Photovoltaic power station operation and maintenance strategy optimization method

Abstract

The invention relates to the technical field of photovoltaic power station operation and maintenance, in particular to a photovoltaic power station operation and maintenance strategy optimization method, which comprises the following steps of: acquiring sunlight intensity, cloud layer coverage and air temperature data in the next week, then performing data cleaning and formatting processing, and executing data preprocessing operation including denoising, data interpolation and time marking; and generating weather prediction data. According to the invention, after sunlight intensity, cloud layer coverage and air temperature data in the next week are collected and processed, the weather prediction data enables the photovoltaic power station to accurately estimate the availability of solar energy, and the power production plan is optimized. The power station can predict the power demand in a short period by using historical power use data analysis, and adjusts the power supply to match the consumption demand, thereby maintaining the balance of supply and demand. The weather and power demand prediction are integrated, the charging and discharging strategy of the energy storage equipment is dynamically adjusted, and the energy management efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power station operation and maintenance, and in particular to a photovoltaic power station operation and maintenance strategy optimization method. Background Art

[0002] The field of photovoltaic power plant operation and maintenance technology involves the maintenance and management of solar photovoltaic power plants to ensure their efficient and reliable operation. This includes monitoring the daily operation of the power plant, detecting and preventing equipment failures, and making necessary repairs and replacements.

[0003] However, the existing photovoltaic power station operation and maintenance technology focuses more on the daily monitoring and maintenance of equipment. The lack of weather and demand forecasts results in the inability of the power station to effectively adjust the power output, which may cause energy shortages during peak demand periods or energy surpluses during low demand periods. In addition, the management of energy storage equipment mostly adopts static strategies and lacks dynamic adjustment capabilities based on real-time data, resulting in insufficient resource utilization. This limits the economic benefits and sustainable operation capabilities of the power station. Therefore, improvements are needed. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a photovoltaic power station operation and maintenance strategy optimization method.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: The photovoltaic power station operation and maintenance strategy optimization method comprises:

[0006] Collect data on sunlight intensity, cloud cover, and temperature for the next week, then clean and format the data, and perform data preprocessing operations, including noise removal, data interpolation, and time tagging, to generate weather forecast data;

[0007] Based on the historical power usage data, data cleaning is performed to remove outliers and fill in missing data points, and then the data is normalized. Then, the power demand of the power station in the short term in the future is predicted based on the data to generate power demand forecast results;

[0008] Receive the weather forecast data and the power demand forecast result, perform energy supply and demand matching analysis based on the weather forecast data and the power demand forecast result, calculate the gap between power supply and demand in each period, dynamically adjust the charging and discharging timing and amount of the energy storage device, and obtain an energy storage scheduling plan;

[0009] Based on the energy storage scheduling scheme, the photovoltaic output and the power released by the energy storage are time-matched, the power output is adjusted to match the consumption curve, and an optimized power scheduling strategy is obtained;

[0010] According to the optimized power dispatching strategy, the operation data of the photovoltaic power station, including power output and energy consumption, is recorded in real time. Whether the operation meets the predetermined efficiency target is verified through data analysis, and the operation parameters are adjusted to maintain efficiency.

[0011] Preferably, the steps of obtaining the weather forecast data are:

[0012] Collect meteorological data within a week, including sunlight intensity, cloud cover and temperature information, perform preliminary cleaning on the collected data, remove damaged or inconsistent records, interpolate and supplement missing data, and obtain cleaned meteorological data;

[0013] Based on the cleaned meteorological data, data preprocessing is performed, a time tag is applied to mark the acquisition time of each data point, and a denoising process is performed to remove random fluctuations and abnormal points in the data to generate preprocessed meteorological data;

[0014] According to the pre-processed meteorological data, the formula is adopted:

[0015]

[0016] Among them, S i Indicates sunlight intensity, W i represents the weight of cloud cover, T i The inverse weight of the temperature is expressed, and the weighted average is calculated to obtain the meteorological condition index for each period and generate weather forecast data.

