A photovoltaic energy storage device power distribution optimization system

Abstract

This invention relates to the field of distributed power control and intelligent auxiliary technology, specifically a photovoltaic energy storage equipment power distribution optimization system; it includes modules for state parameter acquisition, potential energy mapping processing, gradient diffusion calculation, and power execution control; the system constructs virtual fluid pressure values ​​characterizing energy flow trends by collecting node operating data; its core is to obtain the pressure of neighboring nodes and calculate the pressure gradient and rate of change parameters, driving a bidirectional converter to allow energy to flow autonomously to low-pressure nodes until equilibrium is reached; this invention realizes the transformation from global optimization calculation to distributed decision-making based on local pressure differences, the computational complexity of a single node does not increase with the network size, solves the problems of computational deadlock and response lag when large-scale energy access is implemented, and significantly improves scheduling efficiency.

Description

Technical Field

[0001] This invention relates to the field of distributed power control and intelligent auxiliary technology, specifically a power distribution optimization system for photovoltaic energy storage equipment. Background Technology

[0002] In the current distributed energy access environment, the scale of distributed control nodes such as photovoltaic power generation units and energy storage units is expanding. Each node periodically generates diverse and heterogeneous operating data, including state of charge, power generation, load power, and grid connection voltage. To achieve power balance in the distribution network, existing solutions generally adopt a centralized control strategy, which involves a central server collecting local data from each node and executing a global optimization algorithm for power allocation. Although this solution has certain scheduling capabilities in small-scale networks, the complexity of centralized computation increases exponentially when facing large-scale distributed nodes, which can easily lead to computational deadlock and instruction response lag. At the same time, this architecture is highly dependent on the communication link of the central server, which poses a significant risk of single-point failure and makes it difficult to respond to grid frequency or voltage fluctuations within milliseconds, resulting in an inherent contradiction between the global optimization objective and real-time dynamic adjustment.

[0003] Therefore, how to reduce the computational load of large-scale distributed energy nodes and improve the real-time response speed of power allocation and the system's resilience has become an urgent technical problem to be solved. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a power distribution optimization system for photovoltaic energy storage equipment. Specifically, the technical solution of this invention includes:

[0005] The status parameter acquisition module is configured to acquire the local operating data of the target distributed control node in real time. The local operating data includes: the current state of charge, the current generating power, the current load power, and the grid connection point voltage.

[0006] The potential energy mapping processing module is configured to construct a virtual fluid pressure value characterizing the energy flow trend based on the state of charge, the current power generation, and the current load power.

[0007] The gradient diffusion calculation module is configured to obtain the neighboring virtual fluid pressure values ​​of neighboring nodes adjacent to the target distributed control node through the physical communication link, and combine the virtual fluid pressure values ​​to calculate the pressure gradient parameters and pressure gradient change rate parameters within the current network area.

[0008] The power execution control module is configured to generate corresponding power adjustment commands based on the pressure gradient parameters and the pressure gradient change rate parameters.

[0009] The power execution control module drives the bidirectional converter to execute power regulation commands, enabling energy to flow autonomously from nodes with higher virtual fluid pressure values ​​to nodes with lower virtual fluid pressure values ​​until the pressure gradient parameter returns to zero.

[0010] Preferably, the potential energy mapping processing module constructs a virtual fluid pressure value characterizing the energy flow trend based on the state of charge, current power generation, and current load power, including:

[0011] Call the preset state of charge reference value;

[0012] Calculate the charge deviation component between the state of charge and the reference value of the state of charge;

[0013] Calculate the power surplus / deficit component between the current generating capacity and the current load capacity;

[0014] The virtual fluid pressure value is generated by weighted summation of the electrical deviation component and the power surplus / deficit component.

[0015] Preferably, the gradient diffusion calculation module combines virtual fluid pressure values ​​to calculate the pressure gradient parameters within the current network region, including:

[0016] Identify all adjacent nodes that have a physical connection with the target distributed control node and their connection directions;

[0017] Calculate the difference between the virtual fluid pressure value and the virtual fluid pressure value of each neighboring node;

[0018] Based on the connection direction, all the calculated differences are vector-superimposed to generate a pressure gradient parameter that characterizes the local energy potential difference.

