Optical storage power station visual management method and system based on multi-energy data fusion

Abstract

The invention relates to an optical storage power station visual management method based on multi-energy data fusion, and the method is characterized in that the method comprises the following steps: S1, collecting multi-energy heterogeneous data of a multi-energy system; s2, the collected multi-energy heterogeneous data is preprocessed; s3, modeling by using a decomposition-based multi-objective evolutionary algorithm to generate a multi-objective dynamic optimization result; s4, dynamically adjusting and optimizing the target weight for switching of different scenes; s5, generating an action strategy of energy storage charging and discharging, power generation power output and power grid power interaction according to an optimization result; and S6, carrying out visualization processing on the multi-dimensional data and the optimization result. According to the invention, the operation intelligence level and the energy management efficiency of the optical storage power station are effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent management of photovoltaic power plants, and in particular to a visual management method of photovoltaic power plants based on multi-energy data fusion. Background Art

[0002] With the acceleration of energy transformation and the advancement of the "dual carbon" goal (carbon peak and carbon neutrality) strategy, photovoltaic power generation and energy storage systems (photovoltaic storage systems) as important components of clean energy are being widely used. However, multi-energy systems include multi-source heterogeneous data such as photovoltaic power generation, energy storage, grid interaction, and environmental monitoring. These data are highly dynamic in distribution, diverse in type, and have certain uncertainties; and photovoltaic storage power stations need to dynamically adjust optimization goals in different operating scenarios (such as grid peak and valley electricity price mode, off-grid independent grid mode, etc.), such as maximizing economic benefits, improving power supply stability, or reducing carbon emissions. Therefore, data fusion, collaborative optimization, and intelligent management of multi-energy systems have become key technical directions to promote the efficient operation of photovoltaic storage power stations. Summary of the invention

[0003] In order to solve the above problems, the purpose of the present invention is to provide a photovoltaic power station visualization management method and system based on multi-energy data fusion, which effectively improves the operation intelligence level and energy management efficiency of the photovoltaic power station.

[0004] To achieve the above object, the present invention adopts the following technical solutions:

[0005] A method for visual management of a photovoltaic power station based on multi-energy data fusion, comprising the following steps:

[0006] S1: Collect multi-energy heterogeneous data of multi-energy systems, including solar power generation data, battery energy storage data, grid interaction data and environmental monitoring data;

[0007] S2: The collected multi-energy heterogeneous data are synchronously aligned, cleaned and standardized at the edge node;

[0008] S3: Based on the preprocessed multi-energy heterogeneous data, a decomposition-based multi-objective evolutionary algorithm is used to build a model and generate multi-objective dynamic optimization results;

[0009] S4: Dynamically adjust and optimize target weights for switching between different scenarios;

[0010] S5: Generate action strategies for energy storage charging and discharging, power generation output, and grid power interaction based on the optimization results;

[0011] S6: Visualize the multi-dimensional data and optimization results for managers to monitor the operating status in real time, and add a dynamic energy flow diagram to display the energy flow direction and flow size of photovoltaic power generation, energy storage charging and discharging, and grid interaction in real time.

[0012] Furthermore, data from multiple energy systems are collected, including solar power generation data, battery energy storage data, grid interaction data, and environmental monitoring data, as follows:

[0013] The solar power generation data, including the real-time power P after temperature correction PV,T 、The voltage of the photovoltaic panel V PV and current I PV , irradiance G PV and the component temperature T PV ;

[0014] Real-time photovoltaic output power calculation:

[0015] P PV =V PV I PV ·η conv ;

[0016] Among them, η conv is the inverter efficiency;

[0017] Real-time power after temperature correction:

[0018] P PV,T＝ =P PV ·[1-β·(T PV -T ref )];

[0019] Among them, T ref is the normal operating temperature of the component; β is the temperature correction coefficient;

[0020] The battery energy storage data includes the state of charge SOC, the state of health SOH, the battery pack voltage V B and current I B , battery temperature T B ;

[0021] The grid interaction data includes real-time interaction power P grid , interaction frequency f G , voltage amplitude V G ;

[0022] The environmental monitoring data includes solar irradiance G env and ambient temperature T env .

