A method for optimizing operation of an energy system based on distributed power supply and load prediction

Abstract

The present application relates to the field of energy system management, and relates to an energy system optimal operation method based on distributed power supply and load prediction, wherein the present application obtains historical power generation data of distributed power supply in a park comprehensive energy system, establishes a distributed power supply power generation prediction model, analyzes the estimated power generation of the distributed power supply, analyzes the estimated power consumption of the park, further obtains the power supply and demand relationship of the park and the energy storage information of the energy storage device in the comprehensive energy system, obtains the actual power generation of the distributed power supply and the actual power consumption of the park, evaluates the prediction accuracy of the park distributed power supply output and load, obtains the coordinated power, coordination frequency, charging and discharging response speed of the energy storage device in the park comprehensive energy system, and the power outage duration and power outage frequency of the comprehensive energy system, analyzes the comprehensive energy system coordinated operation optimization evaluation index, provides a basis for the optimal operation strategy, and improves the overall performance and operation benefit of the energy system.

Description

Technical Field

[0001] This invention relates to the field of energy system management, and specifically to a method for optimizing the operation of energy systems based on distributed power sources and load forecasting. Background Technology

[0002] With the continuous growth of energy demand and the increasing emphasis on environmental protection and sustainable development, traditional centralized energy systems are facing many challenges, leading to the emergence and expansion of distributed energy systems.

[0003] Distributed power sources offer advantages such as high energy efficiency, low environmental pollution, and flexible installation, providing diversified energy supply options. However, their output is intermittent, fluctuating, and uncertain. Meanwhile, the requirements for reliability, economy, and flexibility in energy supply are constantly increasing. To fully leverage the advantages of distributed power sources and improve the overall performance of energy systems, optimizing the operation of distributed power source energy systems has become a key issue in the energy field, possessing significant practical and strategic value.

[0004] However, existing energy system operation optimization methods still have some limitations and shortcomings in practical applications.

[0005] For example, Chinese Patent CN113393126A discloses a method for alternating parallel and collaborative optimization scheduling of high-energy-consuming industrial parks and the power grid, including the following steps: ① Establishing output models and load forecasting models for conventional generator sets, distributed power sources, energy storage, and other equipment; ② Performing day-ahead forecasts for the park's electricity and heat loads, photovoltaic output, and grid-side wind turbine generator sets; ③ Establishing a decomposed collaborative optimization framework for the park's integrated energy system group and the power grid; ④ Establishing a parallel decomposed collaborative optimization scheduling model for the high-energy-consuming industrial park's integrated energy system and the power grid, considering network security constraints; ⑤ Establishing a solution process based on the parallel alternating direction multiplier method; ⑥ Establishing an interactive solution process on the Matlab platform for the parallel decomposed collaborative optimization scheduling of the high-energy-consuming industrial park's integrated energy system and the power grid. This invention provides technical basis and practical methods for the stable operation of the park's integrated energy system group and the power grid, as well as for the large-scale absorption of wind power.

[0006] The shortcomings of the aforementioned patents are as follows: Firstly, they do not evaluate the accuracy of the distributed power generation output prediction model and the load prediction model, which is not conducive to subsequent targeted correction and optimization of the prediction model. Consequently, the coordinated operation and scheduling of the energy system based on the prediction model may have certain deviations, making it impossible to achieve the optimal operating state.

[0007] Secondly, the lack of evaluation of the effects or benefits of energy system coordination and operation optimization makes it difficult to identify bottlenecks and optimization points in the energy system coordination and operation process, thus failing to provide a basis for further optimization of operation strategies and consequently failing to effectively improve the overall performance and operational efficiency of the energy system. Summary of the Invention

[0008] To address the aforementioned problems, this invention proposes an energy system optimization operation method based on distributed power generation and load forecasting. The specific technical solution is as follows: An energy system optimization operation method based on distributed power generation and load forecasting includes the following steps: Step 1: Prediction of the output day of distributed power generation in the park: Obtain historical power generation data of distributed power generation in the park's integrated energy system, wherein distributed power generation includes solar power generation, wind power generation, and hydropower generation. Analyze the relationship between the power generation of distributed power generation in the park's integrated energy system and the variables related to its power generation, establish a prediction model for the power generation of distributed power generation in the park's integrated energy system, and further obtain the estimated power generation of distributed power generation in the park's integrated energy system within the prediction period.