[0019] Preferably, the gradient diffusion calculation module solves for the pressure gradient change rate parameter, including:

[0020] The pressure gradient parameters are differentiated in the time domain to generate the pressure gradient change rate parameter, which characterizes the degree of fluid turbulence.

[0021] Among them, the pressure gradient change rate parameter is used to introduce a virtual damping effect in the power regulation process to suppress the oscillation amplitude in the energy flow process.

[0022] Preferably, the power execution control module generates corresponding power adjustment commands based on the pressure gradient parameters and the pressure gradient change rate parameters, including:

[0023] Call the preset diffusion coefficient and preset viscous damping coefficient;

[0024] The basic diffusion regulation is obtained by weighting the pressure gradient parameters using the diffusion coefficient;

[0025] The inertial damping adjustment amount is obtained by weighting the pressure gradient change rate parameter using the viscous damping coefficient.

[0026] A power regulation command is generated based on the difference between the basic diffusion regulation and the inertial damping regulation.

[0027] Preferably, the power execution control module executes power regulation commands by driving the bidirectional converter, including:

[0028] Input the power adjustment command to the pulse width modulation encoder;

[0029] The power regulation command is mapped to the switching duty cycle signal of the insulated gate bipolar transistor;

[0030] In response to the switch duty cycle signal, the on / off state of the bidirectional converter is controlled to change the output power or input power of the target distributed control node.

[0031] Preferably, the system also includes:

[0032] The fault self-healing response module is configured to determine that a specific neighboring node has experienced an offline fault when the neighbor virtual fluid pressure value of a specific neighboring node cannot be obtained through the physical communication link.

[0033] The fault self-healing response module is also configured to reset the virtual fluid pressure value of a specific adjacent node to a preset low-pressure extreme value, so as to form a forced pressure gradient between the target distributed control node and the specific adjacent node, triggering an emergency support flow of energy to the fault area.

[0034] Preferably, the physical communication link includes:

[0035] The communication interface based on power line carrier technology is configured to directly load data packets containing virtual fluid pressure values ​​onto power transmission lines.

[0036] Alternatively, a bus interface based on the controller area network may be configured to broadcast data packets of virtual fluid pressure values ​​on a separate data bus.

[0037] Preferably, the target distributed control node is deployed in a digital signal processor or a field-programmable gate array edge gateway;

[0038] Digital signal processors or field-programmable gate arrays (FPGAs) edge gateways are configured to independently perform the mapping of virtual fluid pressure values ​​and the calculation of pressure gradient parameters, without relying on global scheduling instructions from a central server.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] 1. This system maps the power network into a virtual fluid pipeline network, transforming the traditional global optimization calculation into a distributed decision-making process based on local pressure differences for each control node. This approach prevents the computational complexity of a single node from increasing with the expansion of the network size, solves the problem of computational deadlock and instruction response lag that is prone to occur in the central server when large-scale distributed energy access is implemented, and significantly improves the scheduling efficiency of the system.

[0041] 2. This system adopts a decentralized self-organizing control architecture, in which each distributed node independently performs potential energy mapping and gradient calculation without relying on the global scheduling of a central server. This design eliminates the risk of single point of failure in traditional centralized systems. Even if individual nodes or communication links are damaged, the remaining nodes can still maintain power distribution balance. With the fault self-healing response mechanism, emergency support can be triggered autonomously when nodes are offline, which enhances the robustness of the system.

[0042] 3. Utilizing the high-performance computing capabilities deployed on the edge gateway, this system can respond to changes in the power grid status extremely quickly; by introducing a virtual damping effect that characterizes fluid viscosity into power control and using the pressure gradient change rate to dynamically correct the power command, the system effectively suppresses power oscillations and overshoot that may occur during the autonomous flow of energy, ensuring dynamic stability of the power distribution process while maintaining response speed.

[0043] 4. This system uses a potential energy mapping mechanism to homogenize operational data from different dimensions, such as state of charge, power generation, and load power, into a unified virtual fluid pressure value. This fusion approach can simultaneously balance the instantaneous fluctuations in photovoltaic power and the long-term energy health management of energy storage batteries, driving energy to flow spontaneously from surplus nodes to scarce nodes, thus achieving adaptive optimization of power resources in both spatial and temporal dimensions. Attached Figure Description

[0044] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0045] Figure 1 This is a structural diagram of the system of the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0047] Example 1:

[0048] Please see Figure 1 A power distribution optimization system for photovoltaic energy storage equipment, comprising:

[0049] The status parameter acquisition module is configured to acquire the local operating data of the target distributed control node in real time. The local operating data includes: the current state of charge, the current generating power, the current load power, and the grid connection point voltage.