[0023] Furthermore, S2 is specifically:

[0024] Linear interpolation is used to synchronize the low-frequency data and unify the different frequency data into one time scale:

[0025]

[0026] Among them, X high ,X low are high and low frequency data respectively; t high ,t low The time point closest to the current time;

[0027] The sliding window mean filter is used to reduce the noise of high-frequency data, and outliers are eliminated based on the 3σ principle. The multi-energy heterogeneous data are integrated through the physical correlation model of photovoltaic power generation, energy storage system and grid interaction energy to ensure the consistency of the data, and finally the dimension is eliminated through standardization.

[0028] Furthermore, S3 is specifically:

[0029] S31: Decompose the multi-objective optimization problem into several single-objective sub-problems, including: minimizing the system operation cost f1, including the photovoltaic power generation cost, battery energy storage charging and discharging loss, and power grid purchase cost; maximizing the system energy efficiency f2, improving the photovoltaic power generation and energy storage utilization efficiency in the system and reducing energy loss; minimizing carbon emissions f3, and modeling based on the decomposed multi-objective optimization problem;

[0030] S32: The global Pareto optimal solution is obtained by collaboratively solving these sub-problems.

[0031] Further, S31 is as follows:

[0032] In the dynamic optimization problem of the photovoltaic energy storage multi-energy system, the decision variables include:

[0033] Photovoltaic power output distribution P PV (t); charging / discharging power P of the battery energy storage system B (t); grid interaction power P grid (t); State of charge SOC(t) of battery energy storage;

[0034] The system operation cost is minimized f1, as follows:

[0035]

[0036] in, is the unit price of electricity purchased from the power grid, is the power purchased from the grid, The price for selling electricity to the grid. is the power sold to the grid, C PV is the cost per watt of solar power generation, CB The cost of battery energy storage loss;

[0037] Maximize system energy efficiency f2, optimized by reducing energy losses, including inverter efficiency losses and charging and discharging losses:

[0038]

[0039] Among them, η pv (t) is the photovoltaic power generation efficiency; η B (t) is the charging / discharging efficiency of the energy storage system; L loss (t) is the energy loss in the system;

[0040]

[0041] Among them, E grid (t) is the carbon emission coefficient of electricity purchased per unit power;

[0042] Constraints include energy balance constraints, photovoltaic power output limits, battery charge and discharge power limits, battery SOC constraints, and grid power limits:

[0043] P PV (t)+P B (t)+P grid (t) = P load (t);

[0044] 0≤P PV (t)≤P PV，max (t);

[0045] -P B,max ≤P B (t)≤P B,max ;

[0046] SOC min ≤SOC(t)≤SOC max ;

[0047] P grid,min ≤P grid (t)≤P grid，max ;

[0048] Among them, P load (t) is the load demand power; P PV,max (t) is the maximum photovoltaic power calculated based on the irradiance; P B,max is the maximum charge / discharge power of the battery, SOC min is the minimum SOC of the battery, SOC max This is the maximum SOC of the battery.

[0049] Further, S32 is as follows:

[0050] Randomly initialize the solutions in the population, each solution is a set of decision variables: [P PV (t),P B (t),P grid (t)];

[0051] Assign a weight vector λ = (λ1, λ2, λ3) to each subproblem;

[0052] Convert the objective function into a weighted sum form:

[0053]

[0054] Based on the Euclidean distance of the weight vector, the neighborhood of each subproblem is determined to create conditions for local optimization;

[0055] Update the population using differential evolution:

[0056] Selection: Select parent individuals;

[0057] Crossover: Crossover of PPV(t), PB(t), and Pgrid(t) variables;

[0058] Mutation: Apply random perturbations to a small number of solutions to explore the solution space;

[0059] After each update, only the neighborhood associated with the current solution is updated; the non-dominated solutions of all sub-problems are recorded to generate the Pareto frontier;

[0060] Iteratively update the population until the termination condition is met and output a set of optimal solutions.

[0061] Furthermore, S4 is specifically:

[0062] Define scenario signals, including grid connection status S grid , peak and valley signal of electricity price trend S price and load urgency signal D load (t), specifically:

[0063] S grid =1 indicates grid connection, i.e. peak-valley electricity price mode; S grid =0 means no grid connection, i.e. off-grid state;

[0064] S price =1 indicates peak electricity price; S price =0 indicates off-peak electricity price;

[0065]

[0066] Define the weight vector λ of the optimization target as (λ1, λ2, λ3), where λ1 is the weight of the system operating cost f1, λ2 is the weight of the system energy efficiency f2, and λ3 is the weight of the carbon emission minimization f3;

[0067] Based on different scenario parameters, dynamically adjust the weights to generate weights for the Pareto subproblems:

[0068] λ i (t) = w i ·φ i (S grid ,S price ,D load (t));

[0069] Among them, w i is the initial weight parameter; φ i (S grid ,S price ,D load (t)) is the weight enhancement factor, which dynamically adjusts the target weight according to the scene signal.

