system cost, which is composed of the investment cost C of the energy storage equipment SC , Operation and maintenance cost C b , Compensation cost of incentive-based demand-side response C IDR And the cost of electricity purchased by the upper power grid C buy composition; F2 is the system active power fluctuation value; λ1 and λ2 are the weights of the two sub-objective functions F1 and F2, which can be obtained through the hierarchical analysis method, and considering that the two sub-objectives have different dimensions, they must be normalized during the calculation process; K is the number of all load nodes; T is 24 hours a day; α is the discount rate; T SC , T b k is the service life of the supercapacitor and lithium iron phosphate battery; SC1 , k SC2 , k SC3 and k b1 , k b2 , k b3 are the unit power cost, unit capacity cost and unit capacity operation and maintenance cost of supercapacitor and lithium iron phosphate battery respectively; P SC , P b are the power of supercapacitor and lithium iron phosphate battery respectively; E SC 、E b are the capacities of supercapacitor and lithium iron phosphate battery respectively; N is the number of users participating in demand-side response; P i,t and P i,t-1 are the active power values ​​at time t and time t-1 respectively; c DCL 、c NDCL are the unit compensation electricity prices for direct-controlled load and non-direct-controlled load respectively; c is the time-of-use electricity price; P DCL,i (t), P NDCL,i (t) are the actual response quantities of the direct-controlled load and the non-direct-controlled load at node i at time t; P t It is the power purchased by the low-voltage distribution network from the upper power grid at time t.

9. The low-voltage distribution network hybrid energy storage optimization configuration strategy considering incentive-based demand response as claimed in claim 7, characterized in that: The constraints in step 3.2 are as follows: in addition to the most basic power flow constraints, it is also necessary to meet distributed photovoltaic grid-connected constraints, hybrid energy storage charging and discharging constraints, hybrid energy storage SOC constraints, power balance constraints, demand-side response capability constraints, and power constraints with the superior contact point; among which ① distributed photovoltaic grid-connected constraints are: combined with the restrictions on grid-connected fluctuations of renewable energy at different time scales, for any time period t, the distributed photovoltaic volatility constraints are: under short-time-scale high-frequency fluctuations: Under long-time-scale low-frequency fluctuations: In the formula, are the maximum fluctuation power at short time scale and long time scale respectively; P cap (t) is the installed capacity of distributed photovoltaic power generation; ② The hybrid energy storage charging and discharging constraints are: when the remaining capacity of the lithium battery and the supercapacitor cannot meet the charging and discharging power at the next moment, the charging and discharging power of the lithium iron phosphate battery and the supercapacitor should be adjusted: Where η b is the charge and discharge efficiency of lithium iron phosphate battery; η SC is the charging and discharging efficiency of the supercapacitor; ③ The SOC constraint of the hybrid energy storage is: SOC b,min ≤SOC b (t)≤SOC b,max (26), SOC SC,min ≤SOC SC (t)≤SOC SC,max (27), where SOC b (t), SOC SC (t) is the state of charge of the lithium iron phosphate battery and supercapacitor at time t; SOC b,min , SOC b,max The lower and upper limits of the state of charge of the lithium iron phosphate battery; SOC SC,min , SOC SC,max are the lower and upper limits of the supercapacitor state of charge respectively; ④The power balance constraint is: The meaning of each variable in the formula is the same as that of formula (22); ⑤ The demand-side response capacity constraint is that the response capacity of the load participating in the demand response during operation also has certain constraints. The maximum response capacity constraint of DCL and NDCL is: 0≤P DCL,i (t)≤M DCL,i (t)(29), 0≤P NDCL,i (t)≤M NDCL,i (t)(30), where M DCL,i 、M NDCL,i is the maximum response of DCL and NDCL at node i in period t; ⑥ The power constraint with the upper contact point is: 0≤P(t)≤P max (31), where P(t) is the real-time power of the distribution transformer area and the upper power grid connection point; P max is the upper limit of the contact point interaction power.

10. The low-voltage distribution network hybrid energy storage optimization configuration strategy considering incentive-based demand response as claimed in claim 7, characterized in that: The improved whale algorithm in step 3.3 specifically includes the following steps: Step 3.31: Input photovoltaic and load timing output curve parameters and algorithm-related parameters; Step 3.32: Generate the initial whale population by combining the cubic chaos mapping method and calculate the objective function value at the same time; Step 3.33: Perform a fast non-dominated sorting of the position of each individual whale in the algorithm, i.e., the power and capacity of the supercapacitor and lithium iron phosphate battery, and the actual response size of the direct-controlled load and the non-direct-controlled load participating in the demand-side response; Step 3.34: The position of each individual whale is optimized through three steps: encirclement search, bubble attack, and random mutation, and the position of the individual whale is updated; Step 3.35: Determine whether the maximum number of iteration terminations has been reached. If not, return to step 3.32; if reached, output the final optimization solution.

Images