[0050] The potential energy mapping processing module is configured to construct a virtual fluid pressure value characterizing the energy flow trend based on the state of charge, the current power generation, and the current load power.

[0051] The gradient diffusion calculation module is configured to obtain the neighboring virtual fluid pressure values ​​of neighboring nodes adjacent to the target distributed control node through the physical communication link, and combine the virtual fluid pressure values ​​to calculate the pressure gradient parameters and pressure gradient change rate parameters within the current network area.

[0052] The power execution control module is configured to generate corresponding power adjustment commands based on the pressure gradient parameters and the pressure gradient change rate parameters.

[0053] The power execution control module drives the bidirectional converter to execute power regulation commands, enabling energy to flow autonomously from nodes with higher virtual fluid pressure values ​​to nodes with lower virtual fluid pressure values ​​until the pressure gradient parameter returns to zero.

[0054] This embodiment provides a photovoltaic energy storage device power distribution optimization system. The system aims to solve the problems of computational deadlock and response lag in the existing centralized control strategy when dealing with large-scale distributed energy nodes. By mapping the power network as a virtual fluid network, this system achieves adaptive power allocation without global optimization.

[0055] The system comprises multiple distributed control nodes, each corresponding to a photovoltaic power generation unit or an energy storage unit, and is logically divided into the following core modules:

[0056] State parameter acquisition module: Its purpose is to establish the boundary conditions of the virtual fluid field; in this embodiment, the module is configured to acquire the local operating data of the target distributed control node in real time through the sensor interface.

[0057] Local operational data refers to the set of vectors describing the current physical state of a node, specifically including: the current state of charge, the current power generation, etc. Current load power and grid connection point voltage These data form the basis for subsequent potential energy calculations. The sampling frequency is set to the millisecond level to ensure the continuity of the fluid field simulation.

[0058] Potential energy mapping processing module: Its purpose is to homogenize heterogeneous power data into a single scalar field index; in this embodiment, this module is configured to execute a spatial transformation algorithm based on the above. , and Construct virtual fluid pressure values ​​to characterize energy flow trends. ;

[0059] The virtual fluid pressure value is a dimensionless or scalar with a specific dimension definition. In a physical sense, it is equivalent to pressure in fluid mechanics. The physical meaning of this value is: the more abundant the electrical charge and the greater the power surplus of a node, the higher its pressure and the stronger the tendency for energy to overflow; conversely, the lower the pressure, the stronger the tendency for energy to be absorbed.

[0060] Gradient diffusion calculation module: Its purpose is to sense local potential energy differences; in this embodiment, this module is configured to acquire the neighboring virtual fluid pressure values ​​of adjacent nodes that are directly connected to the target distributed control node in the physical topology via a physical communication link. ,in, Representing the A neighbor; combined with one's own This module calculates the pressure gradient parameters within the current network region. and pressure gradient change rate parameter ;

[0061] This process simulates the characteristic that fluid molecules are driven only by local pressure differences, thus reducing computational complexity from centralized methods. This reduces the computational cost of a single node, achieving a quasi-steady characteristic where the computational cost of a single node does not increase linearly with the expansion of the network node size.

[0062] Power execution control module: Its purpose is to transform the calculation results of the virtual field into execution actions in the physical world; in this embodiment, this module is based on and The corresponding power regulation command is generated using the discretized fluid dynamics equations. Furthermore, this module executes the command by driving the bidirectional converter to control the on / off state of the IGBT, allowing energy to flow from... Higher node direction Lower-level nodes flow autonomously until Reset to zero;

[0063] Through the collaborative work of the above modules, this system constructs an energy hydraulic system. Instead of pursuing a mathematically optimal global solution, the system utilizes the principle of least action in physics to allow energy to automatically find the path of least resistance for flow. This not only eliminates the risk of single-point failure of the central server, but also achieves millisecond-level self-healing response to power grid frequency / voltage fluctuations, resolving the inherent contradiction between global optimization and real-time response.

