A control system and method for a distributed photovoltaic optimization controller

Abstract

The present invention relates to the field of controller control technology, and specifically to a control system and method of a distributed photovoltaic optimization controller, including the following steps: installing a data acquisition node on each photovoltaic component and its associated equipment to obtain voltage, current, temperature, irradiance, and component operating status parameters in real time; fusing and preprocessing data from different nodes; dividing the photovoltaic system into multiple subsystems, each subsystem is equipped with a local optimization controller to form a distributed collaborative optimization network; establishing a dynamic mathematical model of the photovoltaic system, designing a global optimization algorithm, and combining the dynamic mathematical model with the real-time data of each subsystem. The present invention can adjust the operation strategy of the photovoltaic system in real time in a complex and changeable environment, maximize the total power generation of the system, and minimize energy loss, thereby significantly improving the overall energy efficiency of the photovoltaic system.

Description

Technical Field

[0001] The present invention relates to the field of controller control technology, and in particular to a control system and method for a distributed photovoltaic optimization controller. Background Art

[0002] With the continuous growth of global energy demand and the enhancement of environmental protection awareness, renewable energy, especially photovoltaic power generation, has become the focus of the energy field. Photovoltaic power generation has been widely used due to its advantages such as cleanliness, high efficiency and pollution-free. Distributed photovoltaic systems have become an important part of photovoltaic power generation due to their flexibility and high efficiency. However, distributed photovoltaic systems still face many challenges in practical applications, especially in system optimization and control.

[0003] The output power of photovoltaic power generation systems is affected by many factors, such as light intensity, temperature, shadow, component performance differences, and grid fluctuations, which can lead to unstable system output. Traditional centralized control strategies often have difficulty achieving optimal control when faced with the complexity and dynamics of distributed systems. In addition, traditional control methods mostly rely on static models and lack real-time adaptive capabilities, making it difficult to achieve optimal control in various complex environments. Therefore, existing technologies are still insufficient in improving the overall energy efficiency and stability of distributed photovoltaic systems.

[0004] At present, the control technology of distributed photovoltaic systems mainly focuses on data acquisition and processing, system modeling and simulation, optimization algorithm design, etc. However, most of the existing technical solutions have the following problems:

[0005] Limited data processing capabilities: Traditional data processing methods are difficult to cope with the fusion and preprocessing of multi-source heterogeneous data, and data accuracy and consistency cannot be guaranteed.

[0006] Single control strategy: Most systems adopt centralized or semi-centralized control strategies, lack effective support for distributed systems, and cannot fully utilize edge computing and distributed collaborative optimization technologies.

[0007] Poor system adaptability: Traditional control methods rely on static models and lack real-time adaptive capabilities, making it difficult to maintain efficient and stable operation in a changing environment. Summary of the invention

[0008] Based on the above objectives, the present invention provides a control system and method for a distributed photovoltaic optimization controller.

[0009] A control method for a distributed photovoltaic optimization controller comprises the following steps:

[0010] S1, install data acquisition nodes on each photovoltaic module and its associated equipment, including inverters and batteries, to obtain voltage, current, temperature, irradiance, and module operating status parameters in real time, and transmit the data to the central control unit through a wireless communication network;

[0011] S2, fuses and preprocesses data from different nodes, including denoising, filtering, and normalization operations to ensure data accuracy and consistency;

[0012] S3, divides the photovoltaic system into multiple subsystems, each of which is equipped with a local optimization controller. It uses edge computing to analyze and process local data in real time, and autonomously adjusts the control strategy according to the global optimization goal of the photovoltaic system to form a distributed collaborative optimization network;

[0013] S4, based on the data of each subsystem, a dynamic mathematical model of the photovoltaic system is established. The dynamic mathematical model reflects the operating status of the photovoltaic system and changes in the external environment in real time, providing support for optimization control;

[0014] S5, designs a global optimization algorithm, combines the dynamic mathematical model and the real-time data of each subsystem, calculates the global optimal control strategy of the photovoltaic system, and transmits the strategy to each local optimization controller for execution.

[0015] Furthermore, the S1 also includes a sensor for data collection, and the sensor includes:

[0016] Voltage sensor: used to measure the output voltage of photovoltaic modules;

[0017] Current sensor: used to measure the output current of photovoltaic modules;

[0018] Temperature sensor: used to monitor the operating temperature of photovoltaic modules;

[0019] Irradiance sensor: used to detect the intensity of solar radiation;

[0020] Inverter sensor group: monitors the input and output status and operating temperature of the inverter;

[0021] Battery sensor group: monitors the voltage, current and temperature of the battery.

