Optical storage hybrid parallel power supply system coordination control method based on transformer substation, electronic equipment and medium

Abstract

The invention discloses an optical storage hybrid parallel power supply system coordination control method based on a transformer substation, electronic equipment and a medium, and the method comprises the steps: collecting historical power supply data and a current operation state of a power grid system comprehensive model, and constructing an energy storage prediction optimization model, the micro-grid system comprehensive model comprises a photovoltaic system dynamic model, a super-capacitor energy storage system dynamic model and a storage battery energy storage system dynamic model; a minimum cost function based on physical constraints is constructed based on the energy storage prediction optimization model, and optimal control input is obtained; generating a power adjustment instruction based on the optimal control input; and based on the power regulation instruction, controlling the super-capacitor energy storage system to charge and discharge, and controlling the storage battery energy storage system to output power. According to the method, the control signal under the physical constraint is optimized by minimizing the cost function, so that the deviation between the actual output and the target output can be reduced, the dynamic response speed and the control precision of the system are improved, and the operation reliability of the system is ensured.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of photovoltaic micro-grid, and particularly relates to a substation-based photovoltaic storage hybrid parallel power system coordinated control method, an electronic device and a medium. BACKGROUND

[0002] With the growth of global energy demand and the emphasis on renewable energy, micro-grid, as an effective framework integrating distributed energy, energy storage units, power conversion systems and loads, has become increasingly important. Photovoltaic systems, as a highly efficient and clean energy supply system, have been widely used. However, the intermittency of renewable energy significantly increases the complexity of micro-grid control and management. In addition, photovoltaic micro-grid faces problems such as large power fluctuation and unstable bus voltage when operating in island mode. These problems seriously affect the power supply reliability and power quality of photovoltaic micro-grid.

[0003] In related technologies, it is difficult to adapt to fluctuating conditions in real time using traditional control strategies, resulting in shortened service life of energy storage elements and poor system stability. Therefore, there is an urgent need for a control strategy that can effectively coordinate the operation of photovoltaic cells, storage batteries and supercapacitors to improve the stability and reliability of photovoltaic micro-grid when operating in island mode. SUMMARY

[0004] The purpose of the present application is to construct a prediction optimization model to predict and optimize the future state of the power grid system comprehensive model and output the optimal control input, and to manage the power output of the energy storage battery and supercapacitor with the optimal control input as the starting point, to improve the dynamic response speed and control accuracy of the system and ensure the reliability of the system operation. The substation-based photovoltaic storage hybrid parallel power system coordinated control method.

[0005] In order to achieve the above object, the application provides a substation-based photovoltaic storage hybrid parallel power supply system coordination control method, which comprises the following steps: collecting historical power supply data and current operating state of a power grid system comprehensive model, and constructing an energy storage prediction optimization model; wherein the historical power supply data comprises power output data of a photovoltaic system in different weather and different time periods, voltage, current and capacity change data in the charging and discharging cycle process of a super capacitor energy storage system, and charging and discharging efficiency, state of charge change and self-discharge rate of a battery energy storage system; the current operating state comprises current irradiance and temperature of a photovoltaic panel, and current voltage, current and state of charge of the super capacitor and the battery; the power grid system comprehensive model comprises a photovoltaic system dynamic model, a super capacitor energy storage system dynamic model and a battery energy storage system dynamic model; a minimum cost function based on physical constraints is constructed based on the energy storage prediction optimization model to obtain optimal control input; power regulation instructions are generated based on the optimal control input; if the output power of the photovoltaic system decreases, the super capacitor energy storage system is controlled to discharge to compensate for the bus voltage based on the power regulation instructions, and the battery energy storage system is controlled to increase the output power; if the output power of the photovoltaic system increases, the super capacitor energy storage system is controlled to charge to absorb the excess power of the photovoltaic system, and the battery energy storage system is controlled to reduce the output power to maintain the stability of the bus voltage.

[0006] In an alternative embodiment, the microgrid system comprehensive model comprises: an alternating current side provided with an alternating current bus, a substation and an alternating current load for the substation; a direct current side provided with a direct current bus, a direct current load for the substation, a unidirectional DC / DC converter and a bidirectional DC / DC converter; a bidirectional AC / DC converter connected with the alternating current bus and the direct current bus; wherein the unidirectional DC / DC converter connects the direct current bus and a photovoltaic array, and the bidirectional DC / DC converter connects the super capacitor energy storage system and the battery energy storage system in parallel.

[0007] In an alternative embodiment, the bidirectional DC / DC converter is provided as at least two, one of which has three ports to connect the super capacitor energy storage system and the battery energy storage system in parallel, and at least one of which has two ports to connect another battery energy storage system, so that the super capacitor energy storage system is connected in parallel with a plurality of battery energy storage systems.

[0008] In an alternative embodiment, the alternating current side and the direct current side are respectively provided as two, and the two alternating current sides and the two direct current sides are respectively connected by connecting lines.

[0009] In one optional implementation, a physical constraint-based cost minimization function is constructed based on the energy storage predictive optimization model to obtain the optimal control input. Specifically, this includes: modeling the electronic circuit of the bidirectional DC / DC converter based on model predictive control, the dynamic model of the supercapacitor energy storage system, and the dynamic model of the battery energy storage system to obtain a continuous-time state-space model; discretizing and linearizing the continuous-time state-space model to obtain a discrete-time state-space model; obtaining the expected control output based on historical data and the discrete-time state-space model; and constructing the physical constraint-based cost minimization function to optimize the expected control output, thereby obtaining the optimal control input.

[0010] In one optional implementation, the continuous-time state-space model is discretized and linearized to obtain a discrete-time state-space model, specifically including: obtaining the Boost transition based on the state equations of the Boost circuit.

[0011] The continuous-time state-space model of the device; where the state equation of the Boost circuit is: In the formula, i L V is the inductor current. s V is the input voltage. b R is the output voltage of the converter. L For the load resistance, C o L is the output capacitor, and L is the converter inductance. For switching states, including 0 or 1; the continuous-time state-space model of the Boost converter's inductor current and output voltage is as follows:

[0012] In the formula, The rate of change of current, The output voltage change rate is used; the continuous-time state-space model is discretized based on the discrete-time state-space model framework to obtain the discrete-time state-space; wherein, the discrete-time state-space model framework is: In the formula, x is the system state variable, u is the control variable, Y(s) is the output variable, A and B are the state matrices, w is the system disturbance, N is the disturbance input matrix, C is the output observation matrix, and s is the sampling time; the discrete-time state space is: In the formula, i L (s+1) represents the inductor current at time s+1, V b (s+1) is the output voltage at time s+1, and T s Let u(s) be the sampling period and u(s) be the control input. Linearize the discrete-time state space to obtain the discrete-time state space model: In the formula, V b (k), i L(k) are steady-state values of the converter output voltage and inductor current, respectively, D is the main switch duty cycle, u b (k) is the auxiliary switch duty cycle.

[0013] In an alternative embodiment, the minimum cost function includes a charge-discharge cost function and a bidirectional DC-DC converter cost function, and the charge-discharge cost function is: wherein N c is a control layer, N p is a prediction layer, u is a future sequence of the input signal, y * is a reference trajectory, w is a manipulated variable, and λ is a weighting factor; and the minimum cost function of the bidirectional DC-DC converter is: wherein i ref (t+k|t) is a reference current at time t+k, i L (t+k) is an actual inductor current at time t+k, δ is a current tracking weight coefficient, Δu(t+k-1) is a control increment, and λ is a control smoothing weight coefficient.