[0064] Example 2:

[0065] The potential energy mapping processing module constructs virtual fluid pressure values ​​characterizing the energy flow trend based on the state of charge, current power generation, and current load power, including:

[0066] Call the preset state of charge reference value;

[0067] Calculate the charge deviation component between the state of charge and the reference value of the state of charge;

[0068] Calculate the power surplus / deficit component between the current generating capacity and the current load capacity;

[0069] The virtual fluid pressure value is generated by weighted summation of the electrical deviation component and the power surplus / deficit component.

[0070] This embodiment is a concretization of the potential energy mapping processing module in Embodiment 1; in order to accurately quantify the energy potential of the nodes, this embodiment uses a weighted coupling algorithm to construct virtual fluid pressure values;

[0071] The specific execution steps of the potential energy mapping processing module are as follows:

[0072] Calling the preset state of charge reference value: Defines the ideal electrical charge level desired by the system, denoted as ;

[0073] Calculate the charge deviation component: Calculate the current state of charge. and The difference between them;

[0074] Calculate the power surplus / deficit component: Calculate the current generating power. With current load power The difference between them;

[0075] Weighted summation generates virtual fluid pressure values:

[0076] To ensure that variables from different physical dimensions can be superimposed in the same scalar field, this embodiment introduces a normalization algorithm that can eliminate dimensional differences; virtual fluid pressure value The calculation formula is defined as follows:

[0077]

[0078] Among them, the embodiments involved in this embodiment and The physical dimensions are all passed through The coefficients are normalized; meanwhile, the system's preset rated power... As a reference limit for power regulation within the system, it does not directly participate in the potential energy scalar mapping, but is used to define the range of the diffusion coefficient in the subsequent process.

[0079] Rated capacity of the target distributed control node energy storage unit, in kilowatt-hours. The purpose of introducing this variable is to convert the dimensionless state of charge into a real physical quantity of energy.

[0080] : Normalization coefficient of stock potential energy, in reciprocal kilowatt-hours This coefficient is used to convert energy deviation into dimensionless pressure potential energy. The higher the value, the higher the system's requirement for the balance of remaining power.

[0081] : Normalized coefficient of flow potential energy, in reciprocal kilowatts This coefficient is used to convert power deviation into dimensionless pressure potential energy. The larger the value, the stronger the system's ability to suppress real-time power throughput;

[0082] : Percentage of power deviation, dimensionless;

[0083] Power surplus or deficit, in kilowatts. ;

[0084] Dimensional verification instructions:

[0085] In the first term of the formula, In the second term of the formula, ;therefore, It is constructed as a pure dimensionless scalar, representing the relative potential energy height of the node in the virtual field, ensuring the physical rigor of subsequent gradient calculations;

[0086] By introducing and With two degrees of freedom, this formula can normalize variables from different physical dimensions into a unified potential energy scalar; this allows the system to cope with both instantaneous fluctuations in photovoltaic power and... This approach can also take into account the long-term health management of energy storage batteries, through... This feature prevents battery overcharging or over-discharging and enables the physical fusion of multi-source heterogeneous data.

[0087] Example 3:

[0088] The gradient diffusion calculation module, in conjunction with virtual fluid pressure values, calculates the pressure gradient parameters within the current network region, including:

[0089] Identify all adjacent nodes that have a physical connection with the target distributed control node and their connection directions;

[0090] Calculate the difference between the virtual fluid pressure value and the virtual fluid pressure value of each neighboring node;

[0091] Based on the connection direction, all the calculated differences are vector-superimposed to generate a pressure gradient parameter that characterizes the local energy potential difference.

[0092] This embodiment is a concretization of the gradient diffusion calculation module in Embodiment 2 for solving pressure gradient parameters; this embodiment uses vector analysis method to accurately calculate the driving force of local energy flow;

[0093] The specific steps for the gradient diffusion calculation module to solve for pressure gradient parameters include:

[0094] Topology identification: Identifying the set of all neighboring nodes that have a physical connection with the target distributed control node. and determine each adjacent node The corresponding connection direction is represented as a unit vector. ;

[0095] Difference calculation: Calculate the difference for this node. With each adjacent node scalar difference between ;

[0096] Vector superposition generates pressure gradient parameters:

[0097] To determine the direction and magnitude of the resultant force of energy flow, a vector synthesis is performed on the pressure differences of all neighbors; pressure gradient parameters. The calculation formula is as follows:

[0098]

[0099] in, The pressure gradient parameter is a vector whose magnitude represents the driving force of energy flow and whose direction represents the optimal energy transfer path.