[0017] Preferably, the steps for obtaining the power demand forecast result are:

[0018] Perform data cleaning on historical power usage data, remove outliers, fill in missing data points, check the integrity and accuracy of the data, and obtain cleaned power data;

[0019] Performing data normalization processing on the cleaned power data, standardizing the data range to eliminate dimensional influence and proportional deviation, and generating normalized power data;

[0020] Using the normalized power data, the formula is adopted:

[0021]

[0022] Calculate the logarithmic growth of electricity demand, obtain the electricity demand forecast value for each period, and generate the electricity demand forecast result; where α represents the adjustment coefficient, reflecting the impact of seasonality and specific day types on electricity demand, β represents the sensitivity of the electricity demand growth rate, and D i Represents electricity consumption data for a single period of time.

[0023] Preferably, the steps of energy supply and demand matching analysis are:

[0024] Receiving the weather forecast data and the power demand forecast result, and performing supply and demand comparison based on the two data, specifically matching the predicted hourly power demand with the predicted power generation to obtain supply and demand matching data;

[0025] By using the supply and demand matching data, the gap between power supply and demand in each time period is calculated, and the difference between the supply and demand data per hour is calculated to determine which time periods require additional power input or can reduce power generation, thereby generating time period power difference data;

[0026] Using the power difference data for the period, the formula is adopted:

[0027]

[0028] Calculate the adjusted power supply and demand balance to obtain the energy supply and demand matching analysis results, where γ adjusts the weight of power demand to give priority to power supply during critical periods, P demand,i represents the power demand in the i-th period, P supply,i represents the power supply in the i-th period; N represents the total number of periods analyzed.

[0029] Preferably, the steps for obtaining the energy storage scheduling plan are:

[0030] Based on the energy supply and demand matching analysis results, determine which periods of time the power demand is higher than the supply and which periods the supply exceeds the demand, make preliminary plans for the charging and discharging timing of the energy storage equipment, determine the priority and period of charging and discharging, and obtain a preliminary energy storage scheduling plan;

[0031] Through the preliminary energy storage scheduling plan, the charging and discharging timing and amount are specifically adjusted, and the operation is performed to refine the charging and discharging amount of each time period, refer to the changes in electricity prices and grid loads, optimize the operating efficiency of energy storage equipment, and generate a detailed energy storage scheduling strategy;

[0032] Using the detailed energy storage scheduling strategy, the formula is adopted:

[0033]

[0034] Calculate the energy storage equipment scheduling plan and generate the energy storage scheduling plan; where N represents the total number of scheduling periods, δ i and ∈ i Refer to the adjustment coefficient of peak and valley power demand to ensure sufficient energy storage output during peak power demand period and actively store energy during valley power demand period. i and D i Indicates the amount of charging and discharging operations during a period.

[0035] Preferably, the steps for obtaining the optimized power dispatching strategy are:

[0036] Based on the results of the energy storage scheduling scheme, the power supply of photovoltaic output and energy storage release is analyzed, and the power output of each period is preliminarily matched, so that the power supply corresponds to the consumer demand curve, and a preliminary power matching result is obtained;

[0037] According to the preliminary power matching result, the power output is adjusted, and the operation is performed to match the power output of photovoltaic power and the power released by energy storage during the period, calculate the power surplus or shortage during the period, optimize the power output matching consumption curve, and generate a detailed power dispatch plan;

[0038] Using the detailed power dispatch plan, the formula is adopted:

[0039]

[0040] Calculate and optimize the final power dispatch and generate the optimized power dispatch strategy, where φ and ψ are the power supply and demand adjustment coefficients, respectively, and S supply,i and D demand,i They represent the power supply and demand in the i-th period respectively, and N represents the total number of scheduling periods.

[0041] Preferably, the steps of data analysis verification are as follows:

[0042] Based on the optimized power dispatching strategy, the operation data of the photovoltaic power station is collected in real time, including the power output and energy consumption in each period, to obtain a real-time operation data set;

[0043] By using the real-time running data set, performing data analysis, calculating the ratio of energy output to consumption in each time period, identifying time periods that do not meet the predetermined efficiency target, and generating efficiency difference analysis results;

[0044] Using the efficiency difference analysis results, the formula is adopted:

[0045]

[0046] The operation efficiency verification results are obtained, where P actual,i represents the power output in the i-th period, P target,i represents the target power output in the i-th period, ω i Represents the weight factor, reflecting the importance of each time period or the electricity price.

