An Optimization Control Method and System for a DC Microgrid Hybrid Energy Storage System

Abstract

The present invention claims protection for an optimized control method and system for a DC microgrid hybrid energy storage system, belonging to the field of industrial control. First, a DC microgrid structure containing a hybrid energy storage device is established, and based on this, a DC microgrid prediction model is established. Then, according to the characteristics of the energy storage device and the conditions for the safe operation of the system, the constraint conditions of the system are set. Finally, an objective function for DC microgrid energy management is established. By determining the indicators to be optimized in the system, an energy management strategy for the DC microgrid system is designed. The present invention uses a two-layer structured model predictive control (MPC) to solve the control and constraints in the operation of the DC microgrid, and provides control signals for each response unit in the system; the present invention sets reasonable constraint conditions for the operation of the energy storage device to extend the service life of the energy storage device and weaken the negative impact on the operation efficiency to a certain extent; the energy management strategy proposed by the present invention can effectively balance the energy fluctuations during load mutations and ensure the stability of the system.

Description

Technical Field

[0001] The present invention belongs to the field of industrial control, and particularly relates to the application of double-layer MPC in the energy management of DC microgrids. Background Art

[0002] In recent years, many countries including China have put forward the strategic goal of "carbon neutrality", requiring a low-carbon transformation in the use of energy. Among them, the high proportion of penetration of renewable energy sources (RES) is crucial, and a complementary development pattern has been formed among various energy systems.

[0003] There are still problems such as intermittency and uncertainty in the development of RES, and the microgrid technology has emerged accordingly. The advantages of the DC microgrid studied in the present invention are reflected in that when distributed power sources are connected to the microgrid, the number of power changes is small, the loss is small, and the converter control is simple. Problems such as phase synchronization that need to be considered in AC power grids do not need to be considered in DC microgrids, which is more beneficial for the optimal management of energy. Energy storage devices in DC microgrids play a very important role in the research of energy management. The energy storage device responds quickly to changes in load and compensates for the power imbalance caused by load changes; during island operation, the energy storage device can absorb and release energy at appropriate times to maintain the stability of the microgrid. However, in actual industrial control, there are problems such as limited lifespan and high cost of energy storage devices, and strict constraints need to be set to ensure the effective operation of energy storage devices. If not reasonably utilized, the service life of energy storage devices will be greatly shortened and the operation efficiency will be unsatisfactory. Therefore, the present invention aims to solve the problem of formulating a more reasonable energy management strategy for DC microgrids, which is more conducive to realizing the optimal scheduling of energy and ensuring the safety and stability of the system during the operation process.

[0004] In recent years, research scholars have explored and studied DC microgrid systems containing energy storage devices. Some studies on DC microgrid energy storage mention using a method combining heuristic rules and particle swarm optimization algorithms to optimize the charging and discharging of energy storage, improving the practicality of the algorithm. However, such algorithms have a long calculation time and are prone to falling into local optima, and are not suitable for operating complex optimization systems. Or, with the minimum operating cost as the objective function, fuzzy control and genetic algorithms are respectively used to achieve the optimal control of the system energy flow. However, the above control strategies need to preset the operation mode of the energy storage system according to the predicted output of RES, and the energy scheduling is greatly affected by the prediction error. There are also those who use MPC to coordinate the operation of the energy management system, which can solve the problem of RES output fluctuations, but there are still limitations for multi-objective optimization problems.

[0005] In view of the above problems, the present invention provides an optimized control strategy for a DC microgrid hybrid energy storage system to solve the above problems. The double-layer MPC control based on the state space model adopted by the present invention can effectively solve the problem of the prediction accuracy of RES power; the present invention divides multiple priorities according to different importance levels, and uses the redundant degrees of freedom to achieve the purpose of economic optimization while ensuring the optimization of system energy scheduling. Summary of the Invention

[0006] The present invention aims to solve the problems of the above prior art. An optimized control method and system for a DC microgrid hybrid energy storage system are proposed. The technical solution of the present invention is as follows:

[0007] An optimized control method for a DC microgrid hybrid energy storage system, comprising the following steps:

[0008] S1: Construct a prediction model for the DC microgrid hybrid energy storage system: A microgrid is composed of a renewable distributed power source, a hybrid energy storage device of a battery and a supercapacitor, and a load, and energy transfer is carried out with the external power grid through the connection point PCC. The photovoltaic operates under the maximum power tracking control to obtain a linearized prediction model of the DC microgrid. Substitute the initial SOC S (k), SOC B (k) and u(k) into the DC microgrid prediction model to obtain the open-loop prediction value at the next moment. Among them, SOC B (k), SOC S (k) are the state of charge of the battery and the supercapacitor respectively, and u(k) represents the input power;

[0009] S2: Steps for calculating the steady-state target of SSTC: First, set the soft constraints and hard constraints of SSTC; then set the priority calculation to ensure the normal operation of the energy storage device and the economic operation of the system; finally, calculate the steady-state target value at time k according to the constraint conditions and the value at the previous moment;

[0010] S3: Dynamic control step: Set the target to be achieved by the DC microgrid energy management optimization strategy and set the objective function accordingly to make the system operate within the constraints.

[0011] Further, the specific steps of constructing the DC microgrid hybrid energy storage system prediction model in step S1 include:

[0012] Step S11: According to the DC microgrid structure, determine the operating variable u(k), the state matrix x(k) and the controlled variable y(k), and combine the characteristics of the hybrid energy storage device to obtain the relationship between the state of charge and power of the energy storage device at adjacent moments, and finally obtain the DC microgrid prediction model;

[0013] Step S12: Substitute the initial value into the prediction model to calculate the external target value of the double-layer MPC control.

[0014] Furthermore, the obtaining of the DC microgrid prediction model specifically includes:

[0015] At sampling time k, the control variables of the system are taken as: u(k) = [P G (k), P B (k), P S (k)] Τ , and its incremental form is: Δu(k) = [ΔP G (k), ΔP B (k), ΔP S (k)],

[0016] The state variable matrix is: x(k) = [P G (k), P B (k), P S (k), SOC B (k), SOC S (k)] Τ ,

[0017] The output variable matrix is: y(k) = [P M (k), SOC B (k), SOC S (k)] Τ

[0018] Among them, P G (k) is the exchanged power between the external power grid and the bus, P B (k) is the output of the battery, P S (k) is the output of the supercapacitor, SOC B (k), SOC S (k) are the state of charge of the battery and the supercapacitor respectively, and P M (k) = P G (k) + P B (k) + P S (k) represents the power provided by other energy storage devices and the external power grid in the DC microgrid except for the photovoltaic power.

[0019] The relationship between the state of charge and power at adjacent sampling times is:

[0020] SOC B (k) = SOC B (k - 1) - P B (k - 1)Δt / E B

[0021] SOC S (k) = SOC S (k - 1) - P S(k - 1)Δt / E S

[0022] Rewritten in incremental form as:

[0023] P B (k) = δSOC B (k) * E B / Δt + ΔP B (k)

[0024] P S (k) = δSOC S (k) * E S / Δt + ΔP S (k)

[0025] Among them, E B 、E S are the capacities of the battery and the supercapacitor; Δt is the sampling step size;

[0026] The linearized prediction model of the DC microgrid can be written as:

[0027]

[0028]

[0029] Furthermore, the steps of calculating the steady - state target SSTC in step S2 specifically include:

[0030] Step S21: Determine the soft constraints and hard constraints: Assume that the ideal charge - discharge interval is a soft constraint, and the maximum charge - discharge interval is a hard constraint. Respectively determine the soft and hard constraints of the battery, supercapacitor output, and external grid power at the (k + i) - th moment. In addition, determine the soft and hard constraints of the state of charge of the battery and supercapacitor at the (k + i) - th moment;

[0031] Step S22: Determine the priority: The priority calculation considers two aspects: One is to place the state of charge of the battery and supercapacitor in the first priority to ensure that the battery and supercapacitor operate within the normal range; the other is to place the power P G of the external grid at the lowest priority;

[0032] Step S23: Considering the steady - state objective function, under the constraint conditions, combined with the SOC S 、SOC B and u solved at the previous moment, obtain the steady - state target values and u * (k).