[0100] Topology direction vector, derived from the network topology matrix, is used to map scalar pressure differences onto physical transmission lines; Topology direction vector During system initialization, the physical wiring relationship is preset in the edge gateway of each node. A single node can determine the vector direction of each adjacent node by reading the locally stored local topology information.

[0101] This calculation method is mathematically equivalent to solving the gradient operator of a discrete spatial field. In this way, nodes do not need to know the topology of the entire microgrid, but only need to determine which line to send energy to and how much to send based on the local pressure difference, thus realizing true decentralized self-organizing control.

[0102] Example 4:

[0103] The gradient diffusion calculation module solves for the pressure gradient change rate parameter, including:

[0104] The pressure gradient parameters are differentiated in the time domain to generate the pressure gradient change rate parameter, which characterizes the degree of fluid turbulence.

[0105] Among them, the pressure gradient change rate parameter is used to introduce a virtual damping effect in the power regulation process to suppress the oscillation amplitude in the energy flow process.

[0106] This embodiment is a concretization of the gradient diffusion calculation module in embodiment 3 for solving the pressure gradient change rate parameter; in order to introduce a viscous damping effect similar to that in fluid mechanics, this embodiment adds differential processing of the time dimension;

[0107] The gradient diffusion calculation module solves for the pressure gradient change rate parameters, including:

[0108] Time-domain differentiation: processing the pressure gradient parameters obtained in real time. Conducting discussions about time Differential operations are used to generate the pressure gradient change rate parameter. The calculation formula is as follows:

[0109]

[0110] in, The pressure gradient change rate parameter characterizes the degree of turbulence in a fluid field or the acceleration of energy flow.

[0111] The system sampling period is determined by the controller's clock setting.

[0112] Virtual damping effect: This parameter is specifically used to introduce virtual damping during power regulation; in fluid dynamics, the viscosity of a fluid hinders drastic changes in velocity; similarly, in power systems, introducing virtual damping... It can counteract sudden changes in power;

[0113] Introducing the pressure gradient change rate parameter essentially adds a differential element to the control loop; this can effectively predict the changing trend of the pressure gradient and generate a reverse viscous force when the gradient fluctuates violently, thereby suppressing the oscillation amplitude during energy flow, preventing overshoot or system instability caused by excessively fast system response, and ensuring the dynamic stability of the distribution network.

[0114] Example 5:

[0115] The power control module generates corresponding power adjustment commands based on the pressure gradient parameters and the rate of change of the pressure gradient parameters, including:

[0116] Call the preset diffusion coefficient and preset viscous damping coefficient;

[0117] The basic diffusion regulation is obtained by weighting the pressure gradient parameters using the diffusion coefficient;

[0118] The inertial damping adjustment amount is obtained by weighting the pressure gradient change rate parameter using the viscous damping coefficient.

[0119] A power regulation command is generated based on the difference between the basic diffusion regulation and the inertial damping regulation.

[0120] This embodiment is a concretization of the power execution control module generating power adjustment commands in Embodiment 4; this embodiment uses an improved diffusion equation as the control law;

[0121] The power execution control module generates the corresponding power adjustment command. include:

[0122] Parameter call: Calls the preset diffusion gain coefficient and viscous damping coefficient Based on the foregoing To define these two coefficients as dimensionless scalars, and in order to restore the output command to a power physical quantity, this embodiment provides explicit dimensional definitions for them: where, The preset value range is ,in, The rated power of the node; The value is determined based on the system sampling period. To perform tuning, the stability criterion must be met. This is to prevent system divergence caused by excessive gain of the differential term; the criterion is based on the stability boundary determination of the discrete control system and is used to ensure that the damping force generated by the differential term does not cause the controlled trajectory to oscillate and spread within the sampling period.