[0014] In an alternative embodiment, the substation-based photovoltaic storage hybrid parallel power system coordination control method further includes: at each sampling time, obtaining a current actual state variable, wherein the current actual state variable includes an inductor current, a DC bus voltage, a battery / supercapacitor terminal voltage, and a state of charge; based on the minimum cost function and the control output expectation, solving a control sequence that minimizes the cost online to obtain an optimal control input, including a current time control amount, using the current actual state variable as an initial value within a prediction range; and inputting the current time control amount into the bidirectional DC / DC converter to enter a next sampling time, thereby realizing rolling optimization.

[0015] The application further provides an electronic device, including: at least one processor; a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute any one of the substation-based photovoltaic storage hybrid parallel power system coordination control methods.

[0016] The application further provides a medium storing a computer program, wherein the computer program is executed by a processor to implement any one of the substation-based photovoltaic storage hybrid parallel power system coordination control methods.

[0017] The application has the beneficial effects that: the control signal under physical constraints is optimized by minimizing the cost function, thereby reducing the deviation between the actual output and the target output, improving the dynamic response speed and control accuracy of the system, and ensuring the reliability of the system operation. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 A flow chart of the method for coordinating and controlling a photovoltaic and energy storage hybrid parallel power supply system based on a substation according to an embodiment of the present application;

[0019] Figure 2 A schematic diagram of a micro-grid system comprehensive model of the method for coordinating and controlling a photovoltaic and energy storage hybrid parallel power supply system based on a substation according to an embodiment of the present application;

[0020] Figure 3 A flow chart of the method for coordinating and controlling a photovoltaic and energy storage hybrid parallel power supply system based on a substation according to another embodiment of the present application;

[0021] Figure 4 A photovoltaic output voltage diagram in a simulation state of the method for coordinating and controlling a photovoltaic and energy storage hybrid parallel power supply system based on a substation according to an embodiment of the present application;

[0022] Figure 5 A direct current bus voltage diagram in a simulation state of the method for coordinating and controlling a photovoltaic and energy storage hybrid parallel power supply system based on a substation according to an embodiment of the present application;

[0023] Figure 6 A photovoltaic system output voltage and current diagram in a simulation state of the method for coordinating and controlling a photovoltaic and energy storage hybrid parallel power supply system based on a substation according to an embodiment of the present application;

[0024] Figure 7 A photovoltaic and hybrid energy storage system power output diagram in a simulation state of the method for coordinating and controlling a photovoltaic and energy storage hybrid parallel power supply system based on a substation according to an embodiment of the present application;

[0025] Figure 8 A super capacitor output voltage and current diagram in a simulation state of the method for coordinating and controlling a photovoltaic and energy storage hybrid parallel power supply system based on a substation according to an embodiment of the present application;

[0026] Figure 9 A storage battery output voltage and current diagram in a simulation state of the method for coordinating and controlling a photovoltaic and energy storage hybrid parallel power supply system based on a substation according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] In instant energy storage technology, super capacitors and batteries are the most critical, but the dynamic characteristics of the two are quite different: batteries are difficult to cope with high transient loads, while super capacitors are more suitable for rapid power fluctuations. This difference requires the coordinated operation of the energy storage system through a coordination control system. At present, although progress has been made in the research of photovoltaic micro-grid system control strategies, there are still deficiencies in dynamic response speed, control accuracy, etc.

[0028] The application is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0029] As shown in Figure 1 and Figure 3 According to an embodiment of the application, in one aspect, a substation-based photovoltaic and energy storage hybrid parallel power supply system coordination control method is provided, comprising the following steps:

[0030] Step S101: Collecting historical power supply data of a power grid system comprehensive model and current operating state, and constructing an energy storage prediction optimization model, wherein the historical power supply data includes power output data of a photovoltaic system in different weather conditions and different time periods, voltage, current, and capacity change data in the charging and discharging cycle process of a super capacitor energy storage system, and charging and discharging efficiency, state of charge change, and self-discharge rate of a battery energy storage system; the current operating state includes current irradiance, temperature of the photovoltaic panel, and current voltage, current, and state of charge of the super capacitor and the battery; and the power grid system comprehensive model includes a photovoltaic system dynamic model, a super capacitor energy storage system dynamic model, and a battery energy storage system dynamic model.

[0031] Step S103: Constructing a minimum cost function based on physical constraints based on the energy storage prediction optimization model, to obtain optimal control input.

[0032] Step S105: Generating a power regulation instruction based on the optimal control input.

[0033] Step S107: If the output power of the photovoltaic system decreases, controlling the super capacitor energy storage system to discharge to compensate for the bus voltage and controlling the battery energy storage system to increase the output power based on the power regulation instruction; if the output power of the photovoltaic system increases, controlling the super capacitor energy storage system to charge to absorb the excess power of the photovoltaic system, and controlling the battery energy storage system to reduce the output power to maintain the stability of the bus voltage.

[0034] In this embodiment, continue to combine Figure 2 As shown in

[0035] On the DC side, dynamic models of a photovoltaic system, a supercapacitor energy storage system, and a battery energy storage system are connected in parallel. The photovoltaic system dynamic model converts solar energy into DC power through a photovoltaic array; its output power is affected by environmental factors such as sunlight and temperature, exhibiting fluctuations and intermittency. The supercapacitor energy storage system dynamic model, with its rapid charge-discharge characteristics (millisecond-level response), can handle short-term power fluctuations on the DC side (such as sudden changes in photovoltaic power or abrupt load changes) and provide instantaneous power support, compensating for the response delays of other systems. The battery energy storage system dynamic model has high energy density, enabling medium- to long-term energy storage and release (such as smoothing intraday photovoltaic power fluctuations and addressing energy gaps at night or on cloudy days), maintaining medium- to long-term power balance on the DC side.

[0036] Photovoltaic system dynamic model

[0037] Photovoltaic panels use semiconductor technology to directly convert sunlight into electrical energy, employing a single-diode model of photovoltaic cells. Using 213.15W solar panels, arranged in a photovoltaic array, six modules are connected in series and seven in parallel. The current-voltage relationship of the photovoltaic modules is shown in equation (1):

[0038]

[0039] Among them, I x Indicates the output photovoltaic current, v x Indicates the output photovoltaic voltage, I photon V represents the photocurrent. thermal Indicates the output thermal voltage. This represents the equivalent series resistance. N represents the shunt resistor. y and N x represents the number of PV modules arranged in series and parallel configurations, respectively, while the ideality factor is represented by a. The DC bus is connected to the PV board via a unidirectional boost DC-DC converter, and the dynamic equation of the PV boost converter is shown in equation (2).

[0040]

[0041] Among them, L x C represents the output filter inductance. x Indicates the PV side capacitance. This represents the output capacitor. The filter current is expressed as... Indicates the output voltage, v x Represents PV voltage. Control input d x(t) ranges from 0 to 1. The output current of the PV converter is i2. To achieve maximum power point tracking, the perturb and observe method is used to optimize the bias voltage and enhance the power extraction of the PV system.

[0042] Super capacitor energy storage system dynamic model

[0043] Super capacitors connected to the microgrid through a bidirectional DC / DC converter can effectively deal with high-frequency time-varying disturbances caused by sudden load changes. Unlike traditional batteries, super capacitors are advanced energy storage devices for modern microgrids. They have a fast charging and discharging rate, which allows them to carry high-frequency current components and effectively extend the life of the battery. Super capacitors can quickly absorb excess energy during periods of high power generation and release energy during peak demand, ensuring balanced and reliable power supply. Integrating super capacitors into the microgrid system can enhance system flexibility and make the DC bus voltage more robust.

