A photovoltaic inverter applied to a photovoltaic micro-grid of an isolated island

Abstract

The application relates to the technical field of energy management, in particular to a photovoltaic inverter applied to a photovoltaic micro-grid of an isolated island, which comprises an energy state classification module; based on photovoltaic grid electricity and weather data, the energy state classification module extracts temperature and light intensity parameters, carries out data classification, and generates a state classification index. The application predicts the energy state in the photovoltaic micro-grid of the isolated island through a hidden Markov model, and improves the accuracy of energy management. The model analyzes energy output according to historical and real-time weather data, can adjust the energy storage system and the inverter output in advance when the energy output fluctuates due to weather changes, allows the system to actively store energy during the energy supply peak period, saves the use when low output is predicted, balances the energy supply and demand, and maximizes the energy efficiency. In addition, an AC / DC complementary dynamic adjustment system is driven by real-time monitoring data, and automatically selects an energy conversion path.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy management, and particularly relates to a photovoltaic inverter applied to a photovoltaic micro-grid on an isolated island. BACKGROUND

[0002] The technical field of energy management focuses on optimizing energy use efficiency and ensuring the reliability and economy of energy supply. In a photovoltaic micro-grid system, energy management involves the integration and coordination of various energy sources (such as solar energy and wind energy) and the effective use of energy storage systems. The technical goals include reducing energy waste, improving system response speed, and achieving optimal control of energy flow through intelligent scheduling and real-time monitoring.

[0003] The photovoltaic inverter applied to the photovoltaic micro-grid on the isolated island is used in the photovoltaic micro-grid system on the isolated island. It can convert the direct current generated by solar panels into alternating current suitable for power grids and household appliances. In the isolated island environment, it can also reduce dependence on external energy. The photovoltaic micro-grid not only meets the basic daily electricity needs on the island, but also serves as a backup power supply in emergency situations.

[0004] Although existing photovoltaic micro-grid systems provide power conversion functions, they are prone to slow adjustment speed and insufficient response under highly variable environmental conditions. In particular, in the closed environment of isolated islands, the independence of the power grid requires devices to respond immediately to various external changes, including weather changes and load fluctuations. However, traditional systems are difficult to effectively anticipate and prepare for sudden weather changes, resulting in a failure to meet energy needs at critical times. In addition, existing technologies are difficult to provide sufficient power immediately when the power grid load suddenly increases. For example, the power grid cannot increase output during peak demand, or reduce energy waste during low demand, limiting the efficiency of the power grid. SUMMARY

[0005] The purpose of the present application is to solve the problems existing in the prior art and to provide a photovoltaic inverter applied to a photovoltaic micro-grid on an isolated island.

[0006] To achieve the above-mentioned purpose, the present application adopts the following technical solution: a photovoltaic inverter applied to a photovoltaic micro-grid on an isolated island, the photovoltaic inverter comprising,

[0007] The energy state classification module extracts temperature and light intensity parameters based on photovoltaic grid power and weather data, classifies data, and generates a state classification index;

[0008] The energy conversion prediction module uses the state classification index to analyze the conversion frequency between each state, integrates future weather prediction data, and predicts energy state changes through probability calculation to obtain energy conversion prediction results.

[0009] The power regulation module adjusts the output parameters of the inverter item by item based on the energy conversion prediction results, synchronously adjusts the power level and frequency matching prediction status, and generates adjustment parameter settings;

[0010] The optimal path analysis module uses the adjustment parameter settings to compare the energy efficiency of AC and DC transmission, selects the transmission mode and path according to the current load of the network, and obtains the path selection result;

[0011] The output adjustment control module adjusts the output settings of the inverter based on the path selection result, updates the power output and frequency, optimizes energy distribution, and generates an output control configuration;

[0012] The energy allocation module reconfigures the charging and discharging timing to balance the load demand, optimizes the energy utilization efficiency, and generates an energy allocation plan through the output control configuration.

[0013] The present invention has the following improvements: the state classification index includes a standard threshold for data segmentation, an energy output range corresponding to each level of state, and a classification weight of an environmental parameter influencing factor; the energy conversion prediction result includes a probability distribution of each state, an expected state conversion time point, and a state duration period; the adjustment parameter setting includes the maximum and minimum power output limits of the inverter, a frequency adjustment range, and a preset power adjustment step value; the output control configuration includes an adjusted power output curve, an adjustment of the set frequency, and an inverter response time setting; the energy allocation plan includes a priority ranking of charging and discharging, an adjusted schedule, and a target energy storage amount for each time period.

[0014] The present invention is improved in that the energy status classification module includes:

[0015] The environmental monitoring submodule collects photovoltaic power grid electricity and weather data, including island temperature and light intensity, based on the isolated island's climatic conditions, and performs real-time monitoring to obtain an island environmental dataset.