[0009] Step 2: Predicting Park Load Day-ahead: Obtain the park's electricity consumption for each historical period in the current year and the park's electricity consumption for the prediction period in each historical year, and analyze the park's estimated electricity consumption for the prediction period.

[0010] Step 3: Park Power Supply and Demand Forecast: Based on the estimated power generation of distributed power sources in the park's integrated energy system and the estimated power consumption of the park during the forecast period, obtain the type of power supply and demand relationship in the park during the forecast period and the energy storage information of energy storage devices in the park's integrated energy system, and provide feedback.

[0011] Step 4: Evaluation of the accuracy of distributed power generation and load forecasting: Obtain the actual power generation of distributed power sources in the park's integrated energy system and the actual power consumption of the park during the forecasting period, analyze the forecasting accuracy of the park's distributed power generation and load, and provide feedback.

[0012] Step 5: Energy System Coordination Operation Optimization Assessment: Obtain the coordinated power volume, coordination frequency, and charge / discharge response speed of the energy storage devices in the park's integrated energy system during the assessment period. Also, obtain the power outage duration and number of power outages in the park's integrated energy system during the assessment period. Analyze the park's integrated energy system coordination operation optimization evaluation index and provide feedback.

[0013] Compared with existing technologies, the energy system optimization operation method based on distributed power generation and load forecasting described in this invention has the following advantages: 1. This invention obtains the dataset for training the distributed power generation prediction model by preprocessing the historical power generation data of distributed power sources in the integrated energy system of the park, thereby avoiding interference from outlier data on the prediction model and improving the reliability of the distributed power generation prediction model.

[0014] 2. By predicting the output and load of distributed power sources in the park, this invention can help adjust energy supply strategies and optimize the output of distributed power sources, thereby achieving the safe and stable operation of the park's integrated energy system.

[0015] 3. This invention combines distributed power generation output forecasting and load forecasting in the park to obtain the power supply and demand relationship type of the park and the energy storage information of energy storage devices in the integrated energy system. This is beneficial to optimize the charging and discharging strategies of energy storage devices in the integrated energy system and better realize the coordinated operation of the integrated energy system.

[0016] 4. This invention obtains the actual power generation of distributed power sources and the actual power consumption of the park in the integrated energy system, and compares them with the predicted values ​​to evaluate the prediction accuracy of the output and load of distributed power sources in the park. This is beneficial for subsequent targeted correction and optimization of the prediction model, making the coordinated operation and scheduling of the energy system based on the prediction model more accurate and achieving the best operating state.

[0017] 5. This invention evaluates the coordinated operation optimization of the integrated energy system in the park by acquiring the coordinated power volume, coordination frequency, charging and discharging response speed of the energy storage devices, as well as the power outage duration and number of power outages of the integrated energy system. This helps to identify deficiencies in the coordinated operation of the energy system, provides a basis for further optimizing the operation strategy, and improves the overall performance and operational efficiency of the energy system. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the method flow of the present invention.

[0020] Figure 2 This is a flowchart of the analysis process for step one of the present invention.

[0021] Figure 3 This is a flowchart of step two of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Please see Figure 1 As shown, the present invention provides an energy system optimization operation method based on distributed power generation and load forecasting, which includes the following steps: Step 1, pre-day forecasting of distributed power generation in the park: obtain historical power generation data of distributed power generation in the park's integrated energy system, wherein distributed power generation includes solar power generation, wind power generation and hydropower generation, analyze the relationship between the power generation of distributed power generation in the park's integrated energy system and the variables related to its power generation, establish a forecasting model for the power generation of distributed power generation in the park's integrated energy system, and further obtain the estimated power generation of distributed power generation in the park's integrated energy system within the forecast period.

[0024] As a preferred option, see [reference] Figure 2 As shown, the specific analysis process of step one includes: extracting historical power generation data of solar power generation, wind power generation and hydropower generation in the park's integrated energy system stored in the database.

[0025] Based on the historical power generation data of solar power generation in the park's integrated energy system, a data set of each historical power generation in the park's integrated energy system is obtained. The data set includes the values ​​of various variables related to power generation and solar power generation, including solar intensity, sunshine duration, solar angle, and ambient temperature.

[0026] It's important to note that sunlight intensity directly determines the amount of energy received by solar panels. The stronger the sunlight, the greater the power generation. For example, power generation efficiency is high on sunny days with ample sunlight at midday; while power generation will decrease significantly on cloudy days or in the early morning or late evening when sunlight is weak.