[0070] Furthermore, the action strategies for energy storage charging and discharging, power generation output and grid power interaction are as follows:

[0071] The charging and discharging strategy of the energy storage system is dynamically adjusted according to the load power demand Pload(t), photovoltaic power generation power PPV(t) and grid power interaction Pgrid(t):

[0072] P B (t) = P load (t)-P PV (t)-P grid (t);

[0073] P B (t)>0: energy storage output power; P B (t)<0: energy storage absorption power; P B (t) = 0: energy storage does not work;

[0074] When executing the above dynamic power allocation, the state of charge (SOC) of the energy storage system is considered to ensure reasonable charging and discharging:

[0075]

[0076] Among them, C rated is the rated capacity of the battery pack; η(t) is the battery charge and discharge efficiency; Δt is the time step;

[0077] SOC limit regulation:

[0078]

[0079] Finally, the energy storage charging and discharging actions at each time step t are determined as:

[0080] If SOC(t)∈[SOC min ,SOC max ], execute PB(t); if it exceeds the SOC limit, execute P after limit adjustment B (t) adjusted ;

[0081] Photovoltaic power output dispatching strategy: Photovoltaic power output is coordinated with the energy storage system and load demand under the condition of ensuring energy utilization efficiency:

[0082] Prioritize the allocation of photovoltaic power generation to real-time loads:

[0083] P PV，load (t) = min(P load (t),P PV (t));

[0084] Among them, P PV ,load(t) is the power obtained by the real-time load;

[0085] The remaining photovoltaic power is used for energy storage charging or sold to the grid:

[0086] P PV，surplus (t) = P PV (t)-P PV，load (t)

[0087] When P PV,surplus (t)>0:

[0088] Priority charging energy storage P B (t)=-P PV,surplus (t);

[0089] If the energy storage is fully loaded SOC(t) = SOC ma , then the remaining power is input into the grid P gridsell (t) = PPV ,surplus (t);

[0090] Grid power interaction strategy, based on electricity price patterns and system energy balance status;

[0091] During off-peak electricity prices, charging power:

[0092]

[0093] At peak electricity prices, output power:

[0094]

[0095] Furthermore, S5 is specifically:

[0096] The collected multi-energy heterogeneous data, including photovoltaic power generation, energy storage SOC, energy storage charging and discharging power, grid interaction power, load demand power, system operation optimization results, etc., are monitored using charts;

[0097] The energy units in the system are represented by nodes, and the size of each node can be fixed or can be changed dynamically according to the data in real time;

[0098] Use arrows to indicate the direction of energy flow, and adjust the thickness of the arrows / lines according to the power of the flow, for example: P PV,load (t)>P PV,storage (t), the arrow from PV to load is thicker; the flow color indicates the status: normal power supply: green; power grid purchase: yellow; peak discharge: red;

[0099] The energy flow simulation is implemented by dynamically moving trajectories, which intuitively displays the flow process, recalculates the power flowing in / out of all nodes according to the preset time, updates the energy flow diagram in real time, and decomposes the energy flow into power flow based on the optimization results.

[0100] A visual management system for a photovoltaic power station based on multi-energy data fusion, comprising a processor, a memory and a computer program stored in the memory. When the processor executes the computer program, it specifically performs the steps of a visual management method for a photovoltaic power station based on multi-energy data fusion as described above.