[0022] Furthermore, the S2 specifically includes:

[0023] S21, data denoising: Kalman filtering is used to denoise the transmitted data. Kalman filtering can effectively reduce the impact of measurement noise and improve the accuracy of data;

[0024] S22, data filtering: use a low-pass filter to filter the data to remove high-frequency noise components in the data while retaining the useful signal part to further improve the quality of the data;

[0025] S23, data normalization: normalize the filtered data and convert data from different sources and different dimensions into the same dimension range to facilitate subsequent data fusion and analysis.

[0026] Furthermore, the S3 specifically includes:

[0027] S31, system division: according to the physical location of photovoltaic modules, power load distribution and network topology, the photovoltaic system is divided into several subsystems, each of which includes several interrelated photovoltaic modules, inverters, batteries and distribution equipment;

[0028] S32, local optimization controller configuration: deploy a local optimization controller in each subsystem. The local optimization controller has data processing, edge computing and communication functions, and independently manages and optimizes the devices in the subsystem;

[0029] S33, edge computing real-time analysis: The local optimization controller uses edge computing technology to analyze and process data from local data acquisition nodes in real time, including voltage, current, temperature, and irradiance parameters. Edge computing can reduce data processing delays and improve response speed;

[0030] S34, autonomous adjustment of control strategy: The local optimization controller autonomously adjusts the control strategy according to the global optimization target and real-time data of the photovoltaic system, including adjusting the operating parameters of the photovoltaic modules, the working state of the inverter and the charging and discharging strategy of the battery to ensure the efficient operation of the subsystem;

[0031] S35, distributed collaborative optimization network: local optimization controllers share data and control strategies through wireless communication networks to form a distributed collaborative optimization network. Each local optimization controller collaboratively adjusts the control strategy based on its own and adjacent subsystems’ data to achieve global optimal control.

[0032] S36, global optimization target coordination: The central control unit regularly sends global optimization targets and strategy guidance to each local optimization controller, and the local optimization controller combines the strategy guidance with local real-time data to ensure the overall coordinated optimization of the photovoltaic system.

[0033] Furthermore, the step of establishing a dynamic mathematical model of the photovoltaic system in S4 specifically includes:

[0034] S41, obtaining real-time operation data from the local optimization controller of each subsystem, including voltage V, current I, temperature T, irradiance G, and component operation status parameters;

[0035] S42, feature parameter extraction: using machine learning algorithms and data analysis techniques to extract key feature parameters from the preprocessed data, the key feature parameters include photovoltaic power output, photovoltaic efficiency change rate, inverter efficiency change rate, battery charge and discharge characteristics, and ambient temperature changes;

[0036] S43, model construction: Combine the extracted key characteristic parameters to establish a dynamic mathematical model of the photovoltaic system. The dynamic mathematical model describes the operating state changes and output characteristics of the photovoltaic modules under different environmental conditions, including: photovoltaic module output power model, photovoltaic module efficiency model, inverter efficiency model, battery charge and discharge characteristic model and state space model;

[0037] S44, real-time update: During operation, the dynamic mathematical model is adaptively updated according to real-time data to ensure that the model can always accurately describe the operating status of the photovoltaic system and changes in the external environment.

[0038] Further, the photovoltaic module output power model is expressed as: P = G·A·η, where P is the output power of the photovoltaic module, G is the solar irradiance, A is the area of ​​the photovoltaic module, and η is the efficiency of the photovoltaic module, which depends on the temperature T and the irradiance G;

[0039] The photovoltaic module efficiency model is expressed as: η = η0 [1-β (TT ref )], where η0 is the PV module at the reference temperature (T Tef ), β is the temperature coefficient, which indicates the ratio of efficiency decrease for each degree of temperature increase, T is the actual operating temperature of the photovoltaic module, T ref is the reference temperature (set to 25°C);

[0040] The inverter efficiency model is expressed as: η inv =η inv，0 -α inv ·(P out -P nom ), where η inv is the efficiency of the inverter, η inv，0 is the inverter at nominal output power (P nom ) under the efficiency, α inv is the efficiency reduction coefficient, P out is the actual output power of the inverter;

[0041] The battery charge and discharge characteristic model is expressed as:

[0042] Among them, E bat (t) is the energy reserve of the battery at time t, E bat(t-1) is the energy reserve of the battery at the previous time point, η ch is the battery charging efficiency, P ch is the charging power, η dis is the battery discharge efficiency, P dis is the discharge power, Δt is the time interval;

[0043] The state space model is expressed as: Among them, x(t) is the system state vector, including voltage and current parameters, u(t) is the input vector, including external influencing factors such as irradiance and temperature, y(t) is the output vector, that is, the observed variable of the photovoltaic system, including power output, A, B, C, D, E, F are photovoltaic system matrices, describing the dynamic behavior of the photovoltaic system and the input-output relationship, and δ(t) is the interference term, describing the random influence of the external environment.