[0027] It should be noted that the duration of sunshine affects the total power generation of solar panels. For example, in summer, when sunshine hours are long, power generation is relatively high; while in winter, when sunshine hours are short, power generation is relatively low.

[0028] It should be noted that the angle of the sun varies depending on the season and time of day. Power generation efficiency is highest when sunlight is perpendicular to the solar panels; the greater the angle deviation, the lower the power generation efficiency.

[0029] It should be noted that solar panels operate at their optimal efficiency within a certain temperature range, typically around 25°C. Temperatures that are too high or too low will reduce the conversion efficiency of the panels, thus affecting power generation.

[0030] Extract the numerical range of single-cycle power generation of solar power in the park's integrated energy system stored in the database, as well as the numerical range of light intensity, sunshine duration, solar angle, and ambient temperature.

[0031] It should be noted that the numerical ranges of power generation per solar power generation in the park's integrated energy system, as well as the numerical ranges of light intensity, sunshine duration, solar angle, and ambient temperature stored in the database, are set based on historical experience and relevant common sense.

[0032] If the value of a certain data item in the historical power generation data set of the integrated energy system of the park exceeds the corresponding value range, then the historical power generation data set of that power generation is abnormal and is removed, resulting in a preprocessed data set of each historical power generation in the integrated energy system of the park. The dataset for the solar power generation prediction model is then constructed. The ratio between the training set and the test set is further set, and the dataset of the solar power generation prediction model is divided to obtain the training set and the test set of the solar power generation prediction model.

[0033] In one specific embodiment, the ratio between the training set and the test set is 3:1.

[0034] In this embodiment, the present invention obtains the dataset for training the distributed power generation prediction model by preprocessing the historical power generation data of the distributed power source in the integrated energy system of the park, thereby avoiding the interference of outlier data on the prediction model and improving the reliability of the distributed power generation prediction model.

[0035] Based on the training set of the solar power generation prediction model, machine learning algorithms are used to analyze the relationship between the power generation of solar power and the variables related to power generation, and a preliminary solar power generation prediction model is established in the park's integrated energy system.

[0036] As a preferred embodiment, the specific analysis process of step one further includes: S1: Based on the test set of the solar power generation prediction model, obtain the power generation of each historical power generation in the test set and the values ​​of each variable associated with solar power generation.

[0037] By substituting the values ​​of the variables associated with each historical power generation in the test set into the power generation prediction model for solar power generation, the estimated power generation for each historical power generation in the test set is obtained.

[0038] The power generation of each historical power generation in the test set is compared with the estimated power generation to obtain the deviation between the power generation of each historical power generation in the test set and the estimated power generation. The average value is then calculated to obtain the prediction deviation of the solar power generation prediction model.

[0039] It should be noted that the deviation between the power generation and the estimated power generation represents the absolute value of the difference between the power generation and the estimated power generation.

[0040] S2: Compare the prediction deviation of the solar power generation prediction model with the preset prediction deviation threshold. If the prediction deviation of the solar power generation prediction model is greater than the preset prediction deviation threshold, the solar power generation prediction model of the initially established integrated energy system of the park needs to be corrected, and S3 is executed. Otherwise, the solar power generation prediction model of the initially established integrated energy system of the park is used as the solar power generation prediction model of the integrated energy system of the park.

[0041] S3: Integrate the training set and test set of the solar power generation prediction model to obtain a new training set for the solar power generation prediction model. Based on the new training set, reconstruct the solar power generation prediction model and correct it to obtain the corrected solar power generation prediction model, which is recorded as the solar power generation prediction model in the park's integrated energy system.

[0042] It should be noted that this invention corrects the solar power generation prediction model by expanding the training set, i.e., by using data augmentation technology. Data augmentation technology increases the diversity and richness of the training data, enabling the model to learn more different features and patterns during the learning process, thereby improving the model's ability to generalize to new data.

[0043] As a preferred option, the specific analysis process in step one also includes: obtaining the power generation of each historical wind power generation in the park's integrated energy system and the values ​​of various variables associated with wind power generation, wherein the variables associated with wind power generation include wind speed, wind direction, air density, and power generation duration; analyzing the relationship between the power generation of wind power generation and the variables associated with it; and establishing a prediction model for the power generation of wind power in the park's integrated energy system.

[0044] It should be noted that wind speed is a key factor affecting wind power generation. Generally speaking, the higher the wind speed, the greater the wind energy received by the wind turbine blades, and the higher the power generation.