[0101] The present invention has the following beneficial effects:

[0102] 1. The present invention effectively improves the intelligent operation level and energy management efficiency of photovoltaic power stations, while reducing costs and operation and maintenance risks. It is suitable for intelligent photovoltaic power stations with multi-energy collaborative optimization and efficient management;

[0103] 2. The present invention proposes a MOEA / D technical solution for dynamic optimization of multi-energy systems. Through target decomposition and physical association, a multi-objective optimization framework is established. By adjusting the objective function and system constraints in real time, the optimal dynamic solution set that meets the scenario requirements can be efficiently generated, providing reliable support for multi-energy scheduling and operation. And through the dynamic weight adjustment method, according to the grid connection status, peak and valley electricity price trends and real-time load pressure, the priority attributes in the multi-objective optimization problem are dynamically allocated to the corresponding weights, realizing scenario adaptive switching and optimal scheduling solution generation;

[0104] 3. The present invention automatically generates specific control strategies for energy storage charging and discharging, power generation output, and power purchase and sales from the power grid based on the optimization results, and improves operation and maintenance efficiency through energy flow dynamic diagrams and multi-dimensional data displays. BRIEF DESCRIPTION OF THE DRAWINGS

[0105] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0106] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0107] refer to Figure 1 In this embodiment, a method for visual management of a photovoltaic power station based on multi-energy data fusion is provided, including the following steps:

[0108] S1: Collect multi-energy heterogeneous data of multi-energy systems, including solar power generation data, battery energy storage data, grid interaction data and environmental monitoring data;

[0109] S2: The collected multi-energy heterogeneous data are synchronously aligned, cleaned and standardized at the edge node;

[0110] S3: Based on the preprocessed multi-energy heterogeneous data, a decomposition-based multi-objective evolutionary algorithm is used to build a model and generate multi-objective dynamic optimization results;

[0111] S4: Dynamically adjust and optimize target weights for switching between different scenarios (grid peak and valley mode, off-grid independent power supply);

[0112] S5: Generate action strategies for energy storage charging and discharging, power generation output, and grid power interaction based on the optimization results;

[0113] S6: Visualize the multi-dimensional data and optimization results for managers to monitor the operating status in real time, and add a dynamic energy flow diagram to display the energy flow direction and flow size of photovoltaic power generation, energy storage charging and discharging, and grid interaction in real time, helping operation and maintenance personnel to quickly grasp the system operation status.

[0114] In this embodiment, data of multiple energy systems are collected, including solar power generation data, battery energy storage data, grid interaction data and environmental monitoring data, as follows:

[0115] The solar power generation data, including the real-time power P after temperature correction PV,T 、The voltage of the photovoltaic panel V PV and current I PV , irradiance G PV and the component temperature T PV ;

[0116] Real-time photovoltaic output power calculation:

[0117] P PV =V PV I PV η conv ;

[0118] Among them, η conv is the inverter efficiency;

[0119] Real-time power after temperature correction:

[0120] P PV,T＝ =P PV ·[1-β·(T PV -T ref )];

[0121] Among them, T ref is the normal operating temperature of the component; β is the temperature correction coefficient;

[0122] The battery energy storage data includes the state of charge SOC, the state of health SOH, the battery pack voltage V B and current I B , battery temperature T B ;

[0123] The grid interaction data includes real-time interaction power P grid , interaction frequency f G , voltage amplitude V G ;

[0124] The environmental monitoring data includes solar irradiance G env and ambient temperature T env .

[0125] In this embodiment, S2 is specifically:

[0126] Linear interpolation is used to synchronize the low-frequency data and unify the different frequency data into one time scale:

[0127]

[0128] Among them, X high ,X low are high and low frequency data respectively; t high ,t low The time point closest to the current time;

[0129] The sliding window mean filter is used to reduce the noise of high-frequency data, and outliers are removed based on the 3σ principle; the multi-energy heterogeneous data are integrated through the physical correlation model of photovoltaic power generation, energy storage system and grid interaction energy to ensure the consistency of data, and finally the dimension is eliminated through standardization. (The physical model includes the energy balance equation of solar power generation and battery energy storage, as well as the interaction model of grid and environmental parameters)

[0130] In this embodiment, S3 is specifically:

[0131] S31: Decompose the multi-objective optimization problem into several single-objective sub-problems, including: minimizing the system operation cost f1, including the photovoltaic power generation cost, battery energy storage charging and discharging loss, and power grid purchase cost; maximizing the system energy efficiency f2, improving the photovoltaic power generation and energy storage utilization efficiency in the system and reducing energy loss; minimizing carbon emissions f3, and modeling based on the decomposed multi-objective optimization problem;

[0132] S32: The global Pareto optimal solution is obtained by collaboratively solving these sub-problems.