[0033] Furthermore, set the soft constraints and hard constraints for the steady - state target calculation, the ideal charge - discharge interval soft constraint and the maximum charge - discharge interval hard constraint, specifically including:

[0034]

[0035] Among them, P S (k + i|k) represents the power of the supercapacitor predicted i steps at time k, and the soft constraint upper and lower bounds of this power are P S,O and The hard constraint upper and lower bounds are P S,h and P B (k + i|k) represents the power of the battery predicted i steps at time k, and the soft constraint upper and lower bounds of this power are P B,o and The hard constraint upper and lower bounds are P B,h and P G (k + i|k) represents the power of the external power grid predicted i steps at time k, and the soft constraint upper and lower bounds of this power are P G,o and The hard constraint upper and lower bounds are P G,h and

[0036] At the same time, it satisfies the power balance equation constraint:

[0037] P M (k) = P N (k), k ≥ 0

[0038] In the formula, P N (k) = P L (k) - P V (k) represents the net load power;

[0039] The ideal range and maximum working range of the SOC values of the supercapacitor and the battery are:

[0040]

[0041] Among them, SOC S (k + i|k) represents the state of charge of the supercapacitor predicted i steps at time k, and the soft constraint upper and lower bounds of this state are SOC S,o and The hard constraint upper and lower bounds are SOC S,h and

[0042] There are also incremental limits for the charge and discharge power:

[0043]

[0044]

[0045]

[0046] Among them, ΔP S (k + i|k) represents the power increment of the supercapacitor predicted i steps at time k, and the upper and lower bounds of this power increment are Δ P ′ S and ΔP B (k + i|k) represents the power increment of the battery predicted i steps at time k, and the upper and lower bounds of this power increment are Δ P ′ B and ΔP G (k + i|k) represents the power increment of the battery predicted i steps at time k, and the upper and lower bounds of this power increment are Δ P ′ G and

[0047] Furthermore, in order to calculate the state of charge at the current moment and the steady-state target value of the input control variable, let δSOC(k) = [δSOC B (k), δSOC S (k)] T , δSOC B (k), δSOC S (k) respectively represent the changes in the state of charge of the supercapacitor and the battery, and δSOC(k) is the state-of-charge vector composed of the changes in the state of charge of the two energy storage devices. Consider the following objective function:

[0048]

[0049] Under its constraint conditions, solve for δSOC B (k), δSOC S (k) and Δu(k). Combining the SOC S (k - 1), SOC B (k - 1) and u(k - 1) that have been solved at the previous moment, the steady-state target values of SOC S (k), SOC B (k) and u(k) are:

[0050]

[0051] u * (k) = u(k - 1) + Δu(k)

[0052] So far, the steady-state target calculation described in step S2 is completed.

[0053] Furthermore, the dynamic control process of step S3 is as follows:

[0054] Set three objectives for the optimization strategy of DC microgrid energy management:

[0055] (1) Make the future controlled variable y(k) as close as possible to the calculated steady-state target value;

[0056] (2) Suppress the drastic change of u(k);

[0057] (3) In the case of no solution, obtain a feasible solution by relaxing the soft constraints;

[0058] The objective function is as follows:

[0059]

[0060] P M and P N respectively represent the power provided by other energy storage devices and the external power grid and the net load power in the DC microgrid except for the photovoltaic power. Among them, Q, R, and L are weight coefficients, SOC(k)* is the ideal value of the state of charge, and SOC(k) includes SOC B (k) and SOC S (k). Specifically,

[0061] At this time, the exchange power P G (k) between the external power grid and the bus, the output power P B (k) of the battery, and the output power P S (k) of the supercapacitor can be obtained. At the same time, SOC B (k) and SOC S (k) are constrained within the ideal operating range.

[0062] An optimized control system for a DC microgrid hybrid energy storage system based on any one of the above methods, which includes:

[0063] DC microgrid hybrid energy storage system prediction model: A microgrid is composed of a renewable distributed power source, a hybrid energy storage device of a battery and a supercapacitor, and a load. Energy transfer is carried out with the external power grid through the connection point PCC. The photovoltaic operates under the maximum power tracking control to obtain a linearized prediction model of the DC microgrid. Substitute the initial SOC S (k), SOC B (k) and u(k) into the DC microgrid prediction model to obtain the open-loop prediction value at the next moment, where SOC B (k) and SOC S (k) are the states of charge of the battery and the supercapacitor respectively, and u(k) represents the input power;

[0064] Steady-state target SSTC calculation module: It is used to first set the soft constraints and hard constraints of SSTC; then set the priority calculation to ensure the normal operation of the energy storage device and the economic operation of the system; finally, calculate the steady-state target value at time k according to the constraint conditions and the value at the previous moment.

[0065] Dynamic control module: It is used to set the goals that need to be achieved by the DC microgrid energy management optimization strategy and set the objective function accordingly, so that the system operates within the constraints.