[0044] Energy stored in the super capacitor during charging which can be expressed as:

[0045]

[0046] In formula (3), R SC is the internal equivalent resistance of the super capacitor, V SC is the terminal voltage, V max is the maximum safe terminal voltage, Q SC is the charge quantity, E o,SC is the initial stored energy, A SC ,B SC ,K SC is a constant parameter.

[0047] Further, the energy released by the super capacitor during discharging which can be expressed as:

[0048]

[0049] Battery energy storage system dynamic model

[0050] Batteries in the microgrid serve as energy storage units to meet the long-term power demand of the system. A single battery energy storage module is composed of series-connected lead-acid (carbon) batteries. In addition, multiple battery modules are connected in parallel to meet the output current demand. The battery module transmits power to or absorbs power from the DC bus through a bidirectional DC / DC converter, and the power exchange between the battery and the DC microgrid affects the remaining capacity (SoC) of the battery, as shown in formula (5).

[0051]

[0052] where Qbat represents the rated battery capacity in ampere-hours (Ah), the battery current is represented by i bat represents the rated battery capacity in ampere-hours (Ah), the battery current is represented by i bat is positive during discharging and negative during charging. The battery current can be represented as

[0053]

[0054] The battery internal resistance in equation (6) is represented by R bat , represents the internal voltage of the battery during charging and discharging modes, respectively, as shown in equations (7), (8).

[0055]

[0056] where σ soc = 0.9 - SoC, E o represents the initial stored energy of the battery, A represents the exponential voltage, B represents the exponential capacity function, and K represents the polarization voltage. The dynamic equation of the DC-DC bidirectional converter is shown in equation (9).

[0057]

[0058] where the duty cycle of the boost converter is represented by d bat ∈ [0, 1], i Lbat (t) represents the inductor current, v bat (t) represents the battery voltage, v o2 represents the output voltage, r Lbat represents the resistance, L bat represents the inductance of the filter, and C bat represents the battery capacitance, C o2 represents the output capacitance of the converter, and i2 represents the output current.

[0059] The photovoltaic system provides basic energy, the super capacitor deals with short-term fluctuations, and the battery handles medium and long-term energy regulation. The combination of the three can cover power demand at different time scales and improve the efficiency of energy utilization on the DC side. The average current and high-frequency current can be separated by a moving average filter, and the high-frequency current is used as the reference signal for the dynamic model of the super capacitor energy storage system; the average current is used as the reference signal for the dynamic model of the energy storage battery system.

[0060] The historical power supply data includes power output data of the photovoltaic system in different weather and different time periods, voltage, current, capacity change data in the charging and discharging cycle process of the super capacitor energy storage system, and historical data such as charging and discharging efficiency, state of charge (SOC) change, self-discharge rate of the battery energy storage system. At the same time, the running state of the micro-grid at the current time is obtained in real time, including the current irradiance and temperature of the photovoltaic panel, the current voltage, current, SOC and other current running state data of the super capacitor and the battery.

[0061] Based on these data, a comprehensive model of the micro-grid system is constructed.

[0062] Photovoltaic system dynamic model: considering the photoelectric conversion characteristics of photovoltaic cells, according to the functional relationship between irradiance, temperature and output power, the dynamic change law of photovoltaic power with environment is accurately simulated.

[0063] Super capacitor energy storage system dynamic model: combining the internal resistance, capacity, charging and discharging rate and other parameters of the super capacitor, the dynamic correlation between voltage, current and SOC is established, which reflects the fast charging and discharging response characteristics of the super capacitor, and simulates the voltage change process of the super capacitor under high-frequency power fluctuation.

[0064] Battery energy storage system dynamic model: considering the charging and discharging efficiency, polarization effect, SOC-voltage characteristics and other parameters of the battery, a model reflecting the slow dynamic response and capacity change law of the battery is constructed, which is used to simulate the state evolution of the battery in the long-time and large-power regulation process.

[0065] Based on the above sub-models, a complete energy storage prediction and optimization model is integrated and built, which provides a system-level analysis basis for subsequent control strategies.

[0066] Taking the energy storage prediction and optimization model constructed in step S101 as the operation core, the historical and current full-quantity data are input, the micro-grid running state in a certain prediction time domain in the future is deduced, the control output expectation is obtained, including the charging and discharging power expectation of the super capacitor and the battery at different future time, bus voltage expectation and other control-related output quantities.

[0067] The target output is clearly set, for example, according to the national standard for stable operation of the microgrid and the actual engineering demand, the bus voltage needs to be strictly maintained within ±3% of the rated voltage (compared with the conventional ±5% range, which can further improve the voltage stability accuracy), the charging and discharging power of the super capacitor needs to be limited within the power boundary corresponding to its maximum charging and discharging current (for example, if the maximum charging and discharging current of the super capacitor is 100A, the maximum charging and discharging power threshold is converted according to the current voltage), and the charging and discharging power of the battery needs to match its charging and discharging capability curve (for example, to avoid high-current charging of the battery at high SOC and high-current discharging of the battery at low SOC, which can damage the battery life). The optimization objective is to minimize the deviation between the actual output and the target output, and a cost function is constructed to minimize the deviation. The optimization problem is solved by a numerical optimization algorithm, which continuously adjusts the control output expectation to obtain the optimal control input, so that the difference between the actual output and the target output is as small as possible, and the microgrid operating state is as close as possible to the ideal stable operating state.

[0068] Based on the optimal control input obtained in step S103, the optimal control input is converted into specific power regulation instructions according to the communication protocol and control interface specification of each energy storage device and control device in the microgrid. For example, for the controller of the super capacitor energy storage system, instructions containing charging and discharging power values, action time, duration, etc. are generated; for the controller of the battery energy storage system, power regulation instructions that adapt to its charging and discharging characteristics are generated, which clearly define the charging and discharging power, SOC control target, etc. in each regulation period; at the same time, the bus voltage stability control requirement is integrated into the instructions to ensure that the power regulation of each energy storage device is coordinated, and a set of power regulation instructions is generated to provide clear and executable operation basis for subsequent execution of the control strategy.

[0069] When receiving the power regulation instructions, the output power of the photovoltaic system is monitored in real time:

[0070] Photovoltaic system output power drop scenario: once it is monitored that the photovoltaic power decreases due to factors such as reduced light (e.g. cloud cover, evening period), reduced temperature, etc., the super capacitor energy storage system immediately discharges according to the power regulation instructions. Since the super capacitor has fast response characteristics (response time can reach milliseconds), it can release electric energy in a very short time to compensate for the bus voltage drop caused by the decrease of photovoltaic power, and quickly pull the bus voltage back to the rated voltage allowed range; at the same time, the battery energy storage system gradually increases the output power according to the power regulation instructions, and since the battery can provide sustained and large-capacity power support, it continuously supplements the power gap after the super capacitor responds quickly, ensuring stable power supply on the load side, and the two work together to achieve bus voltage stability control under the condition of photovoltaic power drop.

[0071] Photovoltaic system output power increase scenario: If the photovoltaic power is monitored to have an upward trend due to factors such as increased light (e.g. cloud dispersal, midday period), the supercapacitor energy storage system is charged according to the instructions, and the supercapacitor's fast charging capability is used to timely absorb the excess photovoltaic power, avoiding a substantial rise in the bus voltage due to excess power. At the same time, the battery energy storage system reduces its output power according to the instructions, reducing its own discharge or switching to a small current charging state (within the allowable range of SOC), cooperating with the supercapacitor to jointly stabilize the bus voltage fluctuations and keep the bus voltage stable around the rated value, achieving reasonable consumption of excess photovoltaic power and stable control of the bus voltage. Throughout the process, the bus voltage, the state of each energy storage device, and other operating parameters are continuously monitored, and the power regulation instructions can be adjusted in real time according to the actual operating conditions (such as when there is a sudden load change), ensuring that the microgrid is always in a stable operating state.