[0133] In this embodiment, S31 is specifically as follows:

[0134] In the dynamic optimization problem of the photovoltaic energy storage multi-energy system, the decision variables include:

[0135] Photovoltaic power output distribution P PV (t); charging / discharging power P of the battery energy storage system B (t); grid interaction power P grid (t); State of charge SOC(t) of battery energy storage;

[0136] The system operation cost is minimized f1, as follows:

[0137]

[0138] in, is the unit price of electricity purchased from the power grid, is the power purchased from the grid, The price for selling electricity to the grid. is the power sold to the grid, C PV is the cost per watt of solar power generation, C B The cost of battery energy storage loss;

[0139] Maximize system energy efficiency f2, optimized by reducing energy losses, including inverter efficiency losses and charging and discharging losses:

[0140]

[0141] Among them, η pv (t) is the photovoltaic power generation efficiency; η B (t) is the charging / discharging efficiency of the energy storage system; L loss (t) is the energy loss in the system;

[0142]

[0143] Among them, E grid (t) is the carbon emission coefficient of electricity purchased per unit power;

[0144] Constraints include energy balance constraints, photovoltaic power output limits, battery charge and discharge power limits, battery SOC constraints, and grid power limits:

[0145] P PV (t)+P B (t)+P grid (t) = P load (t);

[0146] 0≤P PV (t)≤P PV，max (t);

[0147] -P B,max ≤P B (t)≤P B,max ;

[0148] SOC min ≤SOC(t)≤SOC max ;

[0149] P grid，min ≤P grid (t)≤P grid，max ; Among them, P load (t) is the load demand power; PPV ,max (t) is the maximum photovoltaic power calculated based on the irradiance; P B,max is the maximum charge / discharge power of the battery, SOC min is the minimum SOC of the battery, SOC max This is the maximum SOC of the battery.

[0150] In this embodiment, S32 is specifically as follows:

[0151] Randomly initialize the solutions in the population, each solution is a set of decision variables: [P PV (t),P B (t),P grid (t)];

[0152] Assign a weight vector λ = (λ1, λ2, λ3) to each subproblem;

[0153] Convert the objective function into a weighted sum form:

[0154]

[0155] Based on the Euclidean distance of the weight vector, the neighborhood of each subproblem is determined to create conditions for local optimization;

[0156] Update the population using differential evolution:

[0157] Selection: Select parent individuals;

[0158] Crossover: Crossover of PPV(t), PB(t), and Pgrid(t) variables;

[0159] Mutation: Apply random perturbations to a small number of solutions to explore the solution space;

[0160] After each update, only the neighborhood associated with the current solution is updated; the non-dominated solutions of all sub-problems are recorded to generate the Pareto frontier;

[0161] Iteratively update the population until the termination condition is met and output a set of optimal solutions.

[0162] In this embodiment, S4 is specifically:

[0163] Define scenario signals, including grid connection status S grid , peak and valley signal of electricity price trend S price and load urgency signal D load (t), specifically:

[0164] S grid =1 indicates grid connection, i.e. peak-valley electricity price mode; S grid =0 means no grid connection, i.e. off-grid state;

[0165] S price =1 indicates peak electricity price; S price =0 indicates off-peak electricity price;

[0166]

[0167] Define the weight vector λ of the optimization target as (λ1, λ2, λ3), where λ1 is the weight of the system operating cost f1, λ2 is the weight of the system energy efficiency f2, and λ3 is the weight of the carbon emission minimization f3;

[0168] Based on different scenario parameters, dynamically adjust the weights to generate weights for the Pareto subproblems:

[0169] λ i (t) = w i ·φ i (S grid ,S price ,D load (t));

[0170] Among them, w i is the initial weight parameter; φ i (S grid ,S price ,D load (t)) is the weight enhancement factor, which dynamically adjusts the target weight according to the scene signal.