[0047] Compared with the prior art, the advantages and positive effects of the present invention are:

[0048] In the present invention, by collecting and processing the sunlight intensity, cloud cover and temperature data for the next week, the weather forecast data enables the photovoltaic power station to accurately estimate the availability of solar energy and optimize the power production plan. By analyzing the historical power usage data, the power station can predict the power demand in the short term and adjust the power supply to match the consumption demand, thereby maintaining the balance between supply and demand. Integrate weather and power demand forecasts, dynamically adjust the charging and discharging strategies of energy storage equipment, and improve energy management efficiency. Through real-time monitoring and data analysis, the operating efficiency and parameter adjustment of the power station are more accurate, which improves the energy utilization rate and the adaptability of the power station to environmental changes, while ensuring the continuity and reliability of the power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with 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.

[0051] See also Figure 1 The present invention provides a technical solution: a photovoltaic power station operation and maintenance strategy optimization method includes:

[0052] Collect data on sunlight intensity, cloud cover, and temperature for the next week, then clean and format the data, and perform data preprocessing operations, including noise removal, data interpolation, and time tagging, to generate weather forecast data;

[0053] Based on the historical power usage data, data cleaning is performed to remove outliers and fill in missing data points, and then the data is normalized. Then, the power demand of the power station in the short term in the future is predicted based on the data to generate power demand forecast results;

[0054] Receive weather forecast data and power demand forecast results, conduct energy supply and demand matching analysis based on the two, calculate the gap between power supply and demand in each period, dynamically adjust the charging and discharging timing and quantity of energy storage equipment, and obtain an energy storage scheduling plan;

[0055] Based on the energy storage dispatching scheme, the photovoltaic output and the power released by the energy storage are matched in time periods, and the power output is adjusted to match the consumption curve to obtain the optimized power dispatching strategy;

[0056] According to the optimized power dispatch strategy, the operating data of the photovoltaic power station, including power output and energy consumption, is recorded in real time. Through data analysis, it is verified whether the operation meets the predetermined efficiency target, and the operating parameters are adjusted to maintain efficiency.

[0057] The steps to obtain weather forecast data are:

[0058] Collect meteorological data within a week, including sunlight intensity, cloud cover and temperature information, perform preliminary cleaning on the collected data, remove damaged or inconsistent records, interpolate and supplement missing data, and obtain cleaned meteorological data;

[0059] Based on the cleaned meteorological data, data preprocessing is performed, a time tag is applied to mark the collection time of each data point, and denoising is performed to remove random fluctuations and abnormal points in the data to generate preprocessed meteorological data;

[0060] Through the preprocessed meteorological data, the formula is used:

[0061]

[0062] Among them, S i Indicates sunlight intensity, W i represents the weight of cloud cover, T i represents the inverse weight of temperature,

[0063] Calculate the weighted average to obtain the meteorological condition index for each time period and generate weather forecast data.

[0064] Specifically, data on sunlight intensity, cloud cover and temperature for the next week are collected. By using high-precision meteorological observation equipment, the photometric count values ​​of direct sunlight intensity are regularly recorded, and the percentage of cloud cover is recorded through a cloud observation camera. At the same time, temperature sensors are used to continuously monitor temperature changes. After data collection, they are immediately uploaded to the central server. The server uses advanced data processing technology to identify abnormal data. For data that obviously deviates from the normal range, such as data with abnormally high or low sunlight intensity, the data from the previous and next time points are used for reasonable interpolation to fill in these missing data. Through these sophisticated processing steps, the quality and integrity of the collected meteorological data are ensured, thereby obtaining cleaned meteorological data.

[0065] Based on the cleaned meteorological data, data preprocessing is performed. For each data point, a timestamp is added according to the actual collection time to ensure the time accuracy of each data. Further, a high-pass filter is used to remove random noise in the data, especially in the cloud cover data to remove spikes or null values ​​caused by sensor misreading. In addition, a linear interpolation method is used to handle short-term fluctuations in temperature data. Through these detailed preprocessing operations, the availability and accuracy of the data are improved, and preprocessed meteorological data are generated.

[0066] formula The calculation process:

[0067] In the formula, S iRepresents the measured value of sunlight intensity during a specific period of time, W i is a cloud cover adjustment factor that indicates how clouds affect sunlight intensity, ranging from 0 (complete coverage) to 1 (no clouds), T i It is the inverse weight of the effect of temperature on power generation efficiency. Generally, power generation efficiency decreases at higher temperatures.