[0072] By constructing precise dynamic models of photovoltaic systems, supercapacitor and battery energy storage systems, and combining historical full data and real-time state data, the operating characteristics of each device under different operating conditions can be accurately described, significantly improving the prediction accuracy of the energy storage prediction and optimization model. Based on the model predictive control algorithm to solve the optimal control input, the deviation between the actual output and the target output can be effectively reduced, making the control strategy more suitable for the stable operation of the microgrid, and significantly improving the control accuracy, which can stabilize the bus voltage.

[0073] In the control strategy, precise model prediction and optimization control are used to develop charging and discharging instructions for supercapacitors and batteries that are adapted to their characteristics. For example, for batteries, deep charging and discharging, frequent large current charging and discharging, and other operations that damage battery life are avoided; for supercapacitors, the charging and discharging rate and cycle number are reasonably controlled to reduce unnecessary capacity decay. This effectively extends the service life of energy storage devices, reduces system operation and maintenance costs and equipment replacement frequency, and improves the economic efficiency of photovoltaic and battery hybrid parallel power systems.

[0074] This coordinated control method can comprehensively cover various operating conditions of photovoltaic system output power fluctuations (such as random changes in light, seasonal changes, weather changes, etc.), effectively stabilize bus voltage fluctuations through real-time monitoring and rapid response control, and ensure stable power supply on the load side. Whether in extreme operating conditions with large fluctuations in photovoltaic power or in regular operating conditions, the microgrid system can be ensured to operate stably and reliably, improving the microgrid's ability to cope with complex environments and operating condition changes, and providing high-quality power supply for substations and associated loads.

[0075] By reasonably controlling the super capacitor and the battery to absorb the excess photovoltaic power, the effective consumption of photovoltaic energy is realized, and the phenomenon of light abandonment due to excess photovoltaic power is reduced. The excess photovoltaic power is stored in the energy storage device, and the electric energy is released when the photovoltaic power is insufficient, thereby improving the utilization rate of photovoltaic energy, meeting the development needs of new energy consumption and energy saving and carbon reduction, and helping the microgrid to develop in a more green and efficient direction.

[0076] Further, in combination with Figure 2 As shown in the figure, the microgrid system comprehensive model includes: an alternating current side provided with an alternating current bus, a substation and a station alternating current load; a direct current side provided with a direct current bus, a station direct current load, a one-way DC / DC converter and a bidirectional DC / DC converter; a bidirectional AC / DC converter connecting the alternating current bus and the direct current bus; wherein the one-way DC / DC converter connects the direct current bus and the photovoltaic array, and the bidirectional DC / DC converter is connected in parallel with the super capacitor energy storage system and the battery energy storage system.

[0077] Among them, the bidirectional DC / DC converter is provided as at least two, one of which has three ports to connect the super capacitor energy storage system and the battery energy storage system in parallel, and at least one of which has two ports to connect another battery energy storage system, so that the super capacitor energy storage system is connected in parallel with multiple battery energy storage systems.

[0078] The alternating current side and the direct current side are respectively provided as two, and the two alternating current sides and the two direct current sides are connected by connecting lines.

[0079] Alternating current side architecture:

[0080] Electric energy input and conversion: #1, #2 station transformer connects external alternating current power (adapted to the voltage level of station alternating current load) to #1, #2 AC380V bus, and #1, #2 AC380V bus is the alternating current bus; the bidirectional AC / DC converter serves as an AC / DC bridge to convert the alternating current of the AC380V bus into DC220V direct current and send it to the direct current bus; alternatively, if the direct current side has excess electric energy, it can be converted into alternating current and sent back to the alternating current bus.

[0081] Alternating current load power supply: #1, #2 AC380V bus directly supplies power to corresponding station alternating current load to meet the alternating current side power demand.

[0082] The substation is connected to an AC bus, undertakes the functions of voltage transformation and power distribution, and adapts to the power demand of station AC loads of different voltage levels. Station AC loads include lighting, ventilation, conventional power equipment, etc. in the substation. They are connected to the AC bus through a power distribution circuit to obtain power from the AC side to meet the daily operation and emergency power demand. The AC side is set as two independent units, which are electrically connected through a dedicated connection line, can be used as backup for each other to balance the load in normal operation, or quickly switch in case of failure to ensure the continuity of AC power supply.

[0083] DC side architecture

[0084] The DC side takes the DC bus as the hub of power transmission and is connected to station DC loads, such as relay protection devices, automation control equipment, emergency lighting, etc. in the substation, which have very high requirements for power supply reliability and rely on stable power supply from the DC bus. The DC bus is DC220V. The unidirectional DC / DC converter is used as the interface between the photovoltaic array and the DC bus. Due to the output characteristics of the photovoltaic array (only unidirectional output of power), it uses Boost or Buck topology (selected according to the matching requirements of the photovoltaic output voltage and the DC bus voltage) to realize unidirectional conversion and transmission of photovoltaic power. The DC power generated by the photovoltaic array is sent into the system after being adapted to the DC bus voltage.

[0085] Bidirectional DC / DC converter

[0086] The bidirectional DC / DC converter has the ability of bidirectional energy flow. One bidirectional DC / DC converter constructs a parallel channel for the supercapacitor energy storage system and the battery energy storage system. Based on the characteristics of supercapacitors (fast response, high power density) and batteries (high energy density, large capacity), the two are coordinated in power compensation and energy storage through the converter. For example, the supercapacitor can quickly ingest power through the converter to cope with transient fluctuations, and the battery can be continuously charged and discharged through the converter to handle long-term energy regulation. Another bidirectional DC / DC converter is connected to an additional battery energy storage system to expand the flexibility of energy storage configuration, allowing the supercapacitor to be connected in parallel with multiple batteries to adapt to different energy storage capacity and power demand scenarios. The DC side is also set as two independent units connected by a connection line to support cross-unit energy allocation and redundant backup.

[0087] The bidirectional AC / DC converter has the functions of rectification and inversion. In rectification mode, it converts the AC bus power into DC power to supply power to the DC side or charge the energy storage; in inversion mode, it inverts the DC side power (such as the power released by the energy storage) into AC power to feed into the AC bus to support the AC side load or send power to the grid (if allowed). Through the bidirectional AC / DC converter, bidirectional energy flow between the AC side and the DC side is realized, and a hybrid AC / DC microgrid energy interaction network is constructed.

[0088] The super capacitor and the multiple groups of storage batteries are connected in parallel to the DC bus through the bidirectional DC / DC converter to form a layered and collaborative energy storage architecture: the super capacitor relies on its fast response advantage to respond to high-frequency power fluctuations (such as sudden changes in photovoltaic output and load impact) in milliseconds through the corresponding bidirectional DC / DC converter; the storage battery group adjusts energy in the dimension of seconds to minutes through the respective bidirectional DC / DC converter, cooperates with the super capacitor to smooth power fluctuations and optimize energy distribution, and jointly ensures the stability of the DC bus voltage.

[0089] The DC 220V DC bus directly supplies power to the station DC load; the three-port DC / DC converter coordinates the energy interaction between the super capacitor, the energy storage battery, and the bus (for example, the super capacitor quickly responds to power fluctuations, and the energy storage battery bears continuous charging and discharging), and the bidirectional DC / DC converter focuses on the bidirectional power flow between the energy storage battery and the bus, and through the cooperation of multiple types of converters, the energy between the DC side photovoltaic, energy storage, and load can be flexibly allocated, and photovoltaic storage charging and discharging can be realized.