[0171] In this embodiment, the action strategies for energy storage charging and discharging, power generation output and grid power interaction are specifically as follows:

[0172] The charging and discharging strategy of the energy storage system is dynamically adjusted according to the load power demand Pload(t), photovoltaic power generation power PPV(t) and grid power interaction Pgrid(t):

[0173] P B (t) = P load (t)-P PV (t)-P grid (t);

[0174] P B (t)>0: energy storage output power; P B (t)<0: energy storage absorption power; P B (t) = 0: energy storage does not work;

[0175] When executing the above dynamic power allocation, the state of charge (SOC) of the energy storage system is considered to ensure reasonable charging and discharging:

[0176]

[0177] Among them, C rated is the rated capacity of the battery pack; η(t) is the battery charge and discharge efficiency; Δt is the time step;

[0178] SOC limit regulation:

[0179]

[0180] Finally, the energy storage charging and discharging actions at each time step t are determined as:

[0181] If SOC(t)∈[SOC min ,SOC max ], execute PB(t); if it exceeds the SOC limit, execute P after limit adjustment B (t) adjusted ;

[0182] Photovoltaic power output dispatching strategy: Photovoltaic power output is coordinated with the energy storage system and load demand under the condition of ensuring energy utilization efficiency:

[0183] Prioritize the allocation of photovoltaic power generation to real-time loads:

[0184] P PV,load (t) = min(P load (t), P PV (t));

[0185] Among them, P PV,load(t) is the power obtained by the real-time load;

[0186] The remaining photovoltaic power is used for energy storage charging or sold to the grid:

[0187] P PV，surplus (t) = P PV (t)-P PV，load (t)

[0188] When P PV,surplus (t)>0:

[0189] Priority charging energy storage P B (t)=-P PV,surplus (t);

[0190] If the energy storage is fully loaded SOC(t) = SOC ma , then the remaining power is input into the grid P gridsell (t) = PPV ,surplus (t);

[0191] Grid power interaction strategy, based on electricity price patterns and system energy balance status;

[0192] During off-peak electricity prices, charging power:

[0193]

[0194] At peak electricity prices, output power:

[0195]

[0196] In this embodiment, S5 is specifically:

[0197] The collected multi-energy heterogeneous data, including photovoltaic power generation, energy storage SOC, energy storage charging and discharging power, grid interaction power, load demand power, system operation optimization results, etc., are monitored using charts;

[0198] The energy units in the system are represented by nodes (such as "photovoltaic", "energy storage", "grid", "load"). The size of each node can be fixed or can change dynamically in real time according to data (such as dynamically changing the node size according to the power generation);

[0199] Use arrows to indicate the direction of energy flow (e.g., photovoltaic power generation → load, photovoltaic power generation → energy storage charging). Adjust the thickness of the arrow / line according to the flow power, for example: P PV,load (t)>P PV,storage (t), the arrow from PV to load is thicker; the flow color indicates the status: normal power supply: green; power grid purchase: yellow; peak discharge: red;

[0200] The energy flow simulation is implemented by dynamically moving trajectories, which intuitively displays the flow process, recalculates the power flowing in / out of all nodes according to the preset time, updates the energy flow diagram in real time, and decomposes the energy flow into power flow based on the optimization results.

[0201] A visual management system for a photovoltaic power station based on multi-energy data fusion, comprising a processor, a memory, and a computer program stored in the memory. When the processor executes the computer program, the steps in the visual management method for a photovoltaic power station based on multi-energy data fusion are specifically performed.

[0202] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0203] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0204] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0205] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0206] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

Claims

1. A method for visual management of a photovoltaic power station based on multi-energy data fusion, characterized in that: The following steps are involved: S1: Collect multi-energy heterogeneous data of multi-energy systems, including solar power generation data, battery energy storage data, grid interaction data and environmental monitoring data; S2: The collected multi-energy heterogeneous data are synchronously aligned, cleaned and standardized at the edge node; S3: Based on the preprocessed multi-energy heterogeneous data, a decomposition-based multi-objective evolutionary algorithm is used to build a model and generate multi-objective dynamic optimization results; S4: Dynamically adjust and optimize target weights for switching between different scenarios; S5: Generate action strategies for energy storage charging and discharging, power generation output, and grid power interaction based on the optimization results; S6: Visualize the multi-dimensional data and optimization results for managers to monitor the operating status in real time, and add a dynamic energy flow diagram to display the energy flow direction and flow size of photovoltaic power generation, energy storage charging and discharging, and grid interaction in real time.