[0016] The output stratification analysis submodule uses the island environment dataset to classify energy output into three levels: excellent, good, and poor according to light and temperature data, and generates stratified energy data;

[0017] The index construction submodule integrates information from the hierarchical energy data, integrates multiple key factors such as light intensity and temperature fluctuation, encodes them according to the energy output level, and generates a state classification index.

[0018] The present invention is improved in that the energy conversion prediction module includes:

[0019] The state trend analysis submodule analyzes the energy state conversion trend in an isolated island environment through a hidden Markov model based on the state classification index, evaluates the impact of seasonality and climate change on state conversion, and obtains the energy state dynamic analysis results;

[0020] The climate impact integration submodule combines the energy state dynamic analysis results with future climate condition data to generate a climate prediction model;

[0021] The conversion probability assessment submodule uses the climate prediction model and the hidden Markov model to analyze the conversion probability of energy states under given environmental conditions to obtain energy conversion prediction results.

[0022] The present invention is improved in that the hidden Markov model is according to the formula:

[0023]

[0024] Among them, π k represents the probability of state k in the initial observation period, n k is the number of occurrences of state k in the initial observation period, α′ is the weight coefficient of the influence of light intensity on the state, S k is the light intensity adjustment factor corresponding to state k, N is the total number of observations, and ∑S is the sum of the light intensity adjustment factors of all states;

[0025] and

[0026]

[0027] Analyze the conversion trend of energy status in isolated island environments;

[0028] Among them, a jk is the state transition probability matrix, which represents the probability of transitioning from state j to state k, v is the time index, W is the end point of the total observation period, x vj is the indicator function of state j at time v, x (v+1)k is the indicator function of state k at time v+1, β′ is the weight coefficient of temperature influence, P vj is the temperature influence factor at time v state j, γ′ is the weight coefficient of temperature influence, P (v+1)k is the temperature influence factor of state k at time v+1, and δ′ is used to adjust the weight of temperature influence in the calculation of state transition probability.

[0029] The present invention is improved in that the power regulation module includes:

[0030] The energy demand matching submodule analyzes the current energy demand of the isolated island based on the energy conversion prediction result, adjusts the power output of the inverter, and generates power matching parameters;

[0031] The frequency corresponding submodule uses the power matching parameters to adjust the inverter frequency to match the updated power setting to obtain the frequency adaptation setting;

[0032] The output adjustment submodule utilizes the frequency adaptation setting to optimize the overall output configuration of the inverter, including output power regulation and energy distribution, and generates adjustment parameter settings.

[0033] The present invention is improved in that the optimal path analysis module includes:

[0034] The transmission efficiency analysis submodule compares the energy efficiency of AC and DC transmission modes in the photovoltaic power grid of the isolated island based on the adjustment parameter settings, analyzes their respective energy losses and transmission efficiencies, and generates photovoltaic power grid transmission efficiency data;

[0035] The network load evaluation submodule uses the photovoltaic power grid transmission efficiency data to analyze the current power grid load situation and load fluctuation characteristics, compares the performance of various transmission modes, and generates load status analysis results;

[0036] The energy transmission optimization submodule selects an energy transmission path that matches the needs of the isolated island power grid based on the load condition analysis results through a path selection algorithm, including comparing the energy efficiency and stability of each path and generating a path selection result.

[0037] The present invention is improved in that the path selection algorithm is according to the formula:

[0038] I path =ω1·η path +ω2·ΔP path +ω3·W f +ω4·F r +ω5·S f +ω6·R v

[0039] Calculate the optimal path selection index I path ;

[0040] Among them, ω1, ω2, ω3, ω4, ω5 and ω6 are weight coefficients, which are used to balance the impact of energy efficiency, power stability, weather impact, equipment failure rate, seasonal variation factor, and renewable energy output volatility, respectively. path is the energy efficiency of the path, ΔP path is the power stability of the path, W f is the weather influencing factor, F r is the equipment failure rate, S f is the seasonal variation factor, R v The output volatility of renewable energy.

[0041] The present invention is improved in that the output adjustment control module includes:

[0042] The power adjustment submodule reconfigures the power output parameters of the inverter based on the path selection result, including changing the power level to match the needs of the island power grid, and generates an updated power configuration.

[0043] The frequency calibration submodule utilizes the updated power configuration to iteratively adjust the frequency of the inverter, including calibrating the output frequency to match the updated power parameters, to obtain a frequency synchronization configuration.

[0044] The distribution optimization submodule adjusts and optimizes the energy distribution of the inverter by using the frequency synchronization configuration and referring to the current energy output and demand conditions, and generates an output control configuration.

[0045] The present invention is improved in that the energy allocation module includes:

[0046] The timing optimization submodule rearranges the charging and discharging schedules based on the output control configuration to generate an optimized charging and discharging schedule.