[0090] The interconnection structure formed by the AC side and the DC side double units and the connecting line can provide physical layer redundancy for the power supply system. In normal operation, the double units can balance the load and reduce the single device and single loop failure rate; in the event of a fault, the system can quickly switch and supply power across the units through the connecting line, such as when one unit of the AC side fails, the other unit can continue to supply power to the important load through the connecting line, greatly reducing the risk of power failure, and adapting to the stringent requirements of the substation on power supply reliability.

[0091] The super capacitor and the multiple storage batteries are connected in parallel, and cooperate with the bidirectional DC / DC converter, so that even if a group of storage batteries or a converter fails, the remaining energy storage units can still maintain basic functions through the converter. For example, if a group of storage batteries fails, the super capacitor can combine the remaining storage batteries to continue to participate in power regulation through the corresponding converter, ensuring stable operation of the system in fault conditions.

[0092] More specifically, the photovoltaic, hybrid energy storage system (collectively referred to as super capacitor energy storage system and storage battery energy storage system) and the DC load are connected in parallel to the DC bus, and the photovoltaic system optimizes power extraction through the control of the DC-DC converter. The energy storage battery and the super capacitor are connected to the DC bus through the bidirectional DC / DC converter to realize effective energy transmission for charging and discharging. The AC distribution unit can conveniently input or output power from the main grid when needed. In this microgrid framework system, the DC load is coupled to the 220V / 110V DC bus. The microgrid is controlled by a dynamic power management system and operates in autonomous or grid-connected mode as needed.

[0093] The minimum cost function is divided into a charging and discharging minimum cost function and a bidirectional DC-DC converter minimum cost function. The minimum cost function includes a manipulated variable, a reference trajectory, and a weight factor.

[0094] The charge-discharge minimization cost function is:

[0095]

[0096] where N c is the control horizon, N p is the prediction horizon, u is the future sequence of input signals, y * is the reference trajectory, w is the manipulated variable, and λ is the weighting factor.

[0097] The objective of the optimization function is to reduce J at each sampling interval. The control and prediction horizons are denoted by N c and N p , respectively, and the above variables are used to minimize the cost and optimize the duty cycle sequence. In a microgrid, the DC bus connects energy storage devices such as batteries and supercapacitors through DC-DC circuits, and the remaining energy of the photovoltaic system can charge the energy storage devices, which are responsible for meeting the load demand when the AC distribution unit loses power.

[0098] The charge-discharge minimization cost function is used to quantify the deviation between the actual operating state and the desired reference state of the microgrid energy storage system during charging and discharging, as well as the cost of changes in the control input signal.

[0099] N p represents the prediction horizon length, which determines the number of time steps forward that the model predictive control (MPC) looks when predicting future system states. By setting an appropriate N p , the dynamic process of charging and discharging of energy storage systems (supercapacitors, batteries) in a microgrid can be covered, such as the influence period of factors such as photovoltaic output changes and load fluctuations on energy storage charging and discharging, allowing the control strategy to predict these changes in advance and provide sufficient prediction information for optimal control.

[0100] N c is the control horizon length, which determines the number of time steps used to calculate and output control instructions at each control period. It works with N p to achieve rolling optimization of prediction and control, ensuring real-time control while utilizing prediction information to optimize current control decisions.

[0101] y * (t+j|t) is the system reference trajectory predicted at time t for time t+j, which includes the desired state of the energy storage system charging and discharging, such as the desired SOC (State of Charge) of the battery and the desired voltage of the DC bus, and is an ideal target for measuring whether the system is operating optimally.

[0102] w(t+j) as a manipulated variable, reflects the actual running state of the system at t+j time, such as the actual charging and discharging power of the energy storage system, the actual voltage of the bus, etc., and is compared with the reference trajectory y * (t+j|t) difference, the state deviation cost is calculated.

[0103] Δu(t+j-1)] is the control increment, that is, the change of the input signal at adjacent control times. By punishing (multiplying by the weight λ(j)), the frequent and large fluctuations of the control signal can be avoided, the power devices (such as IGBT) of the bidirectional DC-DC converter can be protected, the equipment loss can be reduced, and the stability of the microgrid system operation can be improved, and the violent shock of the bus voltage and power caused by the sudden change of the control quantity can be prevented.

[0104] δ(j) and λ(j) are the weighting factors of state deviation and control increment, respectively, which can be dynamically adjusted according to different operating conditions of the microgrid (such as full-load of photovoltaic, peak load, system fault recovery). For example, during the system fault recovery stage, the weight of δ(j) is increased to preferentially ensure that the system state quickly returns to the reference trajectory, thereby ensuring power supply reliability; during the steady-state operation of the system, the weight of λ(j) is increased to smooth the control signal and reduce the equipment operating noise and loss.

[0105] By minimizing the cost function of charging and discharging, while constraining the system state deviation (the difference between the actual running state and the reference trajectory) and the control increment, the charging and discharging process of the microgrid energy storage system always conforms to the optimal running trajectory. Whether it is to cope with the random fluctuations of photovoltaic output or to match the dynamic changes of load, the charging and discharging power of the energy storage can be accurately adjusted, the AC and DC bus voltage is stably maintained within a reasonable range, and the power quality of the microgrid is improved. For example, when the photovoltaic output suddenly decreases, the battery can be quickly discharged to control the DC bus voltage deviation within ±5% of the rated value, thereby meeting the power supply requirements of relay protection and automatic control equipment.

[0106] The cost function of model predictive control (MPC) includes reference values, state variables, and regulator inputs.

[0107] J = J(x(k), U(k)); (19)

[0108] In equation (19), U(k) represents the control vector. The goal of MPC optimization is to minimize the cost function while satisfying the system performance constraints.

[0109] The bidirectional DC-DC converter is the core device for energy interaction between the AC and DC sides of the microgrid and charging and discharging control of the energy storage system, and its cost function focuses on current tracking accuracy and control smoothness.

[0110] The minimum cost function of the bidirectional DC-DC converter is:

[0111]

[0112] In the formula, i ref (t+k|t) is the reference current at time t+k, i L (t+k) is the actual inductor current at time t+k, δ is the current tracking weight coefficient, Δu(t+k-1) is the control increment, and λ is the control smoothing weight coefficient.

[0113] i ref (t+k|t) is the reference current of the bidirectional DC-DC converter predicted at time t+k. It is determined by the power dispatch requirements and energy storage charging and discharging strategies at the microgrid system level to ensure that the converter operates in the optimal power flow direction (charging or discharging) and power magnitude.

[0114] i L (t+k) is the actual inductor current of the converter at time t+k, directly reflecting the power transfer state of the converter. This is calculated using the reference current i. ref The square of the deviation (t+k|t) is multiplied by the current tracking weighting coefficient δcon to strictly constrain the current tracking accuracy of the converter, ensuring that energy is transferred as expected between the AC / DC side and the energy storage system. For example, when the supercapacitor responds quickly to power fluctuations, its corresponding bidirectional DC-DC converter can accurately track the reference current and achieve millisecond-level power compensation.

[0115] Δu(t+k-1)| is also a control increment. Penalizing it (multiplying it by the control smoothing weight coefficient λcon) can avoid sudden changes in the control signal (such as PWM duty cycle) of the bidirectional DC-DC converter, reduce the switching losses of power devices, and extend the equipment life. At the same time, a stable control signal can make the output power of the converter change smoothly, prevent the DC bus voltage from being impacted by sudden power changes, and improve the voltage stability of the DC side of the microgrid.