[0068] The sunlight intensity in a certain period is 500W / m2, the cloud cover factor is 0.5, and the temperature is 35℃, and the corresponding inverse weight is 1.1. Substituting the values ​​into the formula, the calculation process is as follows:

[0069]

[0070] This result shows that after taking into account the effects of cloud cover and high temperatures, the effective sunlight intensity index during this period is 227W / m2, which will be used in further energy management calculations.

[0071] The steps to obtain the power demand forecast results are as follows:

[0072] Perform data cleaning on historical power usage data, remove outliers, fill in missing data points, check the integrity and accuracy of the data, and obtain cleaned power data;

[0073] The cleaned power data is normalized to standardize the data range to eliminate the dimension effect and proportion deviation, and generate normalized power data;

[0074] Using normalized power data, the formula is:

[0075]

[0076] Calculate the logarithmic growth of electricity demand, obtain the electricity demand forecast value for each period, and generate the electricity demand forecast result; where α represents the adjustment coefficient, reflecting the impact of seasonality and specific day types on electricity demand, β represents the sensitivity of the electricity demand growth rate, and D i Represents electricity consumption data for a single period of time.

[0077] Specifically, based on the historical electricity usage data, the data set is first sorted out to remove abnormal data caused by equipment failure or external factors, such as sudden power peaks or readings of zero. These abnormal points may affect the accuracy of the prediction, and the incomplete data points are completed using linear interpolation to ensure the continuity and integrity of the data. This completion method assumes that electricity consumption changes smoothly in a short period of time and can more accurately reflect the actual situation, so that the cleaned electricity data has higher reliability and reference value, laying the foundation for the next step of data processing and analysis. Through this series of data preprocessing steps, cleaned electricity data that can be directly used for further analysis is obtained.

[0078] After the cleaned power data is processed, the next step is to normalize the data. This process involves converting the power usage of each data point to a unified standard scale. This process is done by calculating the mean and standard deviation of the original data and then applying Each data point is converted using the formula, where x represents the original data point, μ and σ represent the mean and standard deviation of the data, respectively. This method can effectively eliminate the dimensional influence in the original data and the proportional deviation of electricity use in different time periods. The normalized data is convenient for various statistical tests and the application of machine learning algorithms, ensuring the consistency and comparability of subsequent analysis, and generating normalized power data for further analysis.

[0079] formula The calculation process:

[0080] In the electricity demand forecasting model, α and β are adjustment coefficients that are adjusted according to seasonal changes and specific day types to reflect the impact of these factors on electricity demand. For example, electricity demand may increase significantly in winter or summer due to the use of heating and air conditioning.

[0081] Setting α = 0.5 reflects a lower seasonal effect, and β = 0.03 indicates a higher sensitivity to the demand growth rate. i Represents the power consumption data in a single period. In a certain period, the power consumption is 100kWh, so the calculation process is:

[0082] DR=0.5×log(∑e (0.03×100) )=0.5×log(e 3 )=0.5×3=1.5

[0083] This result shows that after considering the effects of seasonality and specific day types, the predicted electricity demand adjustment factor is 1.5, that is, the predicted demand should be multiplied by this factor to estimate the actual electricity demand.

[0084] The steps of energy supply and demand matching analysis are:

[0085] Receive weather forecast data and power demand forecast results, and compare supply and demand based on these two data. The specific operation is to match the predicted hourly power demand with the predicted power generation to obtain supply and demand matching data;

[0086] By matching supply and demand data, the gap between electricity supply and demand in each period is calculated, and the difference between supply and demand data for each hour is calculated to determine which periods require additional power input or can reduce power generation, and generate period power difference data;

[0087] Using the power difference data of time periods, the formula is adopted:

[0088]

[0089] Calculate the adjusted power supply and demand balance to obtain the energy supply and demand matching analysis results, where γ adjusts the weight of power demand to give priority to power supply during critical periods, P demand,i represents the power demand in the i-th period, P supply,i represents the power supply in the i-th period; N represents the total number of periods analyzed.

[0090] Specifically, the weather forecast data received includes the sunlight intensity, cloud coverage and temperature every hour within a week, while the electricity demand forecast results provide the hourly electricity demand during the same period. Supply and demand are compared through these two data. First, the two sets of data are aligned in time to ensure that each data point has corresponding weather and demand data. Next, the predicted power generation is calculated for the hourly data. The power generation is calculated based on sunlight intensity and temperature. Considering the impact of cloud coverage on sunlight intensity, a simplified model is used to estimate the actual available sunlight intensity. This calculation process includes applying cloud coverage as an attenuation coefficient to sunlight intensity. At the same time, the temperature affects the efficiency of photovoltaic panels, and this factor is also taken into account to simulate the actual power generation per hour. The simulated power generation is compared with the electricity demand, so that it can be determined which periods may have excess or insufficient electricity. In this way, the comparison process ensures real-time updating and accuracy of the data, thereby obtaining supply and demand matching data.