[0047] The energy balance submodule uses the optimized charge and discharge schedule to analyze the load situation of the power grid, adjust the charge and discharge priority according to the energy supply situation, and generate a load management plan.

[0048] The equipment efficiency optimization submodule optimizes the operation and scheduling mode of the energy storage equipment through the load management plan and generates an energy distribution plan.

[0049] Compared with the prior art, the advantages and positive effects of the present invention are:

[0050] In the present invention, the energy status in isolated island photovoltaic microgrids is predicted by using a hidden Markov model, thereby improving the accuracy of energy management. The model analyzes energy output based on historical and real-time weather data, and can adjust the energy storage system and inverter output in advance when weather changes cause energy output fluctuations. It allows the system to actively store energy during peak energy supply periods and save energy when low output is predicted, thereby balancing energy supply and demand and maximizing energy efficiency. In addition, the AC / DC complementary dynamic regulation system is driven by real-time monitoring data and automatically selects the energy conversion path. It can also quickly adjust output parameters such as power and frequency to ensure the continuity of power supply and the stable operation of the system, thereby improving the system's adaptability to rapidly changing environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 The present invention proposes a module diagram of a photovoltaic inverter for a photovoltaic microgrid applied to an isolated island;

[0052] Figure 2The present invention proposes a system framework diagram of a photovoltaic inverter for a photovoltaic microgrid applied to an isolated island;

[0053] Figure 3 is a schematic diagram of the energy status classification module of the present invention;

[0054] Figure 4 Schematic diagram of the energy conversion prediction module of the present invention;

[0055] Figure 5 is a schematic diagram of a power regulation module of the present invention;

[0056] Figure 6 is a schematic diagram of the optimal path analysis module of the present invention;

[0057] Figure 7 A schematic diagram of an output adjustment control module of the present invention;

[0058] Figure 8 Schematic diagram of the energy allocation module of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0060] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings and are only for the convenience of describing the present invention and simplifying the description. They do not indicate or imply that the devices or elements referred to must have a specific direction, be constructed and operate in a specific direction, and therefore should not be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0061] See also Figure 1 The present invention provides a technical solution: a photovoltaic inverter for a photovoltaic microgrid on an isolated island, the photovoltaic inverter comprising:

[0062] The energy status classification module extracts multiple parameters such as temperature and light intensity based on photovoltaic grid power and weather data, classifies the data, and generates a status classification index;

[0063] The energy conversion prediction module uses state classification indexing to analyze the conversion frequency between each state, integrates future weather forecast data, and predicts energy state changes through probability calculation to obtain energy conversion prediction results;

[0064] The power adjustment module adjusts the output parameters of the inverter item by item based on the energy conversion prediction result, synchronously adjusts the power level and frequency matching prediction state, and generates adjustment parameter settings;

[0065] The optimal path analysis module compares the energy efficiency of AC and DC transmission based on the adjustment parameter settings, selects the transmission mode and path according to the current load of the network, and obtains path selection results;

[0066] The output adjustment control module adjusts the output settings of the inverter based on the path selection results, updates the power output and frequency, optimizes the energy distribution, and generates output control configurations;

[0067] The energy allocation module reconfigures the charging and discharging time sequence to balance the load demand based on the output control configuration, optimizes the energy use efficiency, and generates an energy distribution plan.

[0068] The state classification index includes standard threshold for data segmentation, energy output range corresponding to each level of state, and classification weight of environmental parameter influence factor. The energy conversion prediction result includes probability distribution of each state, expected state conversion time point, and state duration cycle. The adjustment parameter settings include maximum and minimum power output limits of the inverter, frequency adjustment range, and preset power adjustment step value. The output control configuration includes adjusted power output curve, set frequency adjustment, and inverter response time setting. The energy distribution plan includes priority order of charging and discharging, adjusted schedule, and target energy storage amount of each time period.

[0069] Please refer to Figure 2 and Figure 3 , the energy state classification module includes:

[0070] The environmental monitoring sub-module collects photovoltaic power grid power and weather data based on the climate conditions of isolated islands, including the temperature and light intensity of the island, and performs real-time monitoring of the data to obtain the island environment data set as follows:

[0071] Under the specific climate conditions of isolated islands, the temperature and light intensity of the island are systematically collected and analyzed. These data are collected and transmitted to the central database in real time. When collecting data, temperature sensors and light-sensitive sensors are used, and the data collection frequency is set to once every minute. In addition, the module is provided with a data cleaning function to automatically remove outliers. In this process, real-time data T i and S i are recorded, where i represents the time point of collection. The collected temperature and light data are standardized according to their average values T avg and standard deviations S sd for further analysis.