[0116] The penalty for control increments in the cost minimization function of bidirectional DC-DC converters, and the limitation on control signal fluctuations in the charge / discharge minimization cost function, significantly reduce the switching frequency and switching losses of power devices. Taking IGBTs as an example, smooth changes in the control signal can reduce the number of switching operations, extending the device lifespan from the estimated 3-5 years based on frequent switching to closer to the theoretical lifespan (over 100,000 hours), reducing equipment replacement costs and maintenance workload. At the same time, it avoids overcurrent and overvoltage surges caused by sudden control changes, reducing the probability of equipment failure and improving the reliability of microgrid systems.

[0117] The adjustable nature of the weighting factors δ(j) and λ(j) in the charge-discharge minimization cost function and the adjustable nature of δcon and λcon in the bidirectional DC-DC converter minimization cost function enable the minimization cost function to adapt to various operating conditions of the microgrid. In the grid-connected mode, the control can focus on smoothness, thereby reducing the impact on the power grid. In the off-grid mode, the system state tracking accuracy is prioritized to ensure reliable power supply to important loads (such as emergency lighting and communication equipment). By adjusting the weighting factors flexibly, optimal control in different operating conditions is achieved, and the adaptability of the microgrid in all scenarios is improved.

[0118] Further, in step S103, the energy storage prediction optimization model is an MPC, and a minimum cost function based on physical constraints is constructed based on the energy storage prediction optimization model to obtain optimal control input, specifically including the following steps:

[0119] In step S1031, based on the model predictive control, the dynamic model of the supercapacitor energy storage system, and the dynamic model of the battery energy storage system, an electronic circuit model of the bidirectional DC / DC converter is established to obtain a continuous-time state space model.

[0120] In step S1033, the continuous-time state space model is discretized and linearized to obtain a discrete-time state space model.

[0121] In step S1035, a control output expectation is obtained based on historical data and the discrete-time state space model.

[0122] In step S1037, the minimum cost function is constructed based on physical constraints, and the control output expectation is optimized to obtain optimal control input.

[0123] Through the electronic circuit modeling in step S1031 and the discretization and linearization in step S1033, the dynamic characteristics (such as energy storage charge-discharge response and converter voltage-current transformation law) of the supercapacitor, battery, and bidirectional DC / DC converter can be accurately described, providing a reliable foundation for accurate control and avoiding control deviation caused by model mismatch.

[0124] By planning the future behavior (such as charge-discharge power and voltage regulation trend) of the energy storage system in advance, in the scenarios of photovoltaic power fluctuation and load mutation, the control strategy is adjusted through forward-looking prediction, so that the energy storage response is more timely and more in line with system demand, reducing voltage / power fluctuations.

[0125] Step S1037 physically constraints + minimizes the cost function, and integrates the physical limits of the device (such as the upper limit of the battery charge and discharge current, the voltage range of the super capacitor), and energy transmission constraints (such as power matching and bus voltage stability) into the optimization. Both avoid damage to energy storage devices due to over-limit operation (extend the service life and reduce operation and maintenance costs), and ensure that the microgrid operates within the safe and stable boundary (such as meeting critical loads when power is lost to prevent cascading failures).

[0126] The minimization of the cost function directly relates the energy storage charge and discharge loss, device life cost, and power regulation cost (such as the weight factor balancing the number of energy storage actions and control accuracy). Through MPC rolling optimization, the charge and discharge strategy of the battery and the super capacitor is dynamically allocated (such as using the super capacitor to cut the peak when there is excess photovoltaic power to reduce the high-frequency charge and discharge loss of the battery) under the premise of meeting the physical constraints, which can reduce the system operation cost and improve the photovoltaic consumption rate and energy storage utilization efficiency.

[0127] Step S1033, discretize and linearize the continuous-time state space model to obtain a discrete-time state space model, specifically including the following steps:

[0128] Step S10331: Obtain the continuous-time state space model of the Boost converter based on the state equation of the Boost circuit.

[0129] The state equation of the Boost circuit is:

[0130]

[0131] The continuous-time state space model of the inductor current and the output voltage of the Boost converter is:

[0132]

[0133] In the formula, is the current change rate, is the output voltage change rate;

[0134] According to formula (12), the following can be further obtained:

[0135]

[0136] Step S10333, discretize the continuous-time state space model based on the discrete-time state space model framework to obtain a discrete-time state space.

[0137] The discrete-time state space model framework is

[0138]

[0139] In the formula, x is the system state variable, u is the control variable, Y(s) is the output variable, A and B are the state matrices, w is the system disturbance, N is the disturbance input matrix, C is the output observation matrix, and s is the sampling time.

[0140] The discrete-time state space is:

[0141]

[0142] In the formula, i L (s+1) represents the inductor current at time s+1, V b (s+1) is the output voltage at time s+1, and T s The sampling period is u(s), and the control input is u(s).

[0143] In equation (15), the sampling time is represented by T. s The performance characteristics of a Boost converter at its output can be described as follows:

[0144]

[0145] The derivation and specification of the state space description above are shown in Equation (17).

[0146]

[0147] Step S10335: Linearize the discrete-time state space to obtain the discrete-time state-space model:

[0148]

[0149] In the formula, V b (k), i L (k) represents the steady-state values ​​of the converter output voltage and inductor current, respectively; D is the duty cycle of the main switch; u b (k) represents the duty cycle of the auxiliary switch.

[0150] Weighting factors and physical constraints play a crucial role in influencing the desired behavior of the controller, and appropriate parameters need to be set to achieve the expected performance. The state space representation of the buck mode is similar to that of the boost mode, and its state space representation is shown in equation (21).

[0151]

[0152] Among them, R s For the source internal resistance, C s For the input of the converter, E s The internal voltage of the source.

[0153] In this embodiment, the charging and discharging process of the battery and supercapacitor unit is managed by a bidirectional DC-DC converter, each energy storage module is equipped with an independent converter. To effectively optimize the discharging and charging process, the bidirectional converter uses two independent cost minimization functions. Considering the existence of fluctuations in the output voltage of the battery module, the supercapacitor is used to stabilize the DC bus through the DC-DC converter. The behavior of the converter in the charging (buck) and discharging (boost) modes is predicted by the state-space model to manage the power transfer between the energy storage system and the load bus. The control objective is to minimize the defined cost function, ensuring that the actual system output follows the reference signal,

[0154] Further, the substation-based photovoltaic energy storage hybrid parallel power system coordination control method further comprises the following steps:

[0155] Step S201: At each sampling time, the current actual state variable is obtained, and the current actual state variable includes: inductance current, DC bus voltage, battery / supercapacitor terminal voltage, state of charge;

[0156] Step S203: Within the prediction range, taking the current actual state variable as the initial, based on the minimum cost function and the control output expectation, the control sequence that minimizes the cost is solved online to obtain the optimal control input, and the optimal control input includes the current time control amount;

[0157] Step S205: The current time control amount is input into the bidirectional DC / DC converter, and the next sampling time is entered, and rolling optimization is realized.

[0158] Model predictive control (MPC) is used to manage the electromagnetic and supercapacitor modules, and the control output is predicted every sampling interval t within a defined prediction range P by using historical data and predicted future output and current operating state. Model predictive control optimizes the control signal under physical constraints by minimizing the cost function, thereby reducing the deviation between the actual output and the target output. Physical constraints include: the inductance current does not exceed the rated value of the device to avoid overcurrent damage; the bus voltage is maintained within the allowable fluctuation range to ensure normal operation of the load; the energy storage SOC is within a reasonable range [SOC{min}, SOC{max}] to prevent overcharging / overdischarging.

