An energy optimization scheduling method and system based on data analysis

Abstract

This application provides an energy optimization scheduling method and system based on data analysis. The method includes collecting in real time the power generation data, load demand, and meteorological data of solar energy, wind energy, and energy storage systems, and identifying the power generation patterns and load change trends through data analysis techniques; based on the analysis results, constructing a two-layer energy scheduling model, with the main objectives of maximizing system reliability, scheduling flexibility, and minimizing carbon emissions, and the secondary objective of minimizing scheduling costs; introducing the uncertainties of wind and solar power output and multiple joint constraints of information gap decision-making; using the non-dominated sorting genetic algorithm NSGA-II combined with crowding distance and elitist strategy for optimization and solution, and outputting the Pareto front solution set to obtain the optimal scheduling strategy. This invention ensures the timeliness and accuracy of scheduling decisions, effectively improves the utilization efficiency of renewable energy, reduces carbon emissions, and provides strong support for smart grids and sustainable development.

Description

Technical Field

[0001] The present invention relates to the field of electric energy management and dispatching, and in particular to an energy optimization dispatching method and system based on data analysis. Background Art

[0002] With the continuous growth of global energy demand and the enhancement of environmental protection awareness, the use of traditional fossil energy is facing increasing pressure. Renewable energy, especially solar energy and wind energy, has become an important direction for future energy development due to its clean and renewable characteristics. However, the generation of solar energy and wind energy is highly uncertain and volatile, which brings many challenges to the dispatch and management of power systems.

[0003] Existing energy dispatch methods are often based on static models and lack the ability to respond dynamically to real-time data. This results in the inability to adjust the dispatch strategy in a timely manner when faced with sudden changes in load or meteorological conditions, affecting the reliability and economy of the system. Many traditional dispatch models fail to effectively consider the uncertainty of renewable energy generation, especially when the wind and solar output fluctuates greatly, which can easily lead to an imbalance in electricity supply and demand and increase system risks. Existing methods often single out dispatch objectives and lack comprehensive consideration of multi-objective optimization. For example, the trade-offs between objectives such as carbon emissions, dispatch costs, and system reliability have not been effectively resolved. Although data analysis technology has been widely used in many fields, in the field of energy dispatch, especially the processing and analysis of real-time data is still insufficient, resulting in the inability to accurately identify the changing trends of power generation patterns and load demand.

[0004] Therefore, a more accurate data analysis method and modeling are urgently needed to improve the satisfaction of energy scheduling. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes an energy optimization scheduling method and system based on data analysis.

[0006] The present invention specifically provides the following technical solutions:

[0007] A method for optimizing energy scheduling based on data analysis, the method comprising the following steps:

[0008] Step S1: Real-time collection of power generation data, load demand data and meteorological data of solar energy, wind energy and energy storage systems;

[0009] Step S2: Processing the collected data using data analysis technology to identify the power generation mode of renewable energy and the change trend of load demand;

[0010] Step S3: Based on the results of the above data analysis, the uncertainty of wind and solar power output and the joint constraints of information gap decision-making are introduced, and a two-layer energy scheduling model is constructed with the main goals of maximizing system reliability, optimizing scheduling flexibility, and minimizing carbon emissions, and the secondary goal of minimizing scheduling costs;

[0011] Step S4: using the non-dominated sorting genetic algorithm NSGA-II combined with the optimization method of crowding distance and elite strategy to solve the above energy scheduling model, output the solution set on the Pareto frontier, and obtain the optimal scheduling strategy;

[0012] Step S5: Deploy an energy scheduling plan based on the above optimal scheduling strategy; and dynamically adjust the scheduling strategy based on real-time data and forecast information.

[0013] Optionally, in step S1,

[0014] High-precision photovoltaic power generation monitoring sensors, wind speed sensors and battery management system BMS sensors are used to collect corresponding data at a frequency of once per minute; 5G is used for real-time data transmission.