[0072] The output stratification analysis submodule uses the island environment dataset to classify energy output into three levels: excellent, good, and poor based on light and temperature data. The process of generating stratified energy data is as follows;

[0073] Using the standardized temperature and light data, energy output is divided into three levels: "excellent", "good" and "poor". The specific stratification is achieved by setting the threshold θ T and θ S Score the data points. To do this, define the hierarchical scoring formula: According to R score The value of the R is assigned a grade by a text description, such as a score of more than 100 is considered "excellent", between 50 and 100 is "good", and below 50 is "poor". score is the scoring result of the data point. T is the threshold of temperature data. S is the threshold of illumination data.

[0074] The index construction submodule integrates information from hierarchical energy data, integrates multiple key factors such as light intensity and temperature fluctuation, encodes them according to the energy output level, and generates a state classification index in the following process;

[0075] Integrate information from hierarchical energy data and encode key factors such as temperature fluctuation and light intensity to generate a state classification index. Each state is classified according to its score result R score Assign a code C k , to facilitate quick retrieval and classification of states, the coding is calculated as follows: By dividing the score by 10 and taking the integer part, a code from 0 to 10 is generated to facilitate subsequent data processing and analysis. k According to the scoring result R score The calculated status code.

[0076] See also Figure 2 and Figure 4 , the energy conversion prediction module includes:

[0077] The state trend analysis submodule uses the hidden Markov model to analyze the energy state conversion trend in an isolated island environment based on the state classification index, evaluates the impact of seasonality and climate change on state conversion, and obtains the energy state dynamic analysis results in the following process:

[0078] The transition probabilities between states are calculated based on environmental monitoring data. The state space is defined as the set of all possible energy states, and the initial state probability distribution and state transition probability matrix are calculated.

[0079] The hidden Markov model follows the formula:

[0080]

[0081] Among them, π k represents the probability of state k in the initial observation period, n k is the number of occurrences of state k in the initial observation period, α′ is the weight coefficient of the influence of light intensity on the state, which adjusts the contribution of light to the state probability, S k is the light intensity adjustment factor corresponding to state k, indicating the lighting conditions in this state, N is the total number of observations, and ∑S is the sum of the light intensity adjustment factors of all states, which is used to normalize the influence of light intensity;

[0082] and

[0083]

[0084] Analyze the conversion trend of energy status in isolated island environments;

[0085] Among them, a jk is the state transition probability matrix, which represents the probability of transitioning from state j to state k, v is the time index used to traverse the observation data, W is the end point of the total observation time period, x vj is the indicator function of state j at time v, indicating that state j is observed at time point v, x (v+1)k is the indicator function of state k at time v+1, indicating that state k is observed at time point v+1, β′ is the weight coefficient of temperature influence, which adjusts the influence of the temperature factor at time point v on state j, P vj is the temperature influence factor of state j at time v, which indicates the temperature condition of state j at that time point. γ′ is the weight coefficient of temperature influence, which adjusts the influence of the temperature factor at time point v+1 on state k. P (v+1)k is the temperature influence factor of state k at time v+1, indicating the temperature condition of state k at that time point. δ′ is used to adjust the weight of temperature influence in the state transition probability calculation, affecting the temperature adjustment of states at all time points.

[0086] The calculation process is as follows:

[0087] Set the environment and equipment characteristic parameters:

[0088] Assume that the light intensity adjustment factor S for each state k Obtained through the accumulation of long-term environmental monitoring data. The calculation formula is

[0089]

[0090] s kd is the light intensity in state k on day d, and D is the number of days.

[0091] Temperature influence factor P vj and P (v+1)kCalculated in a similar way:

[0092]

[0093] p vjh is the temperature at state j at hour h, where H is the number of hours.

[0094] State transition probability matrix calculation:

[0095] Define the original state transition probability matrix a jk .

[0096]

[0097] The weight coefficients β′, γ′, and δ′ were determined by regression analysis, analyzing the correlation between state transitions and environmental factors based on past data.

[0098] State transition probability matrix a jk The model more accurately reflects the probability of transitioning from state j to state k, taking into account the influence of environmental factors such as temperature and light. By incorporating actual environmental monitoring data, the model can better predict the state transition probability under different environmental conditions, thereby improving the responsiveness and efficiency of the energy management system.

[0099] The climate impact integration submodule combines the results of the energy state dynamic analysis with the future climate condition data to generate a climate prediction model.

[0100] Analyze the energy state transition probability under different climate conditions in historical data and infer the potential impact of future climate change on energy state. Climate prediction model calculation formula: b jk =β0+β1c k +β2c′ k , b jk Indicated by the climate factor c k and future predicted climate factor c′ k The predicted probability of transitioning from state j to state k under the influence of β0, β1 and β2 are regression coefficients obtained by training historical climate and state data. k and c′ k represent current and predicted climate conditions, respectively.