2. The method for visual management of a photovoltaic power station based on multi-energy data fusion according to claim 1 is characterized in that: The data collected from the multi-energy system includes solar power generation data, battery energy storage data, grid interaction data and environmental monitoring data, as follows: The solar power generation data, including the real-time power P after temperature correction PV,T 、The voltage of the photovoltaic panel V PV and current I PV , irradiance G PV and the component temperature T PV ; Real-time photovoltaic output power calculation: P PV =V PV ·I PV ·or conv ; Among them, η conv is the inverter efficiency; Real-time power after temperature correction: P PV,T＝ =P PV ·[1-β·(T PV -T ref )]; Among them, T ref is the normal operating temperature of the component; β is the temperature correction coefficient; The battery energy storage data includes the state of charge SOC, the state of health SOH, the battery pack voltage V B and current I B , battery temperature T B ; The grid interaction data includes real-time interaction power P grid , interaction frequency f G , voltage amplitude V G ; The environmental monitoring data includes solar irradiance G env and ambient temperature T env .

3. The method for visual management of a photovoltaic power station based on multi-energy data fusion according to claim 2 is characterized in that: The S2 is specifically: Linear interpolation is used to synchronize the low-frequency data and unify the different frequency data into one time scale: Among them, X high ,X low are high and low frequency data respectively; t high ,t low The time point closest to the current time; The sliding window mean filter is used to reduce the noise of high-frequency data, and outliers are removed based on the 3σ principle; The multi-energy heterogeneous data are integrated through the physical correlation model of photovoltaic power generation, energy storage system and grid interaction energy to ensure the consistency of the data, and finally eliminate the dimension through standardization.

4. The method for visual management of a photovoltaic power station based on multi-energy data fusion according to claim 1 is characterized in that: The S3 is specifically: S31: Decompose the multi-objective optimization problem into several single-objective sub-problems, including: minimizing the system operation cost f1, including the photovoltaic power generation cost, battery energy storage charging and discharging loss, and power grid purchase cost; maximizing the system energy efficiency f2, improving the photovoltaic power generation and energy storage utilization efficiency in the system and reducing energy loss; minimizing carbon emissions f3, and modeling based on the decomposed multi-objective optimization problem; S32: The global Pareto optimal solution is obtained by collaboratively solving these sub-problems.

5. The method for visual management of a photovoltaic power station based on multi-energy data fusion according to claim 4 is characterized in that: The S31 is specifically as follows: In the dynamic optimization problem of the photovoltaic energy storage multi-energy system, the decision variables include: Photovoltaic power output distribution P PV (t); charging / discharging power P of the battery energy storage system B (t); grid interaction power P grid (t); State of charge SOC(t) of battery energy storage; The system operation cost is minimized f1, as follows: in, is the unit price of electricity purchased from the power grid, is the power purchased from the grid, The price for selling electricity to the grid. is the power sold to the grid, C PV is the cost per watt of solar power generation, C B The cost of battery energy storage loss; Maximize system energy efficiency f2, optimized by reducing energy losses, including inverter efficiency losses and charging and discharging losses: Among them, η pv (t) is the photovoltaic power generation efficiency; η B (t) is the charging / discharging efficiency of the energy storage system; L loss (t) is the energy loss in the system; Among them, E grid (t) is the carbon emission coefficient of purchased electricity per unit power; Constraints include energy balance constraints, photovoltaic power output limits, battery charge and discharge power limits, battery SOC constraints, and grid power limits: P PV (t)+P B (t)+P grid (t)=P load (t); 0≤P PV (t)≤P PV，max (t); -P B，max ≤P B (t)≤P B,max ; SOC min ≤SOC(t)≤SOC max ; P grid，min ≤P grid (t)≤P grid，max ; Among them, P load (t) is the load demand power; P PV，max (t) is the maximum photovoltaic power calculated based on the irradiance; P B,max is the maximum charge / discharge power of the battery, SOC min is the minimum SOC of the battery, SOC max This is the maximum SOC of the battery.

6. The method for visual management of a photovoltaic power station based on multi-energy data fusion according to claim 1 is characterized in that: The S32 is specifically as follows: Randomly initialize the solutions in the population, each solution is a set of decision variables: [P PV (t),P B (t),P grid (t)]; Assign a weight vector λ = (λ1, λ2, λ3) to each subproblem; Convert the objective function into a weighted sum form: Based on the Euclidean distance of the weight vector, the neighborhood of each subproblem is determined to create conditions for local optimization; Update the population using differential evolution: Selection: Select parent individuals; Crossover: Crossover of PPV(t), PB(t), and Pgrid(t) variables; Mutation: Apply random perturbations to a small number of solutions to explore the solution space; After each update, only the neighborhood associated with the current solution is updated; the non-dominated solutions of all sub-problems are recorded to generate the Pareto frontier; Iteratively update the population until the termination condition is met and output a set of optimal solutions.