[0159] At each sampling time, the new measurement state variables from the device are used as the initial conditions of the system model, and then the optimization problem is recalculated to generate the control input at the next time. That is, the current time control quantity (duty ratio D) is input into the PWM drive module of the bidirectional DC / DC converter to adjust the on / off time of the power tube, so as to realize precise regulation of energy (energy storage when charging, energy supplement when discharging). After completing the control action of the current sampling period, the next sampling time is entered, and the "state sensing-prediction optimization-control execution" process is repeated to form a rolling optimization closed loop. This mechanism can dynamically compensate for the random fluctuations of photovoltaic output, load mutations and other disturbances, and continuously correct the deviation between the model prediction and the actual system.

[0160] Wherein, the prediction range refers to the time interval in which the system output or state change is predicted forward based on the current state at each sampling time, and the core is a set of future time steps (corresponding to multiple sampling intervals T), rather than a single sampling period. The time length of "continuous P sampling intervals" from the current time is the prediction range. For example, if the sampling interval T is 0.1s and the prediction range P is 5, the control output and state change every 0.1s (every sampling interval) in the next 0.1*5=0.5s will be predicted. The control of the bidirectional DC / DC converter needs to be dynamically adjusted over time (such as the duty ratio, current command, etc. at each sampling time). The control sequence is a set of control commands arranged in time sequence, which contains multiple sampling times in the future (in the prediction range), and the control quantity required to be applied to the converter to make the system achieve the optimal operating state (such as stable bus voltage, reasonable energy storage charging and discharging). The control of the bidirectional DC / DC converter needs to be dynamically adjusted over time (such as the duty ratio, current command, etc. at each sampling time). The control sequence is a set of control commands arranged in time sequence, which contains multiple sampling times in the future (in the prediction range), and the control quantity required to be applied to the converter to make the system achieve the optimal operating state (such as stable bus voltage, reasonable energy storage charging and discharging). For example, the duty ratio is adjusted to 0.3 at the first time, to 0.32 at the second time, and so on.

[0161] In actual execution, in order to adapt to the changes of the system in real time (such as sudden change of photovoltaic output, fluctuation of load), all control quantities in the future will not be blindly executed, but only the control quantity at the current time is used to drive the converter, and a new control sequence is calculated at the next time to ensure that the control always follows the actual state of the system.

[0162] The control sequence is a set of control commands arranged in time sequence in the model predictive control, which is planned in the prediction time domain to achieve the optimal operation of the system, and supports the dynamic adaptation logic of rolling optimization, so that the photovoltaic storage system can be accurately coordinated and controlled regardless of the changes of photovoltaic and load.

[0163] By combining real-time status feedback with short-term multi-step prediction, it can quickly respond to disturbances such as sudden changes in photovoltaic output (e.g., a sudden drop in power due to cloud cover) and load switching (e.g., a sudden increase in power output from substation equipment). Compared to traditional PI control, it can reduce DC bus voltage fluctuations by 15%-30% and shorten the dynamic adjustment time to 100-300ms, ensuring the stability of the DC side voltage of the substation and preventing malfunctions of protection devices due to voltage instability.

[0164] To verify the effectiveness of the coordinated control method for photovoltaic hybrid parallel power systems in substations in coordinating the operation of photovoltaic cells, batteries, and supercapacitors, this invention conducted a control simulation experiment on a photovoltaic-storage deep integration microgrid hybrid energy storage system. The experiment tested the adaptability of the control strategy under varying photovoltaic power. The specific process is as follows:

[0165] Simulation parameters

[0166] Simulation time domain: 5s, step size T = 0.0001s;

[0167] Rated power and capacity of load and energy storage system: resistive load (4.8kW), photovoltaic capacity (115kWp), rated power of battery module (5kW), rated capacity of supercapacitor (85F);

[0168] like Figure 1 As shown, the structure of a hybrid energy storage system for a microgrid that deeply integrates photovoltaics and energy storage is demonstrated.

[0169] Simulation verification

[0170] The hybrid energy storage system of a photovoltaic-storage microgrid was simulated using MATLAB / Simulink to test the adaptability of the control strategy under photovoltaic power variation. The total power of the photovoltaic system was 9kW, the power of a single module was 213.15W, the number of series and parallel connections were 6 and 7 respectively, and the switching frequency was 30kHz. The rated power of the battery energy storage module was 5kW, and the rated capacity of the supercapacitor was 85F.

[0171] Among them, the photovoltaic output voltage is as follows Figure 4As shown, during the initial 0-1 second phase of system operation, the photovoltaic system outputs a stable power, working together with the battery energy storage module to meet the load demand. Within 1-3 seconds, the photovoltaic system output fluctuates, and the system power decreases. At this point, the supercapacitor activates first, rapidly discharging to compensate for the bus voltage fluctuations caused by power transients. Subsequently, the battery output power increases to meet the system's long-term power requirements. Within 3-5 seconds, the photovoltaic system output power increases again, and the supercapacitor activates, rapidly charging to absorb excess photovoltaic power. Then, the battery output power decreases to maintain a stable bus voltage. Clearly, compared to the high-frequency oscillations present under PI control, this method can effectively suppress interference caused by photovoltaic fluctuations, ensuring power quality and supply reliability.

[0172] like Figure 5 As shown, under the control strategy, the DC bus voltage can remain stable during photovoltaic power fluctuations, exhibiting only minor fluctuations when photovoltaic output changes, and then quickly stabilizes under control. This demonstrates the strong robustness of the designed controller. However, traditional PI controllers cannot effectively smooth out photovoltaic fluctuations, resulting in larger bus voltage fluctuations and difficulty in ensuring power quality.

[0173] like Figure 6 As shown, changes in solar irradiance cause fluctuations in photovoltaic current and photovoltaic voltage. By leveraging the impact of photovoltaic power fluctuations on the system, the adaptability of microgrid control schemes to harsh operating conditions can be effectively evaluated.

[0174] The power management dynamics of a microgrid system within 0 to 5 seconds are as follows: Figure 7 The diagram shows the output power from key energy storage units, including photovoltaics, batteries, and supercapacitors. Clearly, supercapacitors primarily address the transient power demands of the microgrid, while batteries meet its long-term power requirements. The complementary advantages of supercapacitors and batteries effectively mitigate the impact of photovoltaic fluctuations on the DC bus.

[0175] The current and voltage of the supercapacitor module are as follows: Figure 8 As shown, the reference signal of the supercapacitor is a high-frequency current signal obtained through a moving average filter. Therefore, the supercapacitor has a fast current response speed to meet the transient power demand of the microgrid.

[0176] Figure 9 The image shows the current and voltage of the battery module. The reference signal for the battery is the average current signal obtained through a moving average filter. Therefore, the battery current response speed is relatively slow, which is used to meet the long-term power requirements of the microgrid.

[0177] The application can intuitively show the effectiveness of the control strategy through the control simulation of the light storage deep fusion micro-grid hybrid energy storage system, on the one hand, the dynamic response speed is fast, and the bus voltage fluctuation caused by system disturbance can be effectively suppressed, and the reliability of power supply can be maintained, on the other hand, the output change of the storage battery is smooth under the control, and the normal service life of the storage battery can be ensured. Next, the on-line intelligent management of energy can be further realized by combining the field monitoring data and the artificial intelligence method.

[0178] The application meets the load demand by taking the alternating current distribution unit and the hybrid energy storage system as the main energy. The photovoltaic, energy storage system and load are connected to the DC bus through the power conversion device, the photovoltaic system is operated using the maximum power point tracking (MPPT), and the power extraction is optimized by controlling the respective DC-DC converters. The average current and the high-frequency current are separated by the moving average filter, the high-frequency current is used as the reference signal of the super capacitor, and the energy is provided when the power source fluctuates rapidly; the average current is used as the reference signal of the energy storage battery to meet the long-time power demand of the system. The super capacitor and the energy storage battery operate cooperatively to guarantee the stability of the DC bus side voltage. The power output of the energy storage battery and the super capacitor is managed by the model predictive control method, the dynamic response speed and the control accuracy of the system are improved, and the reliability of the system operation is ensured.