[0091] What is obtained from the supply and demand matching data is the difference between electricity supply and demand in each time period. These data are obtained by calculating the difference between the supply and demand data every hour. The specific execution process includes numerical subtraction of the hourly power generation and power demand to determine which time periods require additional power input or can reduce power generation. For each hour of difference data, positive and negative values ​​are analyzed. Positive values ​​indicate that supply exceeds demand, that is, you can consider reducing power generation or increasing power storage during this period. Negative values ​​indicate that demand exceeds supply, and it is necessary to increase power generation or release power from energy storage devices. Through this analysis, we can have a more detailed understanding of the operating conditions of the power station in different time periods, thereby providing decision support for power dispatching. The hourly power supply and demand difference calculated by this method provides a basis for our subsequent optimization of dispatching strategies, thereby generating time period power difference data.

[0092] formula: The calculation process:

[0093] The formula is used to calculate the adjusted power supply and demand balance, where N represents the total number of analyzed time periods. If there are 24 hours in a day, then N = 24; γ is a weight coefficient used to adjust the power supply to be provided more during the period with higher power demand to ensure the power supply during the critical period; P demand,i and P supply,i Respectively represent the power demand and supply in the i-th period. At the morning peak at 8:00AM, the power demand is 800MW, while the supply is 760MW. At the same time, γ=1.1 to increase the weight of the demand period. The calculation process is as follows:

[0094] (1.1×800-760+…+calculation of other 23 periods)

[0095] This process will traverse each time period, and the final SDS value will reflect the overall supply and demand balance. A higher SDS value may indicate the need to increase power supply or optimize the scheduling strategy to ensure the stable operation of the power system.

[0096] The steps to obtain the energy storage dispatch plan are:

[0097] Based on the results of the energy supply and demand matching analysis, determine which periods of time the power demand is higher than the supply and which periods the supply exceeds the demand, make preliminary plans for the charging and discharging sequence of the energy storage equipment, determine the priority and period of charging and discharging, and obtain a preliminary energy storage scheduling plan;

[0098] Through the preliminary energy storage scheduling plan, the charging and discharging timing and amount are specifically adjusted, and the execution operation is to refine the charging and discharging amount in each period, refer to the changes in electricity prices and grid load, optimize the operating efficiency of energy storage equipment, and generate a detailed energy storage scheduling strategy;

[0099] Using the detailed energy storage dispatch strategy, the formula is adopted:

[0100]

[0101] Calculate the energy storage equipment scheduling plan and generate the energy storage scheduling plan; where N represents the total number of scheduling periods, δ i and ∈ i Refer to the adjustment coefficient of peak and valley power demand to ensure sufficient energy storage output during peak power demand period and actively store energy during valley power demand period. i and D i Indicates the amount of charging and discharging operations during a period.

[0102] Specifically, based on the results of the energy supply and demand matching analysis, a detailed analysis is conducted on the electricity demand and supply situation in each time period, and the time periods when demand is higher than supply and the time periods when supply exceeds demand are identified, so as to make a preliminary plan for the charging and discharging timing of the energy storage equipment. This plan includes determining which time periods should give priority to charging to reserve energy and which time periods should give priority to discharging to meet electricity demand. This step ensures the effective use of energy and the economic benefits of power station operation. In this way, the charging and discharging activities of the energy storage equipment can more effectively respond to the real-time demand and price changes of the electricity market, thereby obtaining a preliminary energy storage scheduling plan.

[0103] Through the preliminary energy storage scheduling plan, the charging and discharging timing and quantity of the energy storage equipment are further refined, taking into account the real-time changes in electricity prices and grid loads. Specific operations include calculating the impact of electricity prices in different time periods on the energy storage strategy and adjusting the priority and intensity of charging and discharging. This process dynamically analyzes the load of the grid and the market electricity price, optimizes the operation strategy of the energy storage equipment, ensures that energy is released when electricity prices are peak and stored when electricity prices are low, thereby maximizing economic benefits and improving energy utilization efficiency, and finally generates a detailed energy storage scheduling strategy.