[0101] The conversion probability assessment submodule uses the climate prediction model and the hidden Markov model to analyze the conversion probability of energy states under given environmental conditions, and the process of obtaining the energy conversion prediction result is as follows:

[0102] Analyze the transition probability of energy states under given environmental conditions. Adjust the initial state transition probability matrix to reflect the expected impact of climate change on energy state transition. State transition probability formula: is the adjusted transition probability from state j to state k, γ k is a tuning parameter that reflects the sensitivity of climate prediction to the probability of state transition, b jk Forecast results from the climate impact integration submodule. This approach refines the original probabilistic model and provides more accurate energy state transition forecasts.

[0103] See also Figure 2 and Figure 5 , the power regulation module includes:

[0104] The energy demand matching submodule analyzes the current energy demand of the isolated island based on the energy conversion prediction results, adjusts the power output of the inverter, and generates the power matching parameters in the following process:

[0105] Using the output of the forecast model, the module calculates the average expected demand D by summarizing the energy demand of each state and the corresponding state probability: D = ∑ k E k ·p k , E k is the energy demand per unit time in state k, p k is the probability that the system is in state k. State probability p k It is derived from the energy state transition model and is calculated as follows: According to the average expected demand D and the capacity limit of the equipment, the output power Q of the inverter is adjusted as follows: Q = min (Q max ,K1·D),Q max is the maximum output capacity of the inverter, and K1 is the regulation coefficient, which is used to adjust the output power according to current demand.

[0106] The frequency corresponding submodule uses the power matching parameters to adjust the inverter frequency to match the updated power setting. The process of obtaining the frequency adaptation setting is as follows:

[0107] After adjusting the inverter's output power, the inverter's output frequency is adjusted to accommodate the updated power setting using the power matching parameter Q. The frequency is adjusted using the following formula: F base is the reference frequency of the inverter, Q nom is the nominal power. This ensures that the inverter's output frequency is synchronized with the power demand, thereby optimizing overall energy efficiency and equipment performance.

[0108] The output adjustment submodule uses frequency adaptation settings to optimize the overall output configuration of the inverter, including output power regulation and energy distribution. The process of generating adjustment parameter settings is as follows:

[0109] Optimize the inverter's overall output configuration, encompassing both power regulation and energy distribution. The adjustment parameter setting, G, is calculated using the following formula: G = α²·F + β²·Q, where α² and β² are weighting coefficients that adjust the effect of frequency F on the output configuration and the effect of power Q on the output configuration, respectively.

[0110] See also Figure 2 and Figure 6 , the optimal path analysis module includes:

[0111] The transmission efficiency analysis submodule compares the energy efficiency of AC and DC transmission modes in the photovoltaic power grid of the isolated island based on the adjustment parameter settings, analyzes their respective energy losses and transmission efficiencies, and generates the photovoltaic power grid transmission efficiency data in the following process:

[0112] The energy efficiency of AC and DC transmission in the photovoltaic power grid of isolated islands is compared. By analyzing the energy loss and transmission efficiency of each transmission method, the energy loss L of AC and DC transmission is defined. AC and L DC , and calculate the transmission efficiency η as follows: and η AC and η DC Represents the transmission efficiency of AC and DC respectively. E in is the total energy input into the system. AC and L DC Represent the energy loss of AC and DC during transmission respectively.

[0113] The network load evaluation submodule uses the PV grid transmission efficiency data to analyze the current grid load and load fluctuation characteristics, compares the performance of various transmission methods, and generates the load condition analysis results in the following process:

[0114] Using the above transmission efficiency data, we can analyze the load situation and load fluctuation characteristics of the current power grid. By comparing the performance of various transmission methods, we can calculate the current load C load and the predicted load P load The relationship between: and P load =∑ n σ n ·P n , C load Indicates the current actual load, which is determined by the power P of each node n With its maximum capacity P max,n The sum of the ratios of P load Represents the predicted load, which is determined by the power fluctuation σ of each node n and power P n The sum of the products of n and n is the grid node index.

[0115] The energy transmission optimization submodule selects the energy transmission path that matches the needs of the isolated island power grid based on the load condition analysis results through the path selection algorithm, including comparing the energy efficiency and stability of each path. The process of generating the path selection result is as follows;

[0116] The path selection algorithm selects the energy transmission path that matches the needs of the isolated island power grid. The energy efficiency and stability of each path are compared to obtain the optimal path selection index.