[0123] Diffusion gain coefficient, in kilowatts Its physical meaning is equivalent to the thermal conductivity in the heat conduction equation, which determines the energy flow rate under a unit pressure gradient.

[0124] Viscous damping coefficient, in kilowatt-seconds or kilojoules Its physical meaning is equivalent to the viscosity in a fluid, and it is used to generate a resistance force that is proportional to the rate of pressure change.

[0125] Calculation of basic diffusion regulation:

[0126] The dimensionless pressure gradient is mapped to a power flow using the diffusion coefficient:

[0127]

[0128] in, For the set of adjacent nodes; correspondingly, the inertial damping adjustment amount The treatment is the time-domain derivative of the sum of neighbor pressure differences:

[0129]

[0130] The final generated power regulation command As a scalar, its calculation formula is defined as follows:

[0131]

[0132] Among them, the above Essentially, it is the pressure gradient vector. The algebraic flux integration across all physical branch directions of the target node represents the total power demand of that node as a source or sink.

[0133] The sign of this scalar instruction represents the direction of energy flow: when When, the node performs power output; when At that time, the node performs power absorption;

[0134] Through this control algorithm, which is similar to force analysis in physics, the system can achieve control that combines speed and stability. This ensures that energy can be transferred rapidly when there is a pressure imbalance, and This ensures that the energy transfer process does not produce violent power surges, perfectly replicating the physical process of a liquid smoothly reaching liquid surface equilibrium in a communicating vessel.

[0135] Example 6:

[0136] The power execution control module executes power regulation commands by driving the bidirectional converter, including:

[0137] Input the power adjustment command to the pulse width modulation encoder;

[0138] The power regulation command is mapped to the switching duty cycle signal of the insulated gate bipolar transistor;

[0139] In response to the switch duty cycle signal, the on / off state of the bidirectional converter is controlled to change the output power or input power of the target distributed control node.

[0140] This embodiment is a specific implementation of the power execution control module driver hardware in Embodiment 1;

[0141] The process of the power execution control module driving the bidirectional converter includes:

[0142] Input pulse width modulation encoder: converts the calculated digital quantity The input is fed into the PWM generator inside the DSP;

[0143] The specific mapping logic adopts a dual closed-loop control structure, which converts the power command into a current reference value. ; where the formula introduces The coefficient is used to convert units to... The power command is restored to watt units to ensure that the calculated current reference value is accurate. The unit is ampere;

[0144] in, The duty cycle signal is calculated using a proportional-integral regulator to determine the effective value of the grid connection point voltage at the location where the target distributed control node is connected to the grid. :

[0145]

[0146] in, The feedback signal of the bidirectional converter inductor current is obtained through a current sensor. and These are the preset proportional coefficient and integral coefficient, respectively;

[0147] Control on / off: Response to signal High-frequency control of the IGBT bridge arm turns it on and off, thereby precisely adjusting the output current amplitude and phase of the target distributed control node, and thus changing its output power or input power.

[0148] This embodiment establishes a complete mapping path from the top-level virtual fluid algorithm to the bottom-level power electronic switch, ensuring that abstract mathematical calculations can be translated into concrete physical energy transmission.

[0149] Example 7:

[0150] The system also includes:

[0151] The fault self-healing response module is configured to determine that a specific neighboring node has experienced an offline fault when the neighbor virtual fluid pressure value of a specific neighboring node cannot be obtained through the physical communication link.

[0152] The fault self-healing response module is also configured to reset the virtual fluid pressure value of a specific adjacent node to a preset low-pressure extreme value, so as to form a forced pressure gradient between the target distributed control node and the specific adjacent node, triggering an emergency support flow of energy to the fault area.

[0153] This embodiment is a further improvement on the system in Embodiment 1, by adding a fault self-healing response module;

[0154] The operating logic of this module is as follows:

[0155] Offline fault diagnosis: When the physical communication link is continuously... If a specific neighboring node cannot be obtained within a certain cycle, it is determined that the node has experienced a communication interruption or is physically offline.