[0044] Furthermore, the dynamic mathematical model also includes real-time updating, which is specifically expressed as:

[0045] in, is the estimated value of the PV system state, K is the Kaldou gain used to adjust the estimated value, and y(t) is the actual observed output.

[0046] Furthermore, the design global optimization algorithm in S5 specifically includes:

[0047] S51, global optimization target determination: according to the operation requirements of the photovoltaic system and the changes in the external environment, the global optimization target is set, including maximizing the total power generation of the photovoltaic system, minimizing energy loss, and optimizing the balance between power supply and demand;

[0048] S52, integration of dynamic mathematical models and real-time data: using the dynamic mathematical models of each subsystem, combined with real-time data, including voltage V, current I, temperature T, irradiance G, and component operating status parameters, to construct the global system state vector X(t) and input vector U(t);

[0049] S53, mathematical expression of the optimization problem:

[0050] Optimization goal 1: Maximize the total power generation of the system Among them, P i (t) is the output power of the i-th PV module at time t, and N is the total number of PV modules;

[0051] Optimization goal 2: Minimize energy loss Among them, L i (t) is the energy loss of the ith subsystem, including transmission loss and conversion loss;

[0052] Optimization goal 3: Optimize the balance between electricity supply and demand:

[0053] Where D(t) is the total power demand of the system at time t;

[0054] S54, comprehensive optimization objectives:

[0055] Among them, w1, w2, w3 are the weight coefficients of the optimization target, which are adjusted according to actual needs;

[0056] S55, constraints:

[0057] Power balance constraints:

[0058] Battery charging and discharging constraints:

[0059] Among them, P ch (t) is the charging power at time t, P dis (t) is the discharge power at time t, P ch，max and P dis，max are the maximum charging and discharging powers, E bat (t) is the energy reserve of the battery at time t, E bat，max It is the maximum energy reserve of the battery.

[0060] Furthermore, S5 also includes solving the optimization problem using an optimization algorithm, specifically including:

[0061] Initialize the initial value of the parameter vector θ (such as random initialization or based on empirical values), set the learning rate α and the number of iterations K;

[0062] Iterative update: Calculate the current parameter vector θ (k) The objective function value J(θ (k) );Calculate the gradient of the objective function Update the parameter vector according to the gradient; check the convergence condition (such as the change in the objective function value is less than the preset threshold or the maximum number of iterations is reached). If the convergence condition is met, stop the iteration; otherwise, return to calculate the current parameter vector θ( k ) under the objective function value J(θ (k) );

[0063] Optimal control strategy calculation: The optimal parameter vector θ finally obtained by the optimization algorithm * , calculate the optimal control strategy for each subsystem, including the operating parameters of the PV panels, the operating status of the inverter, and the battery charging and discharging strategy;

[0064] Strategy transmission and execution: The calculated optimal control strategy is transmitted to the local optimization controller of each subsystem through the wireless communication network. The local optimization controller adjusts the operating parameters of each device in real time according to the received strategy to ensure the execution of the global optimal control strategy.

[0065] A control system of a distributed photovoltaic optimization controller, used to implement the control method of the above-mentioned distributed photovoltaic optimization controller, includes the following modules:

[0066] Data acquisition node: installed on each photovoltaic module and its associated equipment, to obtain voltage, current, temperature, irradiance, and module operating status parameters in real time, and transmit the data to the central control unit through a wireless communication network;

[0067] Data fusion and preprocessing module: fuses and preprocesses data from different nodes, including denoising, filtering, and normalization operations;

[0068] Local optimization controller: deployed in each subsystem, it analyzes and processes local data in real time through edge computing, and autonomously adjusts the control strategy according to the global optimization goal of the photovoltaic system to form a distributed collaborative optimization network;

[0069] Dynamic mathematical model module: Based on the data of each subsystem, a dynamic mathematical model of the photovoltaic system is established to reflect the operating status of the photovoltaic system and changes in the external environment in real time, providing support for optimization control;

[0070] Global optimization algorithm module: Design and implement the global optimization algorithm, combine the dynamic mathematical model and the real-time data of each subsystem, calculate the global optimal control strategy of the photovoltaic system, and pass the strategy to each local optimization controller for execution.

[0071] Beneficial effects of the present invention:

[0072] The present invention installs data acquisition nodes on each photovoltaic module and its associated equipment to obtain detailed operating parameters in real time, and fuses and preprocesses these data to ensure data accuracy and consistency. Through the precise collection and processing of real-time data, combined with dynamic mathematical models and global optimization algorithms, the operating strategy of the photovoltaic system can be adjusted in real time in a complex and changeable environment, the total power generation of the system can be maximized, and energy loss can be minimized, thereby significantly improving the overall energy efficiency of the photovoltaic system.