[0015] Optionally, in step S2,

[0016] The collected data is preprocessed and the partial autocorrelation function PACF is used to extract the trend, seasonality and periodicity characteristics of the data; at the same time, the characteristics of the meteorological data are extracted, including temperature, humidity and wind speed;

[0017] The integrated management system of the virtual power plant is used to integrate the real-time power generation data of different power sources, and the K-means clustering algorithm is used to cluster the power generation data to identify different power generation modes. The autoregressive integral moving average model ARIMA is used to model the load demand data and identify the changing trend of load demand. Polynomial regression analysis is used to explore the relationship between renewable energy power generation and load demand, and to identify the key factors affecting the changes in power generation mode and load demand, including meteorological factors, load characteristics, power source characteristics and market factors.

[0018] Optionally, in step S2, the independent variables specifically include the following:

[0019] The meteorological factors include temperature T, humidity H, wind speed W, and sunshine duration L; the load characteristics include load peak , load valley , load change rate R; the power source characteristics include photovoltaic power generation capacity , Wind power generation capacity , energy storage efficiency E; the market factors include electricity price , demand response D;

[0020] Based on the above independent variables, a second-order polynomial regression model is constructed to predict the load demand Y, specifically:

[0021]

[0022] in, is the intercept term, and is the above independent variable, is the corresponding linear term regression coefficient, is the regression coefficient of the interaction term;

[0023] Utilizing indicators To evaluate the data fitting effect of the above model, The closer it is to 1, the better the effect of predicting load demand is; where RSS represents the sum of squares of the differences between the observed values ​​and the model predicted values, and TSS represents the total variation of the dependent variable.

[0024] Optionally, in step S3, the two-layer energy scheduling model is divided into an upper layer model and a lower layer model; the upper layer model solves the scheduling strategy to maximize the reliability of the system Scheduling flexibility and minimizing carbon emissions is the objective function Z of the upper model up , expressed as:

[0025]

[0026] in, is the availability of the energy dispatch system, represented by the probability of being available within a given time t, is the system available time, is the total time; is the load satisfaction rate, which is the proportion of actual load demand that is satisfied within a certain period of time t. To satisfy the total load, is the total load demand; is the dispatch response time, specifically the response time of the energy dispatch system to load changes. The scheduling adjustment capacity is the amount of power generation that can be adjusted within a certain period of time t, which is calculated by the sum of the differences between the maximum and minimum power generation capacities of each power generation unit; is the total carbon emissions of the dispatching system within a certain period of time t, through the unit carbon emissions of each power generation unit It is calculated by multiplying the power generation P by the sum of the products; , , , , , , is the preset weight of the corresponding parameter, which is used to balance the importance of each goal;

[0027] The goal of the lower model is to minimize the scheduling cost. down , expressed as:

[0028]

[0029] in, , , , , are the combination of photovoltaic power generation, load power, grid power, fuel-related costs of power generation facilities required to meet power demand, and the cost of flexible scheduling; N represents the duration of the forecast horizon given a specific time t, t+N is the cumulative sum of the duration, is the photovoltaic photoelectric quantity in time t, is the constant cost of photovoltaic power generation; is the cost of purchasing from and selling to the grid, and The amount of electricity purchased and sold to the grid, respectively; and are the efficiencies of the pumps and turbines in the grid, and The power generation capacity of the equipment pumps and turbines in the power grid respectively; is the maximum power generated by DG, F is the amount of diesel consumed per kilowatt-hour, is the cost per litre of fuel, and are the cost and quality of CO2 emissions, respectively; and are the flexibility and corresponding cost of scheduling energy in time period k, respectively.

[0030] Optionally, the constraints of the two-layer energy scheduling model include active power balance constraints:

[0031]

[0032] in, , , , They are the photovoltaic power generation, the power generation of the pumps and turbines in the grid equipment, and the power generated by the DG; and They are load power and grid power sales power respectively;

[0033] Power generation capacity constraints:

[0034]

[0035] Among them, and represent the actual output of wind and photovoltaic power generation respectively; and denote the predicted output power of wind and photovoltaic respectively;

[0036] Energy conversion equipment constraints:

[0037]

[0038] in, represents the input power of the type i energy conversion device; Indicates the maximum capacity limit of type i equipment; represents the climbing rate constraint of the i-type device;

[0039] Energy storage constraints:

[0040]

[0041] in, is the energy storage state at time t; and are the charging and discharging efficiencies, respectively; and are the charging and discharging power of the energy storage system, respectively;