[0156] Virtual pressure reset and forced gradient:

[0157] The fault self-healing response module does not set the pressure value of this node. Or retain the old value, but instead... Force reset to preset low-voltage extreme value To avoid numerical calculation overflow, Defined as the negative saturation offset of the current system operating pressure average, i.e.:

[0158]

[0159] Among them, the mean function With standard deviation function It is a spatial domain statistical calculation of the virtual fluid pressure values ​​of all online adjacent nodes at the current sampling time, used to define the pressure trap depth of the fault area in real time; The set of neighboring nodes that have a physical connection with the target distributed control node;

[0160] If the set of adjacent nodes If the number of elements is less than 2, then Directly take the preset lower limit extreme value of the system global pressure;

[0161] To prevent the generation of [something] during the formation of a forced pressure gradient. Exceeding the physical limits of the hardware, this embodiment introduces a limiting operator before power execution. The corrected execution instructions are as follows:

[0162]

[0163] in, It is a symbolic function in the field of mathematics used to extract the direction of power flow; This is the maximum instantaneous power allowed by the bidirectional converter; through this limiting logic, the tendency to provide emergency support to the fault area is ensured, while preventing the control program from crashing due to the calculation results exceeding the limit.

[0164] Trigger emergency support:

[0165] According to the calculation logic of Example 3, since Extremely small, a huge pressure gradient will instantly form between the target node and the faulty node. This will cause the power execution control module to generate a full-load output command, triggering an emergency energy flow to the fault area.

[0166] This is a self-healing mechanism based on pressure traps; when a failure occurs at a certain point, that point instantly becomes a deep pit in the virtual field, and the energy of the surrounding nodes will automatically converge to provide support, just like water flowing to a low-lying area; this mechanism does not require central scheduling to issue instructions and achieves millisecond-level fault traversal capability.

[0167] Example 8:

[0168] Physical communication links include:

[0169] The communication interface based on power line carrier technology is configured to directly load data packets containing virtual fluid pressure values ​​onto power transmission lines.

[0170] Alternatively, a bus interface based on the controller area network may be configured to broadcast data packets of virtual fluid pressure values ​​on a separate data bus.

[0171] This embodiment is a specific implementation of the physical communication link in Embodiment 1;

[0172] To adapt to different application scenarios of photovoltaic energy storage systems, the physical communication link can be implemented in the following two ways:

[0173] Based on power line carrier technology, a modem is used to directly load and carry [carrier] on the power transmission line. High-frequency signal data packets;

[0174] Advantages: No additional wiring is required, enabling communication wherever there is wired connection, thus reducing construction costs;

[0175] Based on Controller Area Network (CAN) bus: A dedicated twisted-pair cable is laid, and the CAN controller broadcasts information including node IDs and other data on the bus. Data frames;

[0176] Advantages: Strong anti-interference ability, high real-time performance, suitable for containerized energy storage power stations with dense equipment;

[0177] It provides a flexible hardware interface solution, ensuring that the core state variable of virtual fluid pressure value can be reliably transmitted between physically adjacent nodes, which is the physical basis for realizing distributed collaborative control.

[0178] Example 9:

[0179] The target distributed control node is deployed in a digital signal processor or field-programmable gate array edge gateway;

[0180] Digital signal processors or field-programmable gate arrays (FPGAs) edge gateways are configured to independently perform the mapping of virtual fluid pressure values ​​and the calculation of pressure gradient parameters, without relying on global scheduling instructions from a central server.

[0181] This embodiment is a specific implementation of the target distributed control node deployment environment in Embodiment 1;

[0182] The software algorithm of the target distributed control node is programmed and run in a digital signal processor or field-programmable gate array edge gateway;

[0183] These edge computing devices are configured to perform all of the above steps independently, without relying on the global scheduling instructions of the central server;

[0184] By performing all calculations at the edge, the latency of cloud communication is eliminated, and the system response speed is improved to within 10ms. At the same time, this also means that there is no central brain in the system, and the failure of any single node will not cause the entire power distribution network to be paralyzed, which greatly improves the robustness of the system.