[0073] The present invention divides the photovoltaic system into multiple subsystems, equips each subsystem with a local optimization controller, and uses edge computing technology to analyze and process local data in real time to form a distributed collaborative optimization network. The local optimization controller autonomously adjusts the control strategy according to the global optimization goal, thereby realizing the collaborative work of the subsystems in the system. This not only improves the response speed and flexibility of the system, but also effectively reduces the delay of data transmission and processing through distributed optimization, ensuring that the system can operate stably and efficiently in various environments.

[0074] The present invention reflects the operating status of the photovoltaic system and changes in the external environment in real time by constructing a dynamic mathematical model. The dynamic mathematical model and the adaptive optimization algorithm can adjust and optimize themselves according to real-time data, so that the system has strong robustness and adaptability. In particular, in the event of an emergency or a dramatic change in the environment, the system can respond quickly and adjust the strategy to avoid performance degradation caused by faults or environmental changes, and ensure the efficient and stable operation of the photovoltaic system. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0076] Figure 1 A schematic diagram of a control method according to an embodiment of the present invention;

[0077] Figure 2 Schematic diagram of functional modules of a control system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0078] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0079] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0080] like Figure 1 As shown, a control method of a distributed photovoltaic optimization controller includes the following steps:

[0081] S1, install data acquisition nodes on each photovoltaic module and its associated equipment, including inverters and batteries, to obtain voltage, current, temperature, irradiance, and module operating status parameters in real time, and transmit the data to the central control unit through a wireless communication network;

[0082] S2, fuses and preprocesses data from different nodes, including denoising, filtering, and normalization operations to ensure data accuracy and consistency;

[0083] S3, divides the photovoltaic system into multiple subsystems, each of which is equipped with a local optimization controller. It uses edge computing to analyze and process local data in real time, and autonomously adjusts the control strategy according to the global optimization goal of the photovoltaic system to form a distributed collaborative optimization network;

[0084] S4, based on the data of each subsystem, a dynamic mathematical model of the photovoltaic system is established. The dynamic mathematical model reflects the operating status of the photovoltaic system and changes in the external environment in real time, providing support for optimization control;

[0085] S5, designs a global optimization algorithm, combines the dynamic mathematical model and the real-time data of each subsystem, calculates the global optimal control strategy of the photovoltaic system, and transmits the strategy to each local optimization controller for execution.

[0086] S1 also includes sensors for data collection, including:

[0087] Voltage sensor: used to measure the output voltage of photovoltaic modules;

[0088] Current sensor: used to measure the output current of photovoltaic modules;

[0089] Temperature sensor: used to monitor the operating temperature of photovoltaic modules;

[0090] Irradiance sensor: used to detect the intensity of solar radiation;

[0091] Inverter sensor group: monitors the input and output status and operating temperature of the inverter;

[0092] Battery sensor group: monitors the voltage, current and temperature of the battery.

[0093] S2 specifically includes:

[0094] S21, data denoising: Kalman filtering is used to denoise the transmitted data. Kalman filtering can effectively reduce the impact of measurement noise and improve the accuracy of data;

[0095] S22, data filtering: use a low-pass filter to filter the data to remove high-frequency noise components in the data while retaining the useful signal part to further improve the quality of the data;

[0096] S23, data normalization: normalize the filtered data to convert data from different sources and different dimensions into the same dimension range to facilitate subsequent data fusion and analysis. Commonly used methods include Min-Max normalization and Z-score normalization:

[0097] Min-Max normalization: scale the data to the range of [0,1];

[0098] Z-score normalization: Convert the data into a standard normal distribution to remove the differences between data of different dimensions.

[0099] S3 specifically includes:

[0100] S31, system division: according to the physical location of photovoltaic modules, power load distribution and network topology, the photovoltaic system is divided into several subsystems, each of which includes several interrelated photovoltaic modules, inverters, batteries and distribution equipment;

[0101] a. Physical location: According to the geographical location and installation location of the PV modules, physically adjacent PV modules are divided into the same subsystem to reduce the distance and delay of data transmission and improve the system response speed.

[0102] b. Power load distribution: Analyze the power load of each photovoltaic module and related equipment in the photovoltaic system, divide the components and equipment with similar or complementary power loads into the same subsystem, balance the power supply and demand within the subsystem, and improve the overall energy efficiency of the system.

[0103] c. Network topology: Based on the network connectivity of the PV system, taking into account the bandwidth and latency of the communication network, the PV modules and devices that are closely connected on the network are divided into the same subsystem to optimize data transmission efficiency and ensure real-time data exchange and collaborative control within the subsystem.