[0066] The advantages and beneficial effects of the present invention are as follows:

[0067] The present invention proposes an optimized control strategy for a DC microgrid hybrid energy storage system based on a two-layer MPC. For the specific two-layer MPC structure diagram, see Figure 2 , including: a DC microgrid prediction model that calculates the initial prediction value of the system; a steady-state target calculation (SSTC) module that can perform local economic optimization and track the prediction value by setting the constraint conditions of MV and CV and setting the optimization performance index; dynamic control that dynamically tracks the optimized setting value obtained by the steady-state target calculation.

[0068] The two-layer MPC adds SSTC compared with the conventional MPC, which can enable the system to play a better control role and economic advantage. First, the DC microgrid prediction model gives the prediction value as the external target for the steady-state target calculation. SSTC is equivalent to completing a feedforward control in the two-layer MPC. On the one hand, it can pre-adjust the external target; on the other hand, when there are disturbances or emergencies during the system operation, it can also be adjusted from the steady-state perspective according to the feedback information. This is reflected in the present invention that when the load changes, the energy storage transposition responds quickly, reasonably distributes the power of the battery and the supercapacitor, and makes the system reach stability. In addition to the above-mentioned advantages of SSTC, it can also use the redundant operation degrees of freedom for economic optimization. This is reflected in the present invention that when setting the priority calculation, the external grid power is set as the lowest priority to achieve economic optimization. Compared with the traditional MPC, the two-layer MPC can improve the economic benefits of the system operation while ensuring the stable operation of the system. Brief Description of the Drawings

[0069] Figure 1 is a schematic structural diagram of a DC microgrid hybrid energy storage system provided by a preferred embodiment of the present invention;

[0070] Figure 2 is a block diagram of the two-layer MPC structure provided by an embodiment of the present application;

[0071] Figure 3 is a flowchart of the two-layer control provided by an embodiment of the present application. Detailed Embodiments

[0072] The technical solutions in the embodiments of the present invention will be clearly and detailedly described below with reference to the accompanying drawings in the embodiments of the present invention. The described embodiments are only a part of the embodiments of the present invention.

[0073] The technical solution for the present invention to solve the above technical problems is:

[0074] In the research of DC microgrid energy management, single-layer MPC is a common control method. However, this method does not include SSTC, which makes it very difficult to achieve zero steady-state error for multivariable constraint tracking control.

[0075] To solve the above problems, this application proposes an optimized control strategy for DC energy management based on double-layer MPC. The double-layer MPC proposed in the invention includes SSTC, which can ensure zero steady-state error characteristics in practical applications. In the present invention, SSTC simultaneously gives the steady-state targets of MV and CV, which is used for the optimization when the system is operating and in the "borderline" state. Here, the so-called "borderline" means that the optimal operating point is located at the boundary of the constraint. In addition, the double-layer MPC also has an economic optimization function, which is an important factor that cannot be ignored for the energy management system.

[0076] Based on this, this application provides an optimized control strategy for the DC microgrid hybrid energy storage system. This method is based on double-layer MPC, combines the complementary advantages of supercapacitors and batteries, and controls the system energy, which can improve the stability of the system, ensure that the battery and supercapacitor work within a reasonable range, and can also improve the economic benefits of the system. Specifically, a prediction model of the system is designed based on the state space model to obtain the target tracking value of the steady-state target calculation module, and optimization is carried out for economic purposes. Finally, the dynamic control module completes the control action according to the steady-state target value.

[0077] Specifically, the establishment process of the DC microgrid prediction model described in step S1 is as follows:

[0078] Refer to Figure 1 , Figure 1 which is a schematic diagram of the DC microgrid structure provided by an embodiment of this application. In the structure diagram, P V is the photovoltaic output, which is the main power source, meets the load demand, and can charge the energy storage device. P B , P S represent the outputs of the battery and the supercapacitor, P L represents the power of the load, P G represents the power of the external power grid. PCC is the connection point, and the microgrid can use PCC to exchange energy with the external power grid. When the energy storage system is saturated and the microgrid has surplus energy, it can be sold to the external power grid through PCC, corresponding to P Gis negative, or electrical energy can be directly purchased from the external power grid through the DC microgrid system, corresponding to P G is positive. When P L is too large, the hybrid energy storage system discharges to supplement the insufficient photovoltaic output. At this time, P B and P S are positive.