[0117] The path selection algorithm is based on the formula:

[0118] I path =ω1·η path +ω2·ΔP path +ω3·W f +ω4·F r +ω5·S f +ω6·R v

[0119] Calculate the optimal path selection index I path ;

[0120] Among them, I path It is used to evaluate the overall performance of different transmission paths. ω1, ω2, ω3, ω4, ω5, and ω6 are weight coefficients, which are used to balance the impact of energy efficiency, power stability, weather impact, equipment failure rate, seasonal variation factor, and renewable energy output volatility, respectively. η path is the energy efficiency of the path, which measures the energy transmission efficiency, ΔP path To evaluate the power stability of the path, the continuity and reliability of energy supply are evaluated, W f is the weather impact factor, which indicates the potential impact of different weather conditions on energy transmission efficiency and stability. r is the equipment failure rate, which measures the frequency of equipment problems that may affect the reliability of energy supply, S f is the seasonal variation factor, reflecting the impact of seasonal changes on energy demand and supply, R v The volatility of renewable energy output refers to the uncertainty and fluctuation of renewable energy output due to weather changes or other factors.

[0121] The calculation process is as follows:

[0122] Determine the value of the impact factor:

[0123] Weather impact factor W f : By analyzing the correlation between historical weather data and energy output, a linear regression model is used to estimate the impact of weather conditions on energy output.

[0124] Assume a wtrepresents the weather conditions at time t, b yt Representing the corresponding energy output, the linear model can be expressed as:

[0125] b yt =μ+θa wt

[0126] θ (weather factor influence coefficient) is calculated by the least square method, then W f =θ.

[0127] Equipment failure rate F r : Calculated by counting the number of equipment failures and the total operating time in the past year. Let f t is the number of failures, h t The total running time is:

[0128]

[0129] Seasonal variation factor S f : Analyze the energy demand change data of each season and use the simple average method to calculate the energy demand ratio of different seasons. Assume that the energy demand in spring is e sp , total demand is e tot ,but:

[0130]

[0131] Renewable energy output volatility R v : Calculate the standard deviation of the output of renewable energy (such as wind and solar) as a measure of volatility. Let r t is the energy output in a specific time, and the average output is but:

[0132]

[0133] Calculate evaluation index I path :

[0134] I path =ω1·η path +ω2·ΔP path +ω3·W f +ω4·F r +ω5·S f +ω6·R v

[0135] The weighting coefficients ω1 to ω6 are adjusted based on historical performance and expert experience to ensure that all factors are appropriately considered according to their importance.

[0136] Evaluation Index I pathIt provides a quantitative indicator to help decision makers comprehensively consider energy efficiency, stability, weather impacts, equipment reliability, seasonality, and the volatility of renewable energy when selecting the most suitable energy transmission path. This comprehensive assessment ensures the efficient and stable operation of the power grid.

[0137] See also Figure 2 and Figure 7 , the output adjustment control module includes:

[0138] The power adjustment submodule reconfigures the inverter’s power output parameters based on the path selection results, including changing the power level to match the island grid’s needs. The process of generating the updated power configuration is as follows:

[0139] Reconfigure the inverter's power output parameters to match the island grid's needs. path and the current energy demand D, the power level Q of the inverter is reconfigured as follows: Q new =Q base ×(1+r·(DD base )), Q new This is the updated inverter power configuration. base is the basic power level, i.e. the power output before adjustment. D is the current energy demand. base is the baseline energy demand, typically the long-term average demand. r is the regulation factor, which adjusts power output to match demand fluctuations. This ensures that the inverter's output power matches current and expected grid demand, optimizing power output.

[0140] The frequency calibration submodule uses the updated power configuration to iteratively adjust the inverter frequency, including calibrating the output frequency to match the updated power parameters. The process of obtaining the frequency synchronization configuration is as follows:

[0141] Using the updated power profile Q new , iteratively adjust the inverter frequency. The new frequency setting F new Adjusted by the following formula: F new is the inverter frequency after adjustment. base is the base frequency, i.e., the frequency before adjustment. φ is the frequency adjustment factor, which adjusts the frequency based on power changes. By taking into account changes in power configuration, the inverter frequency is updated synchronously to ensure that the frequency matches the power output.

[0142] The allocation optimization submodule adjusts and optimizes the energy allocation of the inverter by configuring the frequency synchronization and referring to the current energy output and demand. The process of generating the output control configuration is as follows:

[0143] Configure F through frequency synchronization new, refer to the current energy output and demand situation, adjust and optimize the energy distribution of the inverter. The output control configuration G is calculated by the following formula: G = α·F new +β·Q new , G is the output control configuration, used to adjust the inverter's output settings. α and β are weighting coefficients used to adjust the impact of frequency and power on the final output. This ensures that the inverter's output configuration precisely matches the current grid demand and operating conditions, optimizing the performance and efficiency of the entire system.