[0104] S32, local optimization controller configuration: deploy a local optimization controller in each subsystem. The local optimization controller has data processing, edge computing and communication functions, and independently manages and optimizes the devices in the subsystem;

[0105] S33, edge computing real-time analysis: The local optimization controller uses edge computing technology to analyze and process data from local data acquisition nodes in real time, including voltage, current, temperature, and irradiance parameters. Edge computing can reduce data processing delays and improve response speed;

[0106] S34, autonomous adjustment of control strategy: The local optimization controller autonomously adjusts the control strategy according to the global optimization target and real-time data of the photovoltaic system, including adjusting the operating parameters of the photovoltaic modules, the working state of the inverter and the charging and discharging strategy of the battery to ensure the efficient operation of the subsystem;

[0107] S35, distributed collaborative optimization network: local optimization controllers share data and control strategies through wireless communication networks to form a distributed collaborative optimization network. Each local optimization controller collaboratively adjusts the control strategy based on its own and adjacent subsystems’ data to achieve global optimal control.

[0108] S36, global optimization target coordination: The central control unit regularly sends global optimization targets and strategy guidance to each local optimization controller, and the local optimization controller combines the strategy guidance with local real-time data to ensure the overall coordinated optimization of the photovoltaic system.

[0109] The dynamic mathematical model of the photovoltaic system established in S4 specifically includes:

[0110] S41, obtaining real-time operation data from the local optimization controller of each subsystem, including voltage V, current I, temperature T, irradiance G, and component operation status parameters;

[0111] S42, feature parameter extraction: using machine learning algorithms and data analysis techniques to extract key feature parameters from the preprocessed data, the key feature parameters include photovoltaic power output, photovoltaic efficiency change rate, inverter efficiency change rate, battery charge and discharge characteristics, and ambient temperature changes;

[0112] S43, model construction: Combine the extracted key characteristic parameters to establish a dynamic mathematical model of the photovoltaic system. The dynamic mathematical model describes the operating state changes and output characteristics of the photovoltaic modules under different environmental conditions, including: photovoltaic module output power model, photovoltaic module efficiency model, inverter efficiency model, battery charge and discharge characteristic model and state space model;

[0113] S44, real-time update: During operation, the dynamic mathematical model is adaptively updated according to real-time data to ensure that the model can always accurately describe the operating status of the photovoltaic system and changes in the external environment.

[0114] The output power model of the photovoltaic module is expressed as: P = G·A·η, where P is the output power of the photovoltaic module, G is the solar irradiance, A is the area of ​​the photovoltaic module, and η is the efficiency of the photovoltaic module, which depends on the temperature T and the irradiance G;

[0115] The photovoltaic module efficiency model is expressed as: η=η0[1-β(TT ref )], where η0 is the PV module at the reference temperature (T Tef ), β is the temperature coefficient, which indicates the ratio of efficiency decrease for each degree of temperature increase, T is the actual operating temperature of the photovoltaic module, T ref is the reference temperature (set to 25°C);

[0116] The inverter efficiency model is expressed as: η inv =η inv，0 -α inv ·(P out -P nom ), where η inv is the efficiency of the inverter, η inv，0 is the inverter at nominal output power (P nom ) under the efficiency, α inv is the efficiency reduction coefficient, P out is the actual output power of the inverter;

[0117] The battery charge and discharge characteristic model is expressed as:

[0118] Among them, E bat (t) is the energy reserve of the battery at time t, E bat (t-1) is the energy reserve of the battery at the previous time point, η ch is the battery charging efficiency, P ch is the charging power, η dis is the battery discharge efficiency, P dis is the discharge power, Δt is the time interval;

[0119] The state space model is expressed as: Among them, x(t) is the system state vector, including voltage and current parameters, u(t) is the input vector, including external influencing factors such as irradiance and temperature, y(t) is the output vector, that is, the observed variable of the photovoltaic system, including power output, A, B, C, D, E, F are photovoltaic system matrices, describing the dynamic behavior of the photovoltaic system and the input-output relationship, and δ(t) is the interference term, describing the random influence of the external environment.

[0120] The dynamic mathematical model also includes real-time updates, which are specifically expressed as:

[0121] in, is the estimated value of the PV system state, K is the Kaldou gain used to adjust the estimated value, and y(t) is the actual observed output.

[0122] The dynamic mathematical model is composed of a photovoltaic module output power model, a photovoltaic module efficiency model, an inverter efficiency model, a battery charging and discharging characteristic model, and a state space model.