[0104] formula: The calculation process:

[0105] The formula is used to calculate the energy storage dispatch plan, where N represents the total number of dispatch periods. There are 24 periods in a day, so N = 24; δ i and ∈ i They are the adjustment coefficients for the charging and discharging periods, respectively, and are adjusted according to the peak and valley demand of the electricity market. i Set to 1.2 to promote charging during daytime high electricity price periods∈ i Set to 1.5 to facilitate discharge; C i and D iThey represent the charging and discharging amounts in the ith period respectively. In the 8th period, the charging amount is 100MWh and the discharging amount is 50MWh. The SCS contribution of this period is calculated to be 1.2×100-1.5×50=45MWh. The SCS value of the whole day is obtained by accumulating all periods. This value will be used to evaluate the overall effect of the energy storage strategy within a day. The result shows that by adjusting the charging and discharging strategy to adapt to market price changes, energy storage equipment can effectively balance supply and demand and improve energy utilization and economic benefits.

[0106] The steps to obtain the optimized power dispatch strategy are:

[0107] Based on the results of the energy storage dispatch plan, the power supply of photovoltaic output and energy storage release is analyzed, and the power output of each period is preliminarily matched, so that the power supply corresponds to the consumer demand curve, and the preliminary power matching results are obtained;

[0108] Adjust the power output through the preliminary power matching results, perform operations to match the power output of photovoltaic power and the power released by energy storage, calculate the power surplus or shortage in the time period, optimize the power output matching consumption curve, and generate a detailed power dispatch plan;

[0109] Using the detailed power dispatch plan, the formula is:

[0110]

[0111] Calculate and optimize the final power dispatch and generate the optimized power dispatch strategy, where φ and ψ are the power supply and demand adjustment coefficients, respectively, and S supply,i and D demand,i They represent the power supply and demand in the i-th period respectively, and N represents the total number of scheduling periods.

[0112] Specifically, based on the results of the energy storage scheduling plan, an analysis of the power supply of photovoltaic output and energy storage release is conducted to ensure that the power output in each time period matches the consumer demand curve. The execution operation includes checking the hourly power data to evaluate whether it meets or exceeds the predetermined demand. This includes comparing the real-time power output data with the historical data for the same period, analyzing the differences, and ensuring the real-time adjustment and optimization of the scheduling plan. The adjustment measures include increasing the charging power of the energy storage equipment or adjusting the discharge time to match the consumption peak or trough, thereby obtaining a preliminary power matching result.

[0113] Based on the preliminary power matching results, detailed power dispatch of photovoltaic output and energy storage release is carried out, which involves analyzing the specific power supply and demand status of each time period, and performing operations to calculate in detail the power surplus or shortage in each time period. This includes using a real-time monitoring system to track power output and demand fluctuations, and adjusting the operation strategy of energy storage equipment according to monitoring data, such as increasing energy storage charging when there is an oversupply of electricity or increasing discharge when there is a shortage, optimizing power output to ensure consistency with actual consumer demand, and thus generating a detailed power dispatch plan.

[0114] formula: The calculation process:

[0115] In this formula, φ and ψ are the adjustment coefficients of power supply and demand, which are used to adjust the weights of power supply and demand respectively. The actual power supply S obtained by monitoring supply,i and demand D demand,i , these data are automatically updated every hour by the energy management system. In the i-th period of the day, the actual power supply is 200MW and the demand is 180MW. The adjustment coefficients φ=1.05 and ψ=0.95 reflect the priority adjustment strategy for supply and demand balance. The power dispatch strategy for this period is calculated as follows:

[0116] EDS=1.05×200-0.95×180=210-171=39MW

[0117] This shows that there was an excess supply of 39MW of electricity during that period. Based on this result, the power dispatching system will adjust the discharge plan of the energy storage equipment in subsequent periods to reduce the excess and ensure the efficient operation of the power system.