[0185] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A power distribution optimization system for photovoltaic energy storage equipment, characterized in that, include: The status parameter acquisition module is configured to acquire the local operating data of the target distributed control node in real time. The local operating data includes: the current state of charge, the current generating power, the current load power, and the grid connection point voltage. The potential energy mapping processing module is configured to construct a virtual fluid pressure value characterizing the energy flow trend based on the state of charge, current power generation, and current load power. The gradient diffusion calculation module is configured to obtain the neighboring virtual fluid pressure values ​​of neighboring nodes adjacent to the target distributed control node through the physical communication link, and combine the virtual fluid pressure values ​​to calculate the pressure gradient parameters and pressure gradient change rate parameters within the current network area. The power execution control module is configured to generate corresponding power adjustment commands based on the pressure gradient parameters and the pressure gradient change rate parameters. The power execution control module executes power regulation commands by driving the bidirectional converter, so that energy flows autonomously from the node with higher virtual fluid pressure value to the node with lower virtual fluid pressure value until the pressure gradient parameter returns to zero. The potential energy mapping processing module constructs virtual fluid pressure values ​​characterizing the energy flow trend based on the state of charge, current power generation, and current load power, including: Call the preset state of charge reference value; Calculate the charge deviation component between the state of charge and the reference value of the state of charge; Calculate the power surplus / deficit component between the current generating capacity and the current load capacity; The virtual fluid pressure value is generated by weighted summation of the electrical deviation component and the power surplus / deficit component. The gradient diffusion calculation module, in conjunction with virtual fluid pressure values, calculates the pressure gradient parameters within the current network region, including: Identify all adjacent nodes that have a physical connection with the target distributed control node and their connection directions; Calculate the difference between the virtual fluid pressure value and the virtual fluid pressure value of each neighboring node; Based on the connection direction, all the calculated differences are vector-superimposed to generate a pressure gradient parameter that characterizes the local energy potential difference.

2. The photovoltaic energy storage equipment power distribution optimization system according to claim 1, characterized in that, The gradient diffusion calculation module solves for the pressure gradient change rate parameter, including: The pressure gradient parameters are differentiated in the time domain to generate the pressure gradient change rate parameter, which characterizes the degree of fluid turbulence. Among them, the pressure gradient change rate parameter is used to introduce a virtual damping effect in the power regulation process to suppress the oscillation amplitude in the energy flow process.

3. The photovoltaic energy storage equipment power distribution optimization system according to claim 2, characterized in that, The power control module generates corresponding power adjustment commands based on the pressure gradient parameters and the rate of change of the pressure gradient parameters, including: Call the preset diffusion coefficient and preset viscous damping coefficient; The basic diffusion regulation is obtained by weighting the pressure gradient parameters using the diffusion coefficient; The inertial damping adjustment amount is obtained by weighting the pressure gradient change rate parameter using the viscous damping coefficient. A power regulation command is generated based on the difference between the basic diffusion regulation and the inertial damping regulation.

4. The photovoltaic energy storage equipment power distribution optimization system according to claim 1, characterized in that, The power execution control module executes power regulation commands by driving the bidirectional converter, including: Input the power adjustment command to the pulse width modulation encoder; The power regulation command is mapped to the switching duty cycle signal of the insulated gate bipolar transistor; In response to the switch duty cycle signal, the on / off state of the bidirectional converter is controlled to change the output power or input power of the target distributed control node.

5. The photovoltaic energy storage equipment power distribution optimization system according to claim 1, characterized in that, The system also includes: The fault self-healing response module is configured to determine that a specific neighboring node has experienced an offline fault when the neighbor virtual fluid pressure value of a specific neighboring node cannot be obtained through the physical communication link. The fault self-healing response module is also configured to reset the virtual fluid pressure value of a specific adjacent node to a preset low-pressure extreme value, so as to form a forced pressure gradient between the target distributed control node and the specific adjacent node, triggering an emergency support flow of energy to the fault area.

6. The photovoltaic energy storage equipment power distribution optimization system according to claim 1, characterized in that, Physical communication links include: The communication interface based on power line carrier technology is configured to directly load data packets containing virtual fluid pressure values ​​onto power transmission lines. Alternatively, a bus interface based on the controller area network may be configured to broadcast data packets of virtual fluid pressure values ​​on a separate data bus.

7. The photovoltaic energy storage equipment power distribution optimization system according to claim 1, characterized in that, The target distributed control node is deployed in a digital signal processor or field-programmable gate array edge gateway; Digital signal processors or field-programmable gate arrays (FPGAs) edge gateways are configured to independently perform the mapping of virtual fluid pressure values ​​and the calculation of pressure gradient parameters, without relying on global scheduling instructions from a central server.

Images