[0123] The design global optimization algorithm in S5 specifically includes:

[0124] S51, global optimization target determination: according to the operation requirements of the photovoltaic system and the changes in the external environment, the global optimization target is set, including maximizing the total power generation of the photovoltaic system, minimizing energy loss, and optimizing the balance between power supply and demand;

[0125] S52, integration of dynamic mathematical models and real-time data: using the dynamic mathematical models of each subsystem, combined with real-time data, including voltage V, current I, temperature T, irradiance G, and component operating status parameters, to construct the global system state vector X(t) and input vector U(t);

[0126] S53, mathematical expression of the optimization problem:

[0127] Optimization goal 1: Maximize the total power generation of the system Among them, P i (t) is the output power of the i-th PV module at time t, and N is the total number of PV modules;

[0128] Optimization goal 2: Minimize energy loss Among them, L i (t) is the energy loss of the ith subsystem, including transmission loss and conversion loss;

[0129] Optimization goal 3: Optimize the balance between electricity supply and demand:

[0130] Where D(t) is the total power demand of the system at time t;

[0131] S54, comprehensive optimization objectives:

[0132] Among them, w1, w2, w3 are the weight coefficients of the optimization target, which are adjusted according to actual needs;

[0133] S55, constraints:

[0134] Power balance constraints:

[0135] Battery charging and discharging constraints:

[0136] Among them, P ch (t) is the charging power at time t, P dis (t) is the discharge power at time t, P ch，max and P dis，max are the maximum charging and discharging powers, E bat (t) is the energy reserve of the battery at time t, E bat，max It is the maximum energy reserve of the battery.

[0137] S5 also includes solving the optimization problem using an optimization algorithm, specifically including:

[0138] Initialize the initial value of the parameter vector θ (such as random initialization or based on empirical values), set the learning rate α and the number of iterations K;

[0139] Iterative update: Calculate the current parameter vector θ (k) The objective function value J(θ (k) );Calculate the gradient of the objective function Update the parameter vector according to the gradient; check the convergence condition (such as the change in the objective function value is less than the preset threshold or the maximum number of iterations is reached). If the convergence condition is met, stop the iteration; otherwise, return to calculate the current parameter vector θ (k) The objective function value J(θ (k) );

[0140] Optimal control strategy calculation: The optimal parameter vector θ is finally obtained according to the optimization algorithm * , calculate the optimal control strategy for each subsystem, including the operating parameters of the PV panels, the operating status of the inverter, and the battery charging and discharging strategy;

[0141] Strategy transmission and execution: The calculated optimal control strategy is transmitted to the local optimization controller of each subsystem through the wireless communication network. The local optimization controller adjusts the operating parameters of each device in real time according to the received strategy to ensure the execution of the global optimal control strategy.

[0142] Updating the parameter vector according to the gradient is expressed as: Among them, θ (k) represents the parameter vector at the kth iteration, α is the learning rate, which controls the step size of parameter update, represents the gradient at the kth iteration.

[0143] The gradient is calculated as: in, represents the partial derivative of the objective function with respect to the jth parameter, and m represents the dimension of the parameter vector θ.

[0144] According to the above optimization objectives, the objective function is expressed as:

[0145] Among them, θ represents the parameter vector of the optimization variables, including the control parameters of the PV panels, inverters and batteries.

[0146] like Figure 2 As shown, a control system of a distributed photovoltaic optimization controller is used to implement the control method of the above-mentioned distributed photovoltaic optimization controller, including the following modules:

[0147] Data acquisition node: installed on each photovoltaic module and its associated equipment, to obtain voltage, current, temperature, irradiance, and module operating status parameters in real time, and transmit the data to the central control unit through a wireless communication network;

[0148] Data fusion and preprocessing module: fuses and preprocesses data from different nodes, including denoising, filtering, and normalization operations;

[0149] Local optimization controller: deployed in each subsystem, it analyzes and processes local data in real time through edge computing, and autonomously adjusts the control strategy according to the global optimization goal of the photovoltaic system to form a distributed collaborative optimization network;

[0150] Dynamic mathematical model module: Based on the data of each subsystem, a dynamic mathematical model of the photovoltaic system is established to reflect the operating status of the photovoltaic system and changes in the external environment in real time, providing support for optimization control;

[0151] Global optimization algorithm module: Design and implement the global optimization algorithm, combine the dynamic mathematical model and the real-time data of each subsystem, calculate the global optimal control strategy of the photovoltaic system, and pass the strategy to each local optimization controller for execution.

[0152] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity.