[0045] It should be noted that air density is related to factors such as altitude, temperature, and air pressure. In high-altitude areas, the air is thin and has a lower density, resulting in a relatively lower energy density of wind energy and thus reduced power generation at the same wind speed. Conversely, in low-temperature, high-pressure environments, the air density is higher, resulting in a relatively higher energy density of wind energy and a corresponding increase in power generation.

[0046] It should be noted that the stability and consistency of wind direction also affect power generation. A stable and consistent wind direction allows the blades of a wind turbine to always maintain the optimal angle of attack, thus capturing wind energy more effectively.

[0047] It should be noted that the power generation duration refers to the duration of the wind power generation process.

[0048] The project obtains the historical power generation data of each hydropower generation event in the integrated energy system of the park, as well as the values ​​of various variables associated with hydropower generation, including flow rate, water level difference, and power generation duration. The project analyzes the relationship between the power generation of hydropower and the variables associated with it, and establishes a power generation prediction model for hydropower in the integrated energy system of the park.

[0049] It should be noted that the flow rate refers to the amount of water passing through the turbine per unit time. The greater the flow rate, the more energy the turbine obtains, and the greater the power generation.

[0050] It should be noted that the greater the water level difference, the higher the efficiency of converting the water's potential energy into electrical energy.

[0051] It should be noted that the power generation duration refers to the duration of the wind power generation process.

[0052] It should be noted that the method for obtaining the power generation prediction model for wind power and hydropower in the integrated energy system of the park is based on the same principle as the method for obtaining the power generation prediction model for solar power in the integrated energy system of the park.

[0053] As a preferred option, the specific analysis process of step one also includes: setting the start and end times of the prediction period, obtaining the predicted values ​​of each variable related to solar power generation, wind power generation, and hydropower generation in the park's integrated energy system within the prediction period through the meteorological department, water conservancy department, and energy management department of the park's location, and substituting them into the power generation prediction models of solar power generation, wind power generation, and hydropower generation respectively to obtain the estimated power generation of solar power generation, wind power generation, and hydropower generation in the park's integrated energy system within the prediction period.

[0054] It should be noted that the prediction period can be one day, one week, half a month, etc.

[0055] Step 2: Predicting Park Load Day-ahead: Obtain the park's electricity consumption for each historical period in the current year and the park's electricity consumption for the prediction period in each historical year, and analyze the park's estimated electricity consumption for the prediction period.

[0056] As a preferred option, see [reference] Figure 3 As shown, the specific analysis process of step two is as follows: Obtain the time period of a set duration preceding the start time of the prediction cycle in the current year, and divide it according to the principle of equal duration to the prediction cycle to obtain each historical cycle of the current year. Obtain the electricity consumption of the park within each historical cycle of the current year, plot the first trend curve of the park's electricity consumption over time, obtain the function corresponding to the first trend curve of the park's electricity consumption over time, and substitute the prediction cycle into it to obtain the first estimated electricity consumption of the park within the prediction cycle, which is denoted as... .

[0057] In another specific embodiment, the time period between the start time of the current year and the start time of the prediction period is obtained, and then divided according to the principle of equal duration to the prediction period to obtain the historical periods of the current year.

[0058] It should be noted that the specific method for plotting the first trend curve of the park's electricity consumption over time is as follows: establish a coordinate system with historical cycles as the independent variable and the park's electricity consumption as the dependent variable; mark the corresponding data points in the coordinate system based on the park's electricity consumption in each historical cycle of the current year; and use the mathematical modeling method to plot the first trend curve of the park's electricity consumption over time.

[0059] Extract the park's electricity consumption for each historical year within the prediction period from the database, plot the second trend curve of the park's electricity consumption over time, obtain the function corresponding to the second trend curve of the park's electricity consumption over time, and substitute the prediction period into it to obtain the second estimated electricity consumption of the park within the prediction period, which is denoted as . .

[0060] It should be noted that the method for plotting the second trend curve of electricity consumption in the park over time is the same as the method for plotting the first trend curve of electricity consumption in the park over time.

[0061] By analyzing the formula The estimated electricity consumption of the park during the forecast period is obtained. ,in These represent the weights of the first and second preset estimated electricity consumption, respectively. , This indicates the compensation amount for the estimated electricity consumption of the park within the preset forecast period.

[0062] In one specific embodiment, the weights of the first estimated electricity consumption and the second estimated electricity consumption are 0.5 and 0.5, respectively.

[0063] In this embodiment, the present invention can help adjust the energy supply strategy and optimize the output of distributed power sources by predicting the output and load of distributed power sources in the park, thereby achieving the safe and stable operation of the park's integrated energy system.