[0118] The specific steps verified by data analysis are:

[0119] Based on the optimized power dispatch strategy, the operation data of the photovoltaic power station is collected in real time, including the power output and energy consumption in each period, to obtain a real-time operation data set;

[0120] By running the data set in real time, performing data analysis, calculating the energy output and consumption ratio in each period, identifying the period that does not meet the predetermined efficiency target, and generating efficiency difference analysis results;

[0121] Using the results of the efficiency difference analysis, the formula is adopted:

[0122]

[0123] Among them, P actual,i represents the power output in the i-th period, P target,i represents the target power output in the i-th period, ω iIt represents the weight factor, reflects the importance of each time period or the electricity price, calculates the weighted average efficiency, and verifies whether the operation of the power station reaches or exceeds the predetermined efficiency target.

[0124] Specifically, according to the optimized power dispatch strategy, real-time data collection involves obtaining key performance indicators from the monitoring system of the photovoltaic power station, including power output and energy consumption in each period. These data are automatically recorded by sensors installed at each key node and uploaded to the central monitoring system. Through the data integrity verification process, it is ensured that all collected data are accurate and real-time. This step is the basis for building an accurate analysis model. The obtained real-time operation data set will be used for the next step of efficiency analysis;

[0125] Using real-time operation data sets, the actual power output and energy consumption of the power station in each period are analyzed in detail. By calculating the data of each period, it is analyzed whether the power output reaches the predetermined power target. This analysis is performed by comparing the difference between the actual power and the target power, with special attention paid to those periods where the power is lower than expected, in order to identify possible operating deficiencies or equipment efficiency problems, thereby providing a basis for adjusting the operation strategy and generating efficiency difference analysis results;

[0126] formula The calculation process:

[0127] In a specific evaluation period (such as a day), the photovoltaic power station has data for three periods, and the specific values ​​are: the actual power output P in period 1 actual,1 =150kW, target power output P target,1 =

[0128] 160kW, weight factor ω1 = 1.0; actual power P in period 2 actual,2 =180kW, target power P target,2 =175kW, weight factor ω2 = 1.5; actual power P in period 3 actual,3 =160kW, target power P target,3 =150kW, weight factor ω3 = 0.5;

[0129] Substitute the data into the formula for calculation:

[0130]

[0131] The results show that during the entire evaluation period, the actual operating efficiency of the power plant was slightly higher than the target setting value, indicating that the operating efficiency of the power plant basically reached and slightly exceeded the predetermined efficiency target.

Claims

1. A photovoltaic power station operation and maintenance strategy optimization method, characterized in that: The following steps are involved: Collect data on sunlight intensity, cloud cover, and temperature for the next week, then clean and format the data, and perform data preprocessing operations, including noise removal, data interpolation, and time tagging, to generate weather forecast data; Based on the historical power usage data, data cleaning is performed to remove outliers and fill in missing data points, and then the data is normalized. Then, the power demand of the power station in the short term in the future is predicted based on the data to generate power demand forecast results; Receive the weather forecast data and the power demand forecast result, perform energy supply and demand matching analysis based on the weather forecast data and the power demand forecast result, calculate the gap between power supply and demand in each period, dynamically adjust the charging and discharging timing and amount of the energy storage device, and obtain an energy storage scheduling plan; Based on the energy storage scheduling scheme, the photovoltaic output and the power released by the energy storage are time-matched, the power output is adjusted to match the consumption curve, and an optimized power scheduling strategy is obtained; According to the optimized power dispatching strategy, the operation data of the photovoltaic power station, including power output and energy consumption, is recorded in real time. Whether the operation meets the predetermined efficiency target is verified through data analysis, and the operation parameters are adjusted to maintain efficiency.

2. The photovoltaic power station operation and maintenance strategy optimization method according to claim 1, characterized in that: The steps for obtaining the weather forecast data are as follows: Collect meteorological data within a week, including sunlight intensity, cloud cover and temperature information, perform preliminary cleaning on the collected data, remove damaged or inconsistent records, interpolate and supplement missing data, and obtain cleaned meteorological data; Based on the cleaned meteorological data, data preprocessing is performed, a time tag is applied to mark the acquisition time of each data point, and a denoising process is performed to remove random fluctuations and abnormal points in the data to generate preprocessed meteorological data; According to the pre-processed meteorological data, the formula is adopted: Among them, S i Indicates sunlight intensity, W i represents the weight of cloud cover, T i The inverse weight of the temperature is expressed, and the weighted average is calculated to obtain the meteorological condition index for each period and generate weather forecast data.