[0144] See also Figure 2 and Figure 8 , the energy allocation module includes:

[0145] The timing optimization submodule rearranges the charging and discharging schedules based on the output control configuration, and the process of generating an optimized charging and discharging schedule is as follows;

[0146] Based on the output control configuration G, the charging and discharging schedule is rearranged to adapt to the dynamically changing energy demand and output. The schedule is adjusted by adjusting the coefficient ξ: T new =T old +ξ·G,T new is the optimized charging and discharging schedule, which indicates the new arrangement time of charging and discharging activities. old is the original charge and discharge schedule, representing the previously scheduled charge and discharge times. ξ is the time adjustment factor, used to adjust the schedule based on the output control configuration G, indicating the sensitivity or magnitude of the time adjustment. G is the output control configuration, indicating how closely the current output setting matches demand. Ensure that the charge and discharge schedule is synchronized with the inverter's output control configuration, optimizing the charge and discharge process to suit current energy output and demand.

[0147] The energy balance submodule uses the optimized charging and discharging schedule to analyze the grid load and adjust the charging and discharging priority according to the energy supply situation. The process of generating the load management plan is as follows;

[0148] Using the optimized charge and discharge schedule T new , analyze the actual load of the power grid and adjust the charging and discharging priority to optimize the overall energy balance of the power grid. The load management index Λ is calculated as follows: Λ=∑ i ω i ·(P i -P avg ), Λ is used to evaluate and adjust the charge and discharge priority. i is the grid load in time period i, which represents the actual grid usage in a specific time period. avg is the average grid load, which indicates the average load level of the grid in the past period of time. iThe weight coefficients are assigned based on the importance or priority of time periods, indicating the importance of different time periods. Through these calculations, load management plans are generated to optimize charging and discharging priorities to balance energy supply and demand across the system, reduce peak load pressure, and improve system stability.

[0149] The equipment efficiency optimization submodule optimizes the operation and scheduling of energy storage equipment through load management planning, and the process of generating energy distribution plan is as follows:

[0150] Based on the load management plan Λ, the operation and scheduling of energy storage equipment are optimized. The energy distribution plan is calculated by the equipment efficiency optimization index ∈ to ensure the maximum energy efficiency of the system. The equipment efficiency optimization index ∈ is calculated as follows: ∈ = ∑ m λ m ·ΔE m ,∈ is used to evaluate the energy efficiency improvement under different configurations. ΔE m λ is the energy saving or additional consumption under configuration m, which indicates the energy saved or consumed more under this configuration compared with the standard configuration. m Efficiency weights reflect the contribution of different configurations to overall efficiency, indicating the relative importance or effectiveness of each configuration. This ensures that energy storage equipment operates in an optimized manner to achieve the most efficient use of energy, reduce unnecessary energy waste, and improve overall system energy efficiency. This approach ensures equipment operates at optimal efficiency while adjusting energy allocation plans to adapt to changing needs and conditions.

[0151] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A photovoltaic inverter for a photovoltaic microgrid on an isolated island, characterized by: The photovoltaic inverter comprises: The energy status classification module extracts multiple parameters such as temperature and light intensity based on photovoltaic grid power and weather data, classifies the data, and generates a status classification index; The energy conversion prediction module uses the state classification index to analyze the conversion frequency between each state, integrates future weather forecast data, and predicts energy state changes through probability calculation to obtain energy conversion prediction results; The power regulation module adjusts the output parameters of the inverter item by item based on the energy conversion prediction results, synchronously adjusts the power level and frequency matching prediction status, and generates adjustment parameter settings; The optimal path analysis module uses the adjustment parameter settings to compare the energy efficiency of AC and DC transmission, selects the transmission mode and path according to the current load of the network, and obtains the path selection result; The output adjustment control module adjusts the output settings of the inverter based on the path selection result, updates the power output and frequency, optimizes energy distribution, and generates an output control configuration; The energy allocation module reconfigures the charging and discharging timing to balance the load demand, optimize the energy utilization efficiency, and generate an energy allocation plan through the output control configuration; The energy conversion prediction module includes: The state trend analysis submodule analyzes the energy state conversion trend in an isolated island environment through a hidden Markov model based on the state classification index, evaluates the impact of seasonality and climate change on state conversion, and obtains the energy state dynamic analysis results; The climate impact integration submodule combines the energy state dynamic analysis results with future climate condition data to generate a climate prediction model; The conversion probability assessment submodule uses the climate prediction model and the hidden Markov model to analyze the conversion probability of energy states under given environmental conditions to obtain energy conversion prediction results.

2. The photovoltaic inverter for a photovoltaic microgrid applied to an isolated island according to claim 1, characterized in that: The state classification index includes the standard threshold for data segmentation, the energy output range corresponding to each level of state, and the classification weights of environmental parameter influencing factors. The energy conversion prediction results include the probability distribution of each state, the expected state conversion time point and the state duration period. The adjustment parameter settings include the maximum and minimum power output limits of the inverter, the frequency adjustment range, and the preset power adjustment step value. The output control configuration includes the adjusted power output curve, the adjustment of the set frequency, and the inverter response time setting. The energy allocation plan includes the priority ranking of charging and discharging, the adjusted schedule, and the target energy storage amount for each time period.