[0064] Step 3: Park Power Supply and Demand Forecast: Based on the estimated power generation of distributed power sources in the park's integrated energy system and the estimated power consumption of the park during the forecast period, obtain the type of power supply and demand relationship in the park during the forecast period and the energy storage information of energy storage devices in the park's integrated energy system, and provide feedback.

[0065] As a preferred option, the specific analysis process of step three is as follows: based on the estimated power generation of solar power, wind power and hydropower in the park's integrated energy system during the forecast period, the estimated total power generation of the park during the forecast period is obtained. The estimated total power generation of the park during the forecast period is compared with the estimated power consumption to analyze the type of power supply and demand relationship in the park during the forecast period.

[0066] It should be noted that the specific method for analyzing the power supply and demand relationship type in the park during the forecast period is as follows: if the difference between the estimated total power generation and the estimated power consumption in the park during the forecast period is less than or equal to the set difference threshold, then the power supply and demand relationship type in the park during the forecast period is supply and demand balance; if the difference between the estimated total power generation and the estimated power consumption in the park during the forecast period is greater than the set difference threshold and the estimated total power generation is greater than the estimated power consumption, then the power supply and demand relationship type in the park during the forecast period is supply exceeding demand; if the difference between the estimated total power generation and the estimated power consumption in the park during the forecast period is greater than the set difference threshold and the estimated total power generation is less than the estimated power consumption, then the power supply and demand relationship type in the park during the forecast period is supply falling short of demand.

[0067] It should be noted that the difference between the estimated total power generation and the estimated power consumption refers to the absolute value of the difference between the estimated total power generation and the estimated power consumption.

[0068] By comparing the estimated total power generation and estimated power consumption of the park during the forecast period, the energy storage information of the energy storage devices in the park's integrated energy system during the forecast period is obtained. The energy storage information includes the energy storage status and the amount of energy stored. The energy storage status includes charging, discharging, and neither charging nor discharging.

[0069] It should be noted that the energy storage capacity of the energy storage device can be positive, negative, or zero. When the energy storage device is charging, the energy storage capacity is positive; when the energy storage device is discharging, the energy storage capacity is negative.

[0070] It should be noted that the specific method for obtaining the energy storage information of the energy storage devices in the park's integrated energy system during the forecast period is as follows: compare the estimated total power generation of the park with the estimated power consumption during the forecast period. If the estimated total power generation of the park during the forecast period is greater than the estimated power consumption, the energy storage status of the energy storage devices in the park's integrated energy system during the forecast period is charging. If the estimated total power generation of the park during the forecast period is less than the estimated power consumption, the energy storage status of the energy storage devices in the park's integrated energy system during the forecast period is discharging. If the estimated total power generation of the park during the forecast period is equal to the estimated power consumption, the energy storage status of the energy storage devices in the park's integrated energy system during the forecast period is neither charging nor discharging.

[0071] The difference between the estimated total power generation and the estimated power consumption of the park during the forecast period is recorded as the energy storage capacity of the energy storage devices in the park's integrated energy system during the forecast period.

[0072] In this embodiment, the present invention combines the output forecasting and load forecasting of distributed power sources in the park to obtain the power supply and demand relationship type of the park and the energy storage information of the energy storage devices in the integrated energy system. This is beneficial to optimize the charging and discharging strategy of the energy storage devices in the integrated energy system and better realize the coordinated operation of the integrated energy system.

[0073] Step 4: Evaluation of the accuracy of distributed power generation and load forecasting: Obtain the actual power generation of distributed power sources in the park's integrated energy system and the actual power consumption of the park during the forecasting period, analyze the forecasting accuracy of the park's distributed power generation and load, and provide feedback.

[0074] As a preferred option, the specific analysis process of step four is as follows: obtain the actual power generation of solar power, wind power and hydropower in the park's integrated energy system during the prediction period and the actual power consumption of the park, and compare them with the estimated power generation of solar power, wind power and hydropower and the estimated power consumption of the park to obtain the prediction difference of solar power, wind power and hydropower and the prediction difference of power consumption of the park.

[0075] It should be noted that the predicted difference refers to the absolute value of the difference between the estimated value and the actual value.

[0076] Establish the relationship functions between the forecast difference of solar power generation, wind power generation, and hydropower generation and the forecast accuracy of distributed power generation output, and the relationship functions between the forecast difference of electricity consumption and the forecast accuracy of load, and analyze the forecast accuracy of distributed power generation output and load in the industrial park.