7. The method for visual management of a photovoltaic power station based on multi-energy data fusion according to claim 6 is characterized in that: The S4 is specifically: Define scenario signals, including grid connection status S grid , peak and valley signal of electricity price trend S price and load urgency signal D load (t), specifically: S grid =1 indicates grid connection, i.e. peak-valley electricity price mode; S grid =0 means no grid connection, i.e. off-grid state; S price =1 indicates peak electricity price; S price =0 indicates off-peak electricity price; Define the weight vector λ of the optimization target as (λ1, λ2, λ3), where λ1 is the weight of the system operating cost f1, λ2 is the weight of the system energy efficiency f2, and λ3 is the weight of the carbon emission minimization f3; Based on different scenario parameters, dynamically adjust the weights to generate weights for the Pareto subproblems: λ i (t)=w i ·φ i (S grid ,S price ,D load (t)); Among them, w i is the initial weight parameter; φ i (S grid ,S price ,D load (t)) is the weight enhancement factor, which dynamically adjusts the target weight according to the scene signal.

8. The method for visual management of a photovoltaic power station based on multi-energy data fusion according to claim 1 is characterized in that: The action strategies for energy storage charging and discharging, power generation output and grid power interaction are specifically: The charging and discharging strategy of the energy storage system is dynamically adjusted according to the load power demand Pload(t), photovoltaic power generation power PPV(t) and grid power interaction Pgrid(t): P B (t)=P load (t)-P PV (t)-P grid (t); P B (t)>0: energy storage output power; P B (t)<0: energy storage absorption power; P B (t) = 0: energy storage does not work; When executing the above dynamic power allocation, the state of charge (SOC) of the energy storage system is considered to ensure reasonable charging and discharging: Among them, C rated is the rated capacity of the battery pack; η(t) is the battery charge and discharge efficiency; Δt is the time step; SOC limit regulation: Finally, the energy storage charging and discharging actions at each time step t are determined as: If SOC(t)∈[SOC min ,SOC max ], execute PB(t); if it exceeds the SOC limit, execute P after limit adjustment B (t) adjusted ; Photovoltaic power output dispatching strategy: Photovoltaic power output is coordinated with the energy storage system and load demand under the condition of ensuring energy utilization efficiency: Prioritize the allocation of photovoltaic power generation to real-time loads: P PV,load (t)=min(P load (t),P PV (t)); Among them, P PV,load (t) is the power obtained by the real-time load; The remaining photovoltaic power is used for energy storage charging or sold to the grid: P PV，surplus (t)=P PV (t)-P PV，load (t) When P PV,surplus (t)>0: Priority charging energy storage P B (t)=-P PV,surplus (t); If the energy storage is fully loaded SOC(t) = SOC ma , then the remaining power is input into the grid P gridsell (t) = PPV, surplus (t); Grid power interaction strategy, based on electricity price patterns and system energy balance status; During off-peak electricity prices, charging power: At peak electricity prices, output power:

9. The method for visual management of a photovoltaic power station based on multi-energy data fusion according to claim 1 is characterized in that: The S5 is specifically: The collected multi-energy heterogeneous data, including photovoltaic power generation, energy storage SOC, energy storage charging and discharging power, grid interaction power, load demand power, system operation optimization results, etc., are monitored using charts; The energy units in the system are represented by nodes, and the size of each node can be fixed or can be changed dynamically according to the data in real time; Use arrows to indicate the direction of energy flow, and adjust the thickness of the arrows / lines according to the power of the flow, for example: P PV,load (t)>P PV,storage (t), the arrow from PV to load is thicker; the flow color indicates the status: normal power supply: green; power grid purchase: yellow; peak discharge: red; The energy flow simulation is implemented by dynamically moving trajectories, which intuitively displays the flow process, recalculates the power flowing in / out of all nodes according to the preset time, updates the energy flow diagram in real time, and decomposes the energy flow into power flow based on the optimization results.

10. A visual management system for a photovoltaic power station based on multi-energy data fusion, characterized in that: It includes a processor, a memory and a computer program stored in the memory. When the processor executes the computer program, it specifically executes the steps in the method for visual management of a photovoltaic power station based on multi-energy data fusion as described in any one of claims 1 to 9.

Images