3. The photovoltaic inverter for a photovoltaic microgrid applied to an isolated island according to claim 1, characterized in that: The energy status classification module includes: The environmental monitoring submodule collects photovoltaic power grid electricity and weather data, including island temperature and light intensity, based on the isolated island's climatic conditions, and performs real-time monitoring to obtain an island environmental dataset. The output stratification analysis submodule uses the island environment dataset to classify energy output into three levels: excellent, good, and poor according to light and temperature data, and generates stratified energy data; The index construction submodule integrates information from the hierarchical energy data, integrates multiple key factors such as light intensity and temperature fluctuation, encodes them according to the energy output level, and generates a state classification index.

4. The photovoltaic inverter for a photovoltaic microgrid applied to an isolated island according to claim 1, characterized in that: The hidden Markov model follows the formula: in, Indicates that the status during the initial observation period is The probability of The initial observation period state is The number of occurrences of is the weight coefficient of the influence of light intensity on the state, Status The corresponding light intensity adjustment factor, is the total number of observations, It is the sum of the light intensity adjustment factors of all states; and Analyze the conversion trend of energy status in isolated island environments; in, is the state transition probability matrix, indicating the transition from state Transfer to state The probability of is the time index, is the end point of the total observation period, For in time Status is The indicator function, For in time Status is The indicator function, is the weight coefficient of temperature influence, For in time state The temperature influence factor, is the weight coefficient of temperature influence, For in time state The temperature influence factor, Used to adjust the weight of temperature effect in the state transition probability calculation.

5. The photovoltaic inverter for a photovoltaic microgrid applied to an isolated island according to claim 1, characterized in that: The power regulation module includes: The energy demand matching submodule analyzes the current energy demand of the isolated island based on the energy conversion prediction result, adjusts the power output of the inverter, and generates power matching parameters; The frequency corresponding submodule uses the power matching parameters to adjust the inverter frequency to match the updated power setting to obtain the frequency adaptation setting; The output adjustment submodule utilizes the frequency adaptation setting to optimize the overall output configuration of the inverter, including output power regulation and energy distribution, and generates adjustment parameter settings.

6. The photovoltaic inverter for a photovoltaic microgrid applied to an isolated island according to claim 1, characterized in that: The optimal path analysis module includes: The transmission efficiency analysis submodule compares the energy efficiency of AC and DC transmission modes in the photovoltaic power grid of the isolated island based on the adjustment parameter settings, analyzes their respective energy losses and transmission efficiencies, and generates photovoltaic power grid transmission efficiency data; The network load evaluation submodule uses the photovoltaic power grid transmission efficiency data to analyze the current power grid load situation and load fluctuation characteristics, compares the performance of various transmission modes, and generates load status analysis results; The energy transmission optimization submodule selects an energy transmission path that matches the needs of the isolated island power grid based on the load condition analysis results through a path selection algorithm, including comparing the energy efficiency and stability of each path and generating a path selection result.

7. The photovoltaic inverter for a photovoltaic microgrid applied to an isolated island according to claim 6, characterized in that: The path selection algorithm is based on the formula: Calculate the best path selection index ; in, 、 、 、 、 and are weight coefficients, which are used to balance the impact of energy efficiency, power stability, weather impact, equipment failure rate, seasonal variation factors, and renewable energy output volatility. is the energy efficiency of the path, is the power stability of the path, is the weather influencing factor, is the equipment failure rate, is the seasonal variation factor, The output volatility of renewable energy.

8. The photovoltaic inverter for a photovoltaic microgrid applied to an isolated island according to claim 1, characterized in that: The output adjustment control module includes: The power adjustment submodule reconfigures the power output parameters of the inverter based on the path selection result, including changing the power level to match the needs of the island power grid and generating an updated power configuration; The frequency calibration submodule uses the updated power configuration to iteratively adjust the frequency of the inverter, including calibrating the output frequency to match the updated power parameters to obtain a frequency synchronization configuration; The distribution optimization submodule adjusts and optimizes the energy distribution of the inverter by using the frequency synchronization configuration and referring to the current energy output and demand conditions, and generates an output control configuration.

9. The photovoltaic inverter for a photovoltaic microgrid applied to an isolated island according to claim 1, characterized in that: The energy allocation module includes: The timing optimization submodule rearranges the charging and discharging schedule based on the output control configuration to generate an optimized charging and discharging schedule; The energy balancing submodule uses the optimized charging and discharging schedule to analyze the grid load, adjust the charging and discharging priorities according to the energy supply situation, and generate a load management plan; The equipment efficiency optimization submodule optimizes the operation and scheduling mode of the energy storage equipment through the load management plan and generates an energy distribution plan.

Images