Claims

1. A control method for a distributed photovoltaic optimization controller, characterized in that: The following steps are involved: S1, install data acquisition nodes on each photovoltaic module and its associated equipment, including inverters and batteries, to obtain voltage, current, temperature, irradiance, and module operating status parameters in real time, and transmit the data to the central control unit through a wireless communication network; S2, fuses and preprocesses data from different nodes, including denoising, filtering, and normalization operations to ensure data accuracy and consistency; S3, divides the photovoltaic system into multiple subsystems, each of which is equipped with a local optimization controller. It uses edge computing to analyze and process local data in real time, and autonomously adjusts the control strategy according to the global optimization goal of the photovoltaic system to form a distributed collaborative optimization network; S4, based on the data of each subsystem, a dynamic mathematical model of the photovoltaic system is established. The dynamic mathematical model reflects the operating status of the photovoltaic system and changes in the external environment in real time, and provides support for optimization control. The dynamic mathematical model includes a photovoltaic module output power model, a photovoltaic module efficiency model, an inverter efficiency model, a battery charge and discharge characteristic model, and a state space model; The photovoltaic module output power model is expressed as: P = G·A·η, where P is the output power of the photovoltaic module, G is the solar irradiance, A is the area of ​​the photovoltaic module, and η is the efficiency of the photovoltaic module, which depends on the temperature T and the irradiance G; The photovoltaic module efficiency model is expressed as: η = η0 [1-β (TT ref )], where η0 is the PV module at the reference temperature (T Tef ), β is the temperature coefficient, which indicates the ratio of efficiency decrease for each degree of temperature increase, T is the actual operating temperature of the photovoltaic module, T ref is the reference temperature; The inverter efficiency model is expressed as: η inv =η inv,0 -α inv ·(P out -P nom ), where η inv is the efficiency of the inverter, η inv,0 is the inverter at nominal output power (P nom ) under the efficiency, α inv is the efficiency reduction coefficient, P out is the actual output power of the inverter; The battery charge and discharge characteristic model is expressed as: Among them, E bat (t) is the energy reserve of the battery at time t, E bat (t-1) is the energy reserve of the battery at the previous time point, η ch is the battery charging efficiency, P ch is the charging power, η dis is the battery discharge efficiency, P dis is the discharge power, Δt is the time interval; The state space model is expressed as: Among them, x(t) is the system state vector, including voltage and current parameters, u(t) is the input vector, including irradiance and temperature external factors, y(t) is the output vector, that is, the observed variable of the photovoltaic system, including power output, A, B, C, D, E, F are photovoltaic system matrices, describing the dynamic behavior of the photovoltaic system and the input-output relationship, δ(t) is the interference term, describing the random influence of the external environment; S5, designs a global optimization algorithm, combines the dynamic mathematical model and the real-time data of each subsystem, calculates the global optimal control strategy of the photovoltaic system, and transmits the strategy to each local optimization controller for execution.

2. The control method of a distributed photovoltaic optimization controller according to claim 1, characterized in that: The S1 also includes a sensor for data collection, and the sensor includes: Voltage sensor: used to measure the output voltage of photovoltaic modules; Current sensor: used to measure the output current of photovoltaic modules; Temperature sensor: used to monitor the operating temperature of photovoltaic modules; Irradiance sensor: used to detect the intensity of solar radiation; Inverter sensor group: monitors the input and output status and operating temperature of the inverter; Battery sensor group: monitors the voltage, current and temperature of the battery.

3. The control method of a distributed photovoltaic optimization controller according to claim 1, characterized in that: The S2 specifically includes: S21, data denoising: Kalman filtering is used to denoise the transmitted data; S22, data filtering: use a low-pass filter to filter the data to remove high-frequency noise components in the data; S23, data normalization: normalize the filtered data and convert data from different sources and different dimensions into the same dimension range to facilitate subsequent data fusion and analysis.

4. The control method of a distributed photovoltaic optimization controller according to claim 1, characterized in that: The S3 specifically includes: S31, system division: according to the physical location of photovoltaic modules, power load distribution and network topology, the photovoltaic system is divided into several subsystems, each of which includes several interrelated photovoltaic modules, inverters, batteries and distribution equipment; S32, local optimization controller configuration: deploy a local optimization controller in each subsystem. The local optimization controller has data processing, edge computing and communication functions, and independently manages and optimizes the devices in the subsystem; S33, edge computing real-time analysis: The local optimization controller uses edge computing technology to analyze and process data from local data acquisition nodes in real time, including voltage, current, temperature, and irradiance parameters; S34, autonomous adjustment of control strategy: the local optimization controller autonomously adjusts the control strategy according to the global optimization target and real-time data of the photovoltaic system, including adjusting the operating parameters of the photovoltaic modules, the working state of the inverter and the charging and discharging strategy of the battery; S35, distributed collaborative optimization network: local optimization controllers share data and control strategies through wireless communication networks to form a distributed collaborative optimization network. Each local optimization controller collaboratively adjusts the control strategy based on its own and adjacent subsystems’ data to achieve global optimal control. S36, global optimization target coordination: The central control unit regularly sends global optimization targets and strategy guidance to each local optimization controller, and the local optimization controller combines the strategy guidance with local real-time data to ensure the overall coordinated optimization of the photovoltaic system.