[0077] It should be noted that the relationship between the prediction difference of solar power generation, wind power generation and hydropower generation and the prediction accuracy of distributed power generation output is a negative correlation function. That is, the larger the prediction difference of solar power generation, the larger the prediction difference of wind power generation and the larger the prediction difference of hydropower generation, the lower the prediction accuracy of distributed power generation output.

[0078] It should be noted that the relationship function between the electricity consumption forecast difference and the load forecast accuracy is a negative correlation function, that is, the larger the electricity consumption forecast difference, the lower the load forecast accuracy.

[0079] In this embodiment, the present invention obtains the actual power generation of distributed power sources and the actual power consumption of the park in the integrated energy system of the park, and compares them with the predicted values ​​to evaluate the prediction accuracy of the output and load of the distributed power sources in the park. This is beneficial for subsequent targeted correction and optimization of the prediction model, making the coordinated operation and scheduling of the energy system based on the prediction model more accurate and achieving the best operating state.

[0080] Step 5: Energy System Coordination Operation Optimization Assessment: Obtain the coordinated power volume, coordination frequency, and charge / discharge response speed of the energy storage devices in the park's integrated energy system during the assessment period. Also, obtain the power outage duration and number of power outages in the park's integrated energy system during the assessment period. Analyze the park's integrated energy system coordination operation optimization evaluation index and provide feedback.

[0081] As a preferred embodiment, the specific analysis process in step five includes: setting the duration of the evaluation period, obtaining the coordinated power, number of coordinations, and charge / discharge response speed of the energy storage devices in the park's integrated energy system within the evaluation period, and recording them as follows: , .

[0082] It should be noted that the method for obtaining the coordinated electricity volume and number of coordinations of the energy storage devices in the park's integrated energy system during the assessment period is as follows: obtain the total electricity volume charged and discharged by the energy storage devices in the park's integrated energy system during the assessment period, as well as the total number of charging and discharging times, and sum them up to obtain the coordinated electricity volume and number of coordinations of the energy storage devices in the park's integrated energy system during the assessment period.

[0083] It should be noted that the specific method for obtaining the charging and discharging response speed of the energy storage devices in the integrated energy system of the park during the evaluation period is as follows: obtain the average charging speed and average discharging speed of the energy storage devices in the integrated energy system of the park during the evaluation period, perform weighted average calculation, and obtain the charging and discharging response speed of the energy storage devices in the integrated energy system of the park during the evaluation period.

[0084] Obtain the duration and number of power outages of the park's integrated energy system during the assessment period, and record them as follows: .

[0085] It should be noted that coordinated power refers to the amount of electricity that energy storage devices charge or release within a specific time period according to the control strategy of the integrated energy system, in order to achieve functions such as power balancing and peak shaving in the power system.

[0086] As a preferred embodiment, the specific analysis process in step five further includes: analyzing the formulas... The evaluation index of coordinated operation optimization of the park's integrated energy system was obtained. ,in This represents the correction factor for the preset evaluation index of the coordinated operation of the park's integrated energy system. Represents the natural constant. These represent the preset thresholds for the coordinated power capacity, number of coordination cycles, and charge / discharge response speed of the energy storage device. These represent the influencing factors corresponding to the preset unit power outage duration and unit power outage frequency, respectively.

[0087] It should be noted that the correction factor of the park's integrated energy system coordinated operation optimization evaluation index serves to amplify, reduce, or round the index to facilitate calculation and analysis.

[0088] In this embodiment, the present invention evaluates the coordinated operation optimization of the integrated energy system of the park by acquiring the coordinated power, coordination times, charging and discharging response speed of the energy storage devices in the integrated energy system, as well as the power outage duration and number of power outages of the integrated energy system. This helps to identify the deficiencies in the coordinated operation of the energy system, provides a basis for further optimization of the operation strategy, and improves the overall performance and operational efficiency of the energy system.