[0079] At the sampling moment k, the control variables of the system are taken as: u(k) = [P G (k), P B (k), P S (k)] Τ , and its incremental form is: Δu(k) = [ΔP G (k), ΔP B (k), ΔP S (k)],

[0080] The state variable matrix is: x(k) = [P G (k), P B (k), P S (k), SOC B (k), SOC S (k)] Τ ,

[0081] The output variable matrix is: y(k) = [P M (k), SOC B (k), SOC S (k)] Τ

[0082] Among them, P G (k) is the exchange power between the external power grid and the bus, P B (k) is the output of the battery, P S (k) is the output of the supercapacitor, SOC B (k), SOC S (k) are the state of charge of the battery and the supercapacitor respectively, P M (k) = P G (k) + P B (k) + P S (k)

[0083] The relationship between the state of charge and power at adjacent sampling moments is:

[0084] SOC B (k) = SOC B (k - 1) - P B (k - 1)Δt / E B

[0085] SOC S (k) = SOCS (k - 1)-P S (k - 1)Δt / E S

[0086] Rewritten in incremental form as:

[0087] P B (k)=δSOC B (k)*E B / Δt + ΔP B (k)

[0088] P S (k)=δSOC S (k)*E S / Δt + ΔP S (k)

[0089] Where, E B and E S are the capacities of the battery and the supercapacitor; Δt is the sampling step size.

[0090] The linearized prediction model of the DC microgrid can be written as:

[0091]

[0092]

[0093] Thus far, the prediction model of the DC microgrid described in step S11 is established.

[0094] Specifically, substituting the initial values SOC S (k), SOC B (k) and u(k) into the DC microgrid prediction model to obtain the open-loop prediction value at the next moment.

[0095] Specifically, the steady-state target calculation process described in step S2 is as follows:

[0096] Set the soft constraints and hard constraints for steady-state target calculation, the ideal charge and discharge interval (soft constraint) and the maximum charge and discharge interval (hard constraint):

[0097]

[0098] Simultaneously satisfy the power balance equation constraint:

[0099] P M (k)=P N (k), k≥0

[0100] In the formula, P N (k)=P L (k)-P V (k) represents the net load power.

[0101] The ideal range and maximum operating range of the SOC values of the supercapacitor and the storage battery are as follows:

[0102]

[0103] There are also increment limits for the charge and discharge power:

[0104]

[0105] Specifically, there are two main considerations in the priority calculation: one is to place SOC S and SOC B in the position of priority guarantee to ensure that the supercapacitor and the storage battery are within the normal operating range; the other is to place P G at the lowest priority to achieve economic optimization.

[0106] Specifically, in order to calculate the state of charge at the current moment and the steady-state target value of the input control variable, let δSOC(k) = [δSOC B (k), δSOC S (k)] T , and consider the following objective function:

[0107]

[0108] Under its constraint conditions, solve for δSOC B (k), δSOC S (k) and Δu(k). Combining the SOC S (k - 1), SOC B (k - 1) and u(k - 1) that have been solved in the previous moment, the steady-state target values of SOC S (k), SOC B (k) and u(k) can be obtained as:

[0109]

[0110] u * (k) = u(k - 1) + Δu(k)

[0111] So far, the steady-state target calculation described in step S2 is completed.

[0112] Specifically, the dynamic control process described in step S3 is as follows:

[0113] Assume that the DC microgrid energy management optimization strategy needs to achieve the following three goals:

[0114] (1) Make the future controlled variable y(k) as close as possible to the calculated steady-state target value

[0115] (2) Suppress the drastic change of u(k).

[0116] (3) In the case of no solution, obtain a feasible solution by relaxing the soft constraints.

[0117] The objective function is as follows:

[0118]

[0119] Where Q, R, and L are weight coefficients, SOC(k)* is the ideal value of the state of charge, and SOC(k) includes SOC B (k) and SOC S (k), specifically

[0120] At this time, the exchange power P G (k) between the external power grid and the bus, the output power P B (k) of the battery, and the output power P S (k) of the supercapacitor can be obtained. At the same time, SOC B (k) and SOC S (k) are constrained within the ideal operating range.

[0121] The systems, devices, modules, or units illustrated in the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0122] It should also be noted that the term "comprising", "including", or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, commodity, or device comprising a series of elements not only includes those elements but also includes other elements not explicitly listed, or elements inherent to such process, method, commodity, or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the existence of additional identical elements in the process, method, commodity, or device comprising the said element.