5. The control method of a distributed photovoltaic optimization controller according to claim 1, characterized in that: The establishment of the dynamic mathematical model of the photovoltaic system in S4 specifically includes: S41, obtaining real-time operation data from the local optimization controller of each subsystem, including voltage V, current I, temperature T, irradiance G, and component operation status parameters; S42, feature parameter extraction: using machine learning algorithms and data analysis techniques to extract key feature parameters from the preprocessed data, the key feature parameters include photovoltaic power output, photovoltaic efficiency change rate, inverter efficiency change rate, battery charge and discharge characteristics, and ambient temperature changes; S43, model construction: Combine the extracted key characteristic parameters to establish a dynamic mathematical model of the photovoltaic system. The dynamic mathematical model describes the operating state changes and output characteristics of the photovoltaic components under different environmental conditions; S44, real-time update: During operation, the dynamic mathematical model is adaptively updated according to real-time data to ensure that the model can always accurately describe the operating status of the photovoltaic system and changes in the external environment.

6. The control method of a distributed photovoltaic optimization controller according to claim 1, characterized in that: The dynamic mathematical model also includes real-time updating, which is specifically expressed as: in, is the estimated value of the PV system state, K is the Kalman gain used to adjust the estimated value, and y(t) is the actual observed output.

7. The control method of a distributed photovoltaic optimization controller according to claim 1, characterized in that: The design global optimization algorithm in S5 specifically includes: S51, global optimization target determination: according to the operation requirements of the photovoltaic system and the changes in the external environment, the global optimization target is set, including maximizing the total power generation of the photovoltaic system, minimizing energy loss, and optimizing the balance between power supply and demand; S52, integration of dynamic mathematical models and real-time data: using the dynamic mathematical models of each subsystem, combined with real-time data, including voltage V, current I, temperature T, irradiance G, and component operating status parameters, to construct the global system state vector X(t) and input vector U(t); S53, mathematical expression of the optimization problem: Optimization goal 1: Maximize the total power generation of the system Among them, P i (t) is the output power of the i-th PV module at time t, and N is the total number of PV modules; Optimization goal 2: Minimize energy loss Among them, L i (t) is the energy loss of the ith subsystem, including transmission loss and conversion loss; Optimization goal 3: Optimize the balance between electricity supply and demand: Where D(t) is the total power demand of the system at time t; S54, comprehensive optimization objectives: Among them, w1, w2, w3 are the weight coefficients of the optimization target; S55, constraints: Power balance constraints: Battery charging and discharging constraints: Among them, P ch (t) is the charging power at time t, P dis (t) is the discharge power at time t, P ch,max and P dis,max are the maximum charging and discharging powers, E bat (t) is the energy reserve of the battery at time t, E bat,max It is the maximum energy reserve of the battery.

8. The control method of a distributed photovoltaic optimization controller according to claim 7, characterized in that: The S5 also includes solving the optimization problem using an optimization algorithm, specifically including: Initialize the initial value of the parameter vector θ, set the learning rate α and the number of iterations K; Iterative update: Calculate the current parameter vector θ (k) The objective function value J(θ (k) );Calculate the gradient of the objective function Update the parameter vector according to the gradient; check the convergence condition, if it meets the convergence condition, stop the iteration; otherwise, return to calculate the current parameter vector θ (k) The objective function value J(θ (k) ); Optimal control strategy calculation: The optimal parameter vector θ is finally obtained according to the optimization algorithm * , calculate the optimal control strategy for each subsystem, including the operating parameters of the PV panels, the operating status of the inverter, and the battery charging and discharging strategy; Strategy transmission and execution: The calculated optimal control strategy is transmitted to the local optimization controller of each subsystem through the wireless communication network. The local optimization controller adjusts the operating parameters of each device in real time according to the received strategy to ensure the execution of the global optimal control strategy.

9. A control system of a distributed photovoltaic optimization controller, used to implement a control method of a distributed photovoltaic optimization controller as claimed in any one of claims 1 to 8, characterized in that: Includes the following modules: Data acquisition node: installed on each photovoltaic module and its associated equipment, to obtain voltage, current, temperature, irradiance, and module operating status parameters in real time, and transmit the data to the central control unit through a wireless communication network; Data fusion and preprocessing module: fuses and preprocesses data from different nodes, including denoising, filtering, and normalization operations; Local optimization controller: deployed in each subsystem, it analyzes and processes local data in real time through edge computing, and autonomously adjusts the control strategy according to the global optimization goal of the photovoltaic system to form a distributed collaborative optimization network; Dynamic mathematical model module: Based on the data of each subsystem, a dynamic mathematical model of the photovoltaic system is established to reflect the operating status of the photovoltaic system and changes in the external environment in real time, providing support for optimization control; Global optimization algorithm module: Design and implement the global optimization algorithm, combine the dynamic mathematical model and the real-time data of each subsystem, calculate the global optimal control strategy of the photovoltaic system, and pass the strategy to each local optimization controller for execution.

Images