[0089] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A method for optimizing the operation of an energy system based on distributed power sources and load forecasting, characterized in that, Includes the following steps: Step 1: Forecast of Distributed Power Output in the Park: Obtain historical power generation data of distributed power sources in the park's integrated energy system, including solar power, wind power, and hydropower. Analyze the relationship between the power generation of distributed power sources in the park's integrated energy system and the variables related to their power generation. Establish a forecast model for the power generation of distributed power sources in the park's integrated energy system. Further obtain the estimated power generation of distributed power sources in the park's integrated energy system within the forecast period. Step 2: Predicting Park Load Day-ahead: Obtain the park's electricity consumption for each historical period in the current year and the park's electricity consumption for the prediction period in each historical year, and analyze the park's estimated electricity consumption for the prediction period. Step 3: Park Power Supply and Demand Forecast: Based on the estimated power generation of distributed power sources in the park's integrated energy system and the estimated power consumption of the park during the forecast period, obtain the type of power supply and demand relationship in the park during the forecast period and the energy storage information of energy storage devices in the park's integrated energy system, and provide feedback. Step 4: Evaluation of the accuracy of distributed power generation and load forecasting: Obtain the actual power generation of distributed power sources in the park's integrated energy system and the actual electricity consumption of the park during the forecasting period, analyze the forecasting accuracy of the park's distributed power generation and load, and provide feedback. Step 5: Energy System Coordination Operation Optimization Assessment: Obtain the coordinated power volume, coordination frequency, and charge / discharge response speed of the energy storage devices in the park's integrated energy system during the assessment period. Also, obtain the power outage duration and number of power outages in the park's integrated energy system during the assessment period. Analyze the park's integrated energy system coordination operation optimization evaluation index and provide feedback.

2. The energy system optimization operation method based on distributed power generation and load forecasting according to claim 1, characterized in that: The specific analysis process in step one includes: Extract historical power generation data for solar, wind, and hydropower generation from the integrated energy system of the park stored in the database; Based on the historical power generation data of solar power generation in the park's integrated energy system, a data set of each historical power generation of solar power generation in the park's integrated energy system is obtained. The data set includes the values ​​of power generation and various variables related to solar power generation, including solar intensity, sunshine duration, solar angle, and ambient temperature. Extract the numerical range of power generation per solar power generation in the park's integrated energy system, as well as the numerical range of light intensity, sunshine duration, solar angle, and ambient temperature stored in the database. If the value of a certain data item in the historical power generation data set of the integrated energy system of the park exceeds the corresponding value range, then the historical power generation data set of that power generation is abnormal and is removed, thus obtaining the preprocessed data set of each historical power generation in the integrated energy system of the park. The dataset for the solar power generation prediction model is constructed, and the ratio between the training set and the test set is further set to divide the dataset of the solar power generation prediction model, thus obtaining the training set and test set of the solar power generation prediction model. Based on the training set of the solar power generation prediction model, machine learning algorithms are used to analyze the relationship between the power generation of solar power and the variables related to power generation, and a preliminary solar power generation prediction model is established in the park's integrated energy system.

3. The energy system optimization operation method based on distributed power generation and load forecasting according to claim 2, characterized in that: The specific analysis process in step one also includes: S1: Based on the test set of the solar power generation prediction model, obtain the power generation of each historical power generation in the test set and the values ​​of each variable associated with solar power generation; By substituting the values ​​of the variables associated with each historical power generation in the test set into the power generation prediction model for solar power generation, the estimated power generation of each historical power generation in the test set is obtained. The power generation of each historical power generation in the test set is compared with the estimated power generation to obtain the deviation between the power generation of each historical power generation in the test set and the estimated power generation. The average value is then calculated to obtain the prediction deviation of the solar power generation prediction model. S2: Compare the prediction deviation of the solar power generation prediction model with the preset prediction deviation threshold. If the prediction deviation of the solar power generation prediction model is greater than the preset prediction deviation threshold, the solar power generation prediction model of the initially established integrated energy system of the park needs to be corrected and S3 is executed. Otherwise, the solar power generation prediction model of the initially established integrated energy system of the park is used as the solar power generation prediction model of the integrated energy system of the park. S3: Integrate the training set and test set of the solar power generation prediction model to obtain a new training set for the solar power generation prediction model. Based on the new training set, reconstruct the solar power generation prediction model and correct it to obtain the corrected solar power generation prediction model, which is recorded as the solar power generation prediction model in the park's integrated energy system.

4. The energy system optimization operation method based on distributed power generation and load forecasting according to claim 3, characterized in that: The specific analysis process in step one also includes: The project aims to obtain the historical power generation data of each wind power generation event in the integrated energy system of the park, as well as the values ​​of various variables associated with wind power generation, including wind speed, wind direction, air density, and power generation duration. The project will analyze the relationship between the power generation of wind power generation and the variables associated with it, and establish a prediction model for the power generation of wind power in the integrated energy system of the park. The project obtains the historical power generation data of each hydropower generation event in the integrated energy system of the park, as well as the values ​​of various variables associated with hydropower generation, including flow rate, water level difference, and power generation duration. The project analyzes the relationship between the power generation of hydropower and the variables associated with it, and establishes a power generation prediction model for hydropower in the integrated energy system of the park.