[0042] Carbon emission constraints:

[0043]

[0044] in, is the carbon emission coefficient of the ith energy source; is the power generation of the i-th energy source; is the maximum carbon emission allowed;

[0045] Information gap parameter constraints:

[0046]

[0047] in, Represents the actual objective function value Estimation of objective function value based on uncertainty information The information gap between It is the set information gap threshold, which indicates the risk range that decision makers can tolerate;

[0048] Uncertainty constraints on wind and solar power output:

[0049]

[0050] in, and are wind and solar power generation respectively; To meet the load requirements, is the safety factor; For maximum power generation capacity.

[0051] Optionally, in step S4, the specific process of optimization solution is as follows:

[0052] S41. Set the size N of the initial population; each individual represents a scheduling strategy, encode the individual, and the individual includes the scheduling amount of solar energy, wind energy, and energy storage; randomly generate individuals in the initial population to ensure that the entire feasible solution space is covered;

[0053] S42. Calculate the fitness of each individual according to the objective function defined in step S3;

[0054] S43. Perform non-dominated sorting on the population and determine the dominance rank of each individual;

[0055] S44. For each individual in each level, calculate its crowding distance; when selecting individuals, first sort them according to the non-dominated level, and then sort them according to the crowding distance in the same level;

[0056] S45. Roulette selection is used to give priority to individuals with low non-dominated levels and large crowding distances to ensure the retention of excellent individuals;

[0057] S46. Use uniform crossover to generate new individuals; mutate the newly generated individuals;

[0058] S47. In each generation, the top 10% of excellent individuals are retained and directly passed on to the next generation to ensure the inheritance of excellent genes;

[0059] S48. Set the maximum number of iterations; monitor the change of population fitness, and terminate the algorithm when the fitness change is less than the set threshold;

[0060] S49. Finally, the solution set on the Pareto frontier is output, representing the optimal scheduling strategy.

[0061] Optionally, in step S4,

[0062] A sharing mechanism is introduced in the fitness calculation process to reduce competition between similar individuals and encourage the population to explore different solutions. At the same time, parallel computing is used to reduce the time complexity of the algorithm to O(nlogn).

[0063] Optionally, the optimization scheduling method further includes:

[0064] Step S6: Evaluate the deployed energy scheduling scheme, compare the actual operation data with the predicted data, and verify the effectiveness and reliability of the scheduling strategy;

[0065] Step S7: Establish a feedback mechanism to feed back the actual operation results to the data analysis system;

[0066] Step S8: Conduct multi-scenario simulation, consider different load demands, meteorological conditions and equipment failures, and formulate corresponding emergency plans;

[0067] Step S9: Provide visualized scheduling results and real-time data, allowing users to interact and make manual adjustments.

[0068] The present invention also provides an energy optimization scheduling system based on data analysis, the system comprising:

[0069] Data acquisition module, used to collect real-time power generation data, load demand data and meteorological data of solar energy, wind energy and energy storage systems;

[0070] A data analysis module, used to process the collected data using data analysis technology to identify the power generation mode of renewable energy and the change trend of load demand;

[0071] The dispatch model building module is used to introduce uncertainty considerations of wind and solar power output and information gap decision theory based on data analysis results to build a two-layer energy dispatch model with the main goals of achieving maximum system reliability, optimal dispatch flexibility, and minimum carbon emissions, and the secondary goal of minimizing dispatch costs;

[0072] The optimization solution module uses the non-dominated sorting genetic algorithm NSGA-II combined with the optimization method of crowding distance and elite strategy to solve the above energy scheduling model, output the solution set on the Pareto frontier, and obtain the optimal scheduling strategy;

[0073] The scheduling execution module deploys the energy scheduling plan according to the optimal scheduling strategy and dynamically adjusts the scheduling strategy according to real-time data and forecast information.