[0179] In another aspect, the application provides an electronic device, characterized in that it comprises at least one processor, a memory in communication connection with the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any one of the substation-based light storage hybrid parallel power supply system coordination control methods.

[0180] In another aspect, the application provides a medium, which is a computer storage medium, and stores a computer program, and the computer program is executed by a processor to implement any one of the substation-based light storage hybrid parallel power supply system coordination control methods.

[0181] The computer storage medium can be referred to as a medium simply. Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware. The program can be stored in a non-volatile computer readable storage medium, and when the program is executed, the processes of the above-mentioned embodiments can be included. Any reference to memory, storage, database or other medium used in each embodiment provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM). Each embodiment in the specification is described in a progressive manner, and the same or similar parts of each embodiment can be referred to each other. Each embodiment focuses on the difference from other embodiments. Especially, for the device, equipment, non-volatile computer storage medium embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant part can be referred to the part of the method embodiment.

[0182] The above embodiments are only examples for clearly illustrating, and not limiting the embodiments. Based on the above description, other different forms of changes or variations can be made by those skilled in the art. Here, all the embodiments cannot be exhausted, and the obvious changes or variations still fall within the protection scope of the present application.

Claims

1. A substation-based coordination control method for a hybrid parallel power system of optical storage, characterized in that, The application relates to a micro-grid system comprehensive model and a power regulation method thereof. The application comprises the following steps: Collecting historical power data and current operation state of a power grid system comprehensive model to construct an energy storage prediction optimization model; wherein the historical power data comprises power output data of a photovoltaic system under different weather conditions and at different time periods, voltage, current and capacity change data of a super capacitor energy storage system in a charging and discharging cycle process, and charging and discharging efficiency, state of charge change and self-discharge rate of a storage battery energy storage system; the current operation state comprises current irradiance and temperature of a photovoltaic panel and current voltage, current and state of charge of the super capacitor and the storage battery; and the power grid system comprehensive model comprises a photovoltaic system dynamic model, a super capacitor energy storage system dynamic model and a storage battery energy storage system dynamic model; Constructing a minimum cost function based on physical constraints based on the energy storage prediction optimization model to obtain optimal control input; Generating a power regulation instruction based on the optimal control input; 2.The substation-based coordination control method of hybrid parallel power system of optical storage according to claim 1, characterized in that, If the output power of the photovoltaic system decreases, the super capacitor energy storage system is controlled to discharge to compensate for the bus voltage based on the power regulation instruction, and the storage battery energy storage system is controlled to increase the output power; if the output power of the photovoltaic system increases, the super capacitor energy storage system is controlled to charge to absorb the excess power of the photovoltaic system, and the storage battery energy storage system is controlled to reduce the output power to maintain the stability of the bus voltage. The micro-grid system comprehensive model comprises: An alternating current side provided with an alternating current bus, a substation and a station alternating current load; A direct current side provided with a direct current bus, a station direct current load, a one-way DC / DC converter and a two-way DC / DC converter; A two-way AC / DC converter connected with the alternating current bus and the direct current bus; 3.The substation-based coordination control method of hybrid parallel power system of optical storage according to claim 2, characterized in that, The one-way DC / DC converter is connected with the direct current bus and a photovoltaic array, and the two-way DC / DC converter is connected in parallel with the super capacitor energy storage system and the storage battery energy storage system. 4.The substation-based coordination control method of hybrid parallel power system of optical storage according to claim 2, characterized in that, The two-way DC / DC converter is provided as at least two, one of which has three ports to connect the super capacitor energy storage system and the storage battery energy storage system in parallel, and at least one of which has two ports to connect another storage battery energy storage system, so that the super capacitor energy storage system is connected in parallel with a plurality of storage battery energy storage systems.

5. The substation-based photovoltaic storage hybrid parallel power system coordinated control method according to any one of claims 1 to 4, characterized in that, The alternating current side and the direct current side are respectively provided as two, and the two alternating current sides and the two direct current sides are connected through connecting lines. Based on the energy storage prediction optimization model, a minimum cost function based on physical constraints is constructed to obtain optimal control input, which specifically comprises: Based on model prediction control, the super capacitor energy storage system dynamic model and the storage battery energy storage system dynamic model, an electronic circuit model of the two-way DC / DC converter is established to obtain a continuous time state space model; The continuous time state space model is discretized and linearized to obtain a discrete time state space model; Based on the historical power data and the discrete time state space model, a control output expectation is obtained; Based on physical constraints, the minimum cost function is constructed to optimize the control output expectation to obtain the optimal control input. 6.The substation-based photovoltaic storage hybrid parallel power system coordination control method according to claim 5, characterized in that, Discretize and linearize the continuous-time state space model to obtain a discrete-time state space model, specifically including: A continuous-time state space model of the Boost converter is obtained based on a state equation of the Boost circuit. The state equation of the Boost circuit is: where i L is the inductor current, V s is the input voltage, V b is the converter output voltage, R L is the load resistance, C o is the output capacitance, L is the converter inductance, is the switch state, including 0 or 1 ; The continuous-time state space model of the inductor current and the output voltage of the Boost converter is: wherein is the rate of change of current, is the rate of change of output voltage; Discretize the continuous-time state space model based on a discrete-time state space model framework to obtain a discrete-time state space; The discrete-time state space model framework is: In the formula, x is a system state variable, u is a control variable, Y(s) is an output variable, A and B are state matrices respectively, w is a system disturbance, N is a disturbance input matrix, C is an output observation matrix, and s is a sampling time; The discrete-time state space is: In the formula, i L (s+1) is the inductance current at s+1 moment, V b (s+1) is the output voltage at s+1 moment, T s is the sampling period, and u(s) is the control input. Linearize the discrete-time state space to obtain the discrete-time state space model: In the formula, V b (k), i L (k) are the steady-state values of the converter output voltage and inductor current, respectively, D is the main switch duty cycle, u b (k) is the auxiliary switch duty cycle. 7.The substation-based photovoltaic storage hybrid parallel power system coordinated control method according to claim 5, characterized in that, The minimum cost function includes a charging and discharging cost function and a bidirectional DC-DC converter cost function, and the charging and discharging cost function is: where N c is the control layer, N p is the prediction layer, u * is the reference trajectory, w is the manipulated variable, and λ is a weighting factor. The cost function of the bidirectional DC-DC converter is: In the formula, i ref (t+k) is the reference current at t+k, i L (t+k) is the actual inductor current at t+k, δ is the current tracking weight coefficient, Δu(t+k-1) is the control increment, and λ is the control smoothing weight coefficient. 8.The substation-based photovoltaic storage hybrid parallel power system coordination control method according to claim 5, characterized in that, Also including: At each sampling time, the current actual state variable is obtained, including: inductor current, DC bus voltage, battery / supercapacitor terminal voltage, state of charge; Within the prediction range, based on the minimum cost function and the control output expectation, the control sequence that minimizes the cost is solved online with the current actual state variable as the initial value to obtain the optimal control input, including the current time control amount; The current time control amount is input into the bidirectional DC / DC converter to enter the next sampling time, realizing rolling optimization.

9. An electronic device, comprising: Including: At least one processor; A memory in communication connection with the at least one processor; The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the substation-based photovoltaic storage hybrid parallel power supply system coordination control method of any one of claims 1-8.

10. A medium characterized by, A computer program is stored, and the computer program is executed by a processor to implement the substation-based photovoltaic storage hybrid parallel power supply system coordination control method of any one of claims 1-8.

Images