[0123] The above embodiments should be understood as being only used to illustrate the present invention and not to limit the protection scope of the present invention. After reading the content recorded in the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. An optimization control method for a DC microgrid hybrid energy storage system, characterized in that, it includes the following steps: S1: Construct a prediction model for the DC microgrid hybrid energy storage system: A microgrid is composed of a renewable distributed power source, a hybrid energy storage device of a battery and a supercapacitor, and a load, and energy transfer is carried out with the external power grid through the point of common coupling PCC. The photovoltaic operates under the maximum power tracking control to obtain a linearized prediction model of the DC microgrid. Substitute the SOC S (k), SOC B (k) and u(k) into the DC microgrid prediction model. SOC B (k), SOC S (k) are the state of charge of the battery and the supercapacitor respectively, and u(k) represents the input power to obtain the open-loop prediction value at the next moment; S2: Steps for calculating the steady-state target SSTC: Step S21: Determine the soft constraints and hard constraints: Assume that the ideal charge and discharge interval is a soft constraint, and the maximum charge and discharge interval is a hard constraint. Respectively determine the soft constraints and hard constraints of the battery, supercapacitor output, and external grid power at the (k + i)-th moment. In addition, determine the soft constraints and hard constraints of the battery and supercapacitor state of charge at the (k + i)-th moment; Step S22: Determine the priority: The priority calculation considers two aspects: one is to place the state of charge of the battery and the supercapacitor at the first priority to ensure that the battery and the supercapacitor operate within the normal range; the other is to place the power P of the external power grid G at the lowest priority; Step S23: Considering the steady-state objective function, under the constraint conditions, combined with the SOC S (k), SOC B (k) and u(k), obtain the steady-state objective value at time k and u * (k); S3: Dynamic control step: Set three goals for the DC microgrid energy management optimization strategy: (1) Make the future controlled variable y(k) as close as possible to the calculated steady-state target value; (2) Suppress the drastic change of u(k); (3) In the case of no solution, obtain a feasible solution by relaxing the soft constraints; The objective function is as follows: Among them, P M and P N respectively represent the power provided by other energy storage devices and the external power grid and the net load power in the DC microgrid except for the photovoltaic power; Q, R, and L are weight coefficients, SOC(k)* is the ideal value of the state of charge, and SOC(k) includes SOC B (k) and SOC S (k). Specifically, At this time, the exchange power P G (k) between the external power grid and the bus, the output power P B (k) of the battery, and the output power P S (k) of the super capacitor can be obtained. At the same time, SOC B (k) and SOC S (k) are constrained within the range of ideal operation.

2. According to the optimization control method for a DC microgrid hybrid energy storage system described in claim 1, characterized in that, the specific construction of the DC microgrid hybrid energy storage system prediction model in step S1 includes: Step S11: According to the DC microgrid structure, determine the operating variable u(k), state matrix x(k), and controlled variable y(k). Combine the characteristics of the hybrid energy storage device to obtain the relationship between the state of charge and power of the energy storage device at adjacent moments. Finally, obtain the DC microgrid prediction model; Step S12: Substitute the initial value into the prediction model to calculate the external target value of the two-layer MPC control.

3. According to the optimization control method for a DC microgrid hybrid energy storage system described in claim 2, characterized in that, the specific obtaining of the DC microgrid prediction model includes: At sampling instant k, the control variable of the system is taken as: u(k) = [P G (k), P B (k), P S (k)] Τ , and its incremental form is: Δu(k) = [ΔP G (k), ΔP B (k), ΔP S (k)], The state variable matrix is: x(k) = [P G (k), P B (k), P S (k), SOC B (k), SOC S (k)] Τ , The output variable matrix is: y(k) = [P M (k), SOC B (k), SOC S (k)] Τ Among them, P G (k) is the exchanged power between the external power grid and the bus, P B (k) is the output power of the battery, P S (k) is the output power of the supercapacitor, SOC B (k), SOC S (k) are the state of charge of the battery and the supercapacitor respectively, P M (k) = P G (k) + P B (k) + P S (k), P M (k) represents the power provided by other energy storage devices and the external power grid except for the photovoltaic power inside the DC microgrid; The relationship between the state of charge and power at adjacent sampling moments is: SOC B SOC(k) = B SOC(k - 1)-P B (k - 1)Δt / E B SOC S SOC(k) = S SOC(k - 1)-P S (k - 1)Δt / E S Rewrite it in incremental form as: P B (k) = δSOC B (k) * E B / Δt + ΔP B (k) P S P(k) = δSOC S P(k) * E S / Δt + ΔP S P(k) Among them, E B and E S are the capacities of the storage battery and the super capacitor; Δt is the sampling step size; The DC microgrid linearized prediction model can be written as:

4. According to the optimization control method for a DC microgrid hybrid energy storage system described in claim 1, characterized in that, Set the soft constraints and hard constraints for steady-state target calculation, including the ideal charge and discharge interval soft constraint and the maximum charge and discharge interval hard constraint, specifically including: Among them, P S (k + i|k) represents the power of the supercapacitor obtained by predicting i steps at time k. The soft constraint upper and lower bounds of this power are P S,O and The hard constraint upper and lower bounds are P S,h and P B (k + i|k) represents the power of the battery obtained by predicting i steps at time k. The soft constraint upper and lower bounds of this power are P B,o and The hard constraint upper and lower bounds are P B,h and P G (k + i|k) represents the power of the external power grid obtained by predicting i steps at time k. The soft constraint upper and lower bounds of this power are P G,o and The hard constraint upper and lower bounds are P G,h and Simultaneously satisfy the power balance equation constraint: P M P(k) = N P(k), k ≥ 0 where P N (k) = P L (k) - P V (k) represents the payload power; The ideal range and maximum working range of the SOC values of the supercapacitor and the battery are: Among them, SOC S (k + i|k) represents the state of charge of the supercapacitor predicted i steps at time k, and the soft constraint upper and lower bounds of this state are SOC S,o and The hard constraint upper and lower bounds are SOC S,h and There are also incremental limits for the charge and discharge power: where, ΔP S (k+i|k) represents the power increment of the supercapacitor predicted i steps at time k, and the upper and lower bounds of this power increment are Δ P ′ S and ΔP B (k+i|k) represents the power increment of the battery predicted i steps at time k, and the upper and lower bounds of this power increment are Δ P ′ B and ΔP G (k+i|k) represents the power increment of the battery predicted i steps at time k, and the upper and lower bounds of this power increment are Δ P ′ G and 5. According to the optimization control method for a DC microgrid hybrid energy storage system described in claim 4, characterized in that, To calculate the state of charge at the current moment and the steady-state target value of the input control variable, let δSOC(k) = [δSOC B (k), δSOC S (k)] T , where δSOC B (k) and δSOC S (k) respectively represent the changes in the state of charge of the supercapacitor and the battery, and δSOC(k) is the state-of-charge matrix composed of the changes in the state of charge of the two energy storage devices. Consider the following objective function: Solve for δSOC under its constraints B (k), δSOC S (k) and Δu(k), combined with the SOC S (k - 1) that has been solved at the previous moment, SOC B (k - 1) and u(k - 1), the SOC S (k) can be obtained. The steady-state target values of SOC B (k) and u(k) are: u * u(k) = u(k - 1)+Δu(k) At this point, the steady-state target calculation in step S2 is completed.

6. A DC microgrid hybrid energy storage system optimization control system based on the method according to any one of claims 1-5, characterized in that, it includes: Prediction Model of DC Microgrid Hybrid Energy Storage System: A microgrid is composed of a renewable distributed power source, a hybrid energy storage device of a battery and a supercapacitor, and a load. Energy transfer is carried out with the external power grid through the point of common coupling (PCC). The photovoltaic operates under the maximum power tracking control to obtain a linearized prediction model of the DC microgrid. Substitute the SOC S (k), SOC B (k) and u(k) values into the DC microgrid prediction model. SOC B (k), SOC S (k) are the state of charge of the battery and the supercapacitor respectively, and u(k) represents the input power to obtain the open-loop prediction value at the next moment; Steady-state target SSTC calculation module: used to first set the soft constraints and hard constraints of SSTC; then set the priority calculation to ensure the normal operation of the energy storage device and the economic operation of the system; finally, calculate the steady-state target value at the k-th moment according to the constraint conditions and the value of the previous moment; Dynamic control module: used to set the goals that need to be completed by the DC microgrid energy management optimization strategy and set the objective function accordingly, so that the system operates within the constraints.

Images