3. The photovoltaic power station operation and maintenance strategy optimization method according to claim 1, characterized in that: The steps for obtaining the power demand forecast result are: Perform data cleaning on historical power usage data, remove outliers, fill in missing data points, check the integrity and accuracy of the data, and obtain cleaned power data; Performing data normalization processing on the cleaned power data, standardizing the data range to eliminate dimensional influence and proportional deviation, and generating normalized power data; Using the normalized power data, the formula is adopted: Calculate the logarithmic growth of electricity demand, obtain the electricity demand forecast value for each period, and generate the electricity demand forecast result; where α represents the adjustment coefficient, reflecting the impact of seasonality and specific day types on electricity demand, β represents the sensitivity of the electricity demand growth rate, and D i Represents electricity consumption data for a single period of time.

4. The photovoltaic power station operation and maintenance strategy optimization method according to claim 1, characterized in that: The steps of energy supply and demand matching analysis are: Receiving the weather forecast data and the power demand forecast result, and performing supply and demand comparison based on the two data, specifically matching the predicted hourly power demand with the predicted power generation to obtain supply and demand matching data; By using the supply and demand matching data, the gap between power supply and demand in each time period is calculated, and the difference between the supply and demand data per hour is calculated to determine which time periods require additional power input or can reduce power generation, thereby generating time period power difference data; Using the power difference data for the period, the formula is adopted: Calculate the adjusted power supply and demand balance to obtain the energy supply and demand matching analysis results, where γ adjusts the weight of power demand to give priority to power supply during critical periods, P demand,i represents the power demand in the i-th period, P supply,i represents the power supply in the i-th period; N represents the total number of time periods analyzed.

5. The photovoltaic power station operation and maintenance strategy optimization method according to claim 4, characterized in that: The steps for obtaining the energy storage scheduling plan are: Based on the energy supply and demand matching analysis results, determine which periods of time the power demand is higher than the supply and which periods the supply exceeds the demand, make preliminary plans for the charging and discharging timing of the energy storage equipment, determine the priority and period of charging and discharging, and obtain a preliminary energy storage scheduling plan; Through the preliminary energy storage scheduling plan, the charging and discharging timing and amount are specifically adjusted, and the operation is performed to refine the charging and discharging amount of each time period, refer to the changes in electricity prices and grid loads, optimize the operating efficiency of energy storage equipment, and generate a detailed energy storage scheduling strategy; Using the detailed energy storage scheduling strategy, the formula is adopted: Calculate the energy storage equipment scheduling plan and generate the energy storage scheduling plan; where N represents the total number of scheduling periods, δ i and ∈ i Refer to the adjustment coefficient of peak and valley power demand to ensure sufficient energy storage output during peak power demand period and actively store energy during valley power demand period. i and D i Indicates the amount of charging and discharging operations during a period.

6. The photovoltaic power station operation and maintenance strategy optimization method according to claim 1, characterized in that: The steps for obtaining the optimized power dispatching strategy are: Based on the results of the energy storage scheduling scheme, the power supply of photovoltaic output and energy storage release is analyzed, and the power output of each period is preliminarily matched, so that the power supply corresponds to the consumer demand curve, and a preliminary power matching result is obtained; According to the preliminary power matching result, the power output is adjusted, and the operation is performed to match the power output of photovoltaic power and the power released by energy storage during the period, calculate the power surplus or shortage during the period, optimize the power output matching consumption curve, and generate a detailed power dispatch plan; Using the detailed power dispatch plan, the formula is adopted: Calculate and optimize the final power dispatch and generate the optimized power dispatch strategy, where φ and ψ are the power supply and demand adjustment coefficients, respectively, and S supply,i and D demand,i They represent the power supply and demand in the i-th period respectively, and N represents the total number of scheduling periods.

7. The photovoltaic power station operation and maintenance strategy optimization method according to claim 1, characterized in that: The steps verified by data analysis are as follows: Based on the optimized power dispatching strategy, the operation data of the photovoltaic power station is collected in real time, including the power output and energy consumption in each period, to obtain a real-time operation data set; By using the real-time running data set, performing data analysis, calculating the ratio of energy output to consumption in each time period, identifying time periods that do not meet the predetermined efficiency target, and generating efficiency difference analysis results; Using the efficiency difference analysis results, the formula is adopted: The operation efficiency verification results are obtained, where P actual,i represents the power output in the i-th period, P target,i represents the target power output in the i-th period, ω i Represents the weight factor, reflecting the importance of each time period or the electricity price.

Images