5. The energy system optimization operation method based on distributed power generation and load forecasting according to claim 4, characterized in that: The specific analysis process in step one also includes: The start and end dates of the forecast period are set. The predicted values ​​of each variable related to solar power generation, wind power generation, and hydropower generation in the park's integrated energy system are obtained from the meteorological department, water conservancy department, and energy management department of the park's location within the forecast period. These values ​​are then substituted into the power generation forecast models for solar power generation, wind power generation, and hydropower generation to obtain the estimated power generation of solar power generation, wind power generation, and hydropower generation in the park's integrated energy system within the forecast period.

6. The energy system optimization operation method based on distributed power generation and load forecasting according to claim 1, characterized in that: The specific analysis process for step two is as follows: Obtain the time period of a set duration preceding the start time of the forecast period in the current year, and divide it according to the principle of equal duration to the forecast period to obtain the historical periods for the current year. Obtain the electricity consumption of the park within each historical period of the current year, plot the first trend curve of the park's electricity consumption over time, obtain the function corresponding to the first trend curve of the park's electricity consumption over time, and substitute the forecast period into it to obtain the first estimated electricity consumption of the park within the forecast period, which is denoted as... ; Extract the park's electricity consumption for each historical year within the prediction period from the database, plot the second trend curve of the park's electricity consumption over time, obtain the function corresponding to the second trend curve of the park's electricity consumption over time, and substitute the prediction period into it to obtain the second estimated electricity consumption of the park within the prediction period, which is denoted as . ; By analyzing the formula The estimated electricity consumption of the park during the forecast period is obtained. ,in These represent the weights of the first and second preset estimated electricity consumption, respectively. , This indicates the compensation amount for the estimated electricity consumption of the park within the preset forecast period.

7. The energy system optimization operation method based on distributed power generation and load forecasting according to claim 1, characterized in that: The specific analysis process in step three is as follows: Based on the estimated power generation of solar power, wind power and hydropower in the park's integrated energy system during the forecast period, the estimated total power generation of the park during the forecast period is obtained. The estimated total power generation of the park during the forecast period is compared with the estimated power consumption to analyze the type of power supply and demand relationship in the park during the forecast period. By comparing the estimated total power generation and estimated power consumption of the park during the forecast period, the energy storage information of the energy storage devices in the park's integrated energy system during the forecast period is obtained. The energy storage information includes the energy storage status and the amount of energy stored. The energy storage status includes charging, discharging, and neither charging nor discharging.

8. The energy system optimization operation method based on distributed power generation and load forecasting according to claim 1, characterized in that: The specific analysis process in step four is as follows: The actual power generation of solar power, wind power and hydropower in the park's integrated energy system and the actual electricity consumption of the park during the forecast period are obtained. They are then compared with the estimated power generation of solar power, wind power and hydropower and the estimated electricity consumption of the park to obtain the forecast difference of solar power, wind power and hydropower and the forecast difference of electricity consumption of the park. Establish the relationship functions between the forecast difference of solar power generation, wind power generation, and hydropower generation and the forecast accuracy of distributed power generation output, and the relationship functions between the forecast difference of electricity consumption and the forecast accuracy of load, and analyze the forecast accuracy of distributed power generation output and load in the industrial park.

9. The energy system optimization operation method based on distributed power generation and load forecasting according to claim 1, characterized in that: The specific analysis process in step five includes: Set the duration of the evaluation period, obtain the coordinated power consumption, number of coordinations, and charge / discharge response speed of the energy storage devices in the park's integrated energy system within the evaluation period, and record them as follows: , ; Obtain the duration and number of power outages of the park's integrated energy system during the assessment period, and record them as follows: .

10. The energy system optimization operation method based on distributed power generation and load forecasting according to claim 9, characterized in that: The specific analysis process in step five also includes: By analyzing the formula The evaluation index of coordinated operation optimization of the park's integrated energy system was obtained. ,in This represents the correction factor for the preset evaluation index of the coordinated operation of the park's integrated energy system. Represents the natural constant. These represent the preset thresholds for the coordinated power capacity, number of coordination cycles, and charge / discharge response speed of the energy storage device. These represent the influencing factors corresponding to the preset unit power outage duration and unit power outage frequency, respectively.

Images