[0074] The present invention has the following beneficial technical effects: The present invention provides an energy optimization scheduling method and system based on data analysis. By real-time collection and analysis of data from solar energy, wind energy and energy storage systems, it is possible to accurately identify power generation modes and load demand changes, thereby optimizing scheduling strategies and maximizing the utilization rate of renewable energy; combining machine learning and virtual power plant data analysis methods to improve the prediction accuracy and processing efficiency of power generation and demand response; a two-layer energy scheduling model is designed, with maximum system reliability, optimal scheduling flexibility and minimum carbon emissions as the main goals, and minimum scheduling cost as the secondary goal, and the introduction of the uncertainty of wind and solar output and the joint constraints of information gap decision-making, so that the scheduling model can maintain high reliability and flexibility of the system in the face of uncertainty, ensuring effective operation under various meteorological conditions and load demands; combined with the non-dominated sorting genetic algorithm NSGA-II, and the introduction of the optimization method of congestion distance and elite strategy, the selection of the optimal strategy is guaranteed, which can reduce the overall scheduling cost while ensuring the reliability of the scheduling system. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0076] Figure 1 A flow chart of an energy optimization scheduling method based on data analysis provided in an embodiment of the present invention.

[0077] Figure 2 A flowchart for solving a two-layer energy scheduling model provided for an embodiment of the present invention.

[0078] Figure 3 A block diagram of an energy optimization scheduling system based on data analysis provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0079] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0080] The following specific examples are used to illustrate the embodiments of the present application. Those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other in the absence of conflict. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work belong to the scope of protection of the present application.

[0081] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any labeled structures and / or functions described herein are merely illustrative. Based on the present application, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number and aspect set forth herein may be used to implement an apparatus and / or practice a method.

[0082] It should also be noted that the illustrations provided in the following embodiments are only used to schematically illustrate the basic concept of the present application.

[0083] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the examples can be practiced without these marked details.

[0084] The purpose of the present invention is to provide an energy optimization scheduling method and system based on data analysis, aiming to improve the satisfaction of energy scheduling strategy.

[0085] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0086] Reference Figure 1 , showing an energy optimization scheduling method based on data analysis according to an embodiment of the present application, the method comprises the following steps:

[0087] Step S1: Real-time collection of power generation data, load demand data and meteorological data of solar energy, wind energy and energy storage systems.

[0088] Optionally, high-precision photovoltaic power generation monitoring sensors are used to monitor the power generation, efficiency and operating status of the photovoltaic power generation system in real time to ensure accurate power generation data; wind speed sensors are used to measure wind speed and wind direction to evaluate the potential and real-time power generation capacity of wind power generation; and battery management system BMS sensors are used to monitor the charge and discharge status, remaining power and health status of the energy storage system to ensure efficient operation of the energy storage system. The data collection frequency is once a minute to ensure sufficient real-time data is obtained to capture the rapidly changing power generation and load demand conditions; 5G is used for real-time data transmission to ensure high speed and low latency of data transmission, so that the collected data can be quickly transmitted to the data analysis system for processing. Through efficient data collection and transmission technology, the intelligence level and responsiveness of the entire energy dispatching system are improved.

[0089] Step S2: Process the collected data using data analysis technology to identify the power generation mode of renewable energy and the change trend of load demand.

[0090] Optionally, the collected data is preprocessed, and the trend, seasonality and periodicity characteristics of the data are extracted using the partial autocorrelation function PACF; the characteristics of the meteorological data are also extracted, including temperature, humidity and wind speed;

[0091] The integrated management system of the virtual power plant is used to integrate the real-time power generation data of different power sources, and the K-means clustering algorithm is used to cluster the power generation data to identify different power generation modes. The autoregressive integral moving average model ARIMA is used to model the load demand data and identify the changing trend of load demand. Polynomial regression analysis is used to explore the relationship between renewable energy power generation and load demand, and to identify the key factors affecting the changes in power generation mode and load demand, including meteorological factors, load characteristics, power source characteristics and market factors.

[0092] Specifically, the partial autocorrelation function PACF is used to extract the trend, seasonality and periodicity characteristics of time series data. PACF can help identify the degree of autocorrelation in the data, thus providing a basis for subsequent modeling. This process helps to remove noise and enhance the interpretability of the signal. Meteorological data feature extraction Meteorological factors: including temperature, humidity and wind speed, which directly affect the power generation capacity of renewable energy. The integrated management system of the virtual power plant is used for data integration and clustering analysis, and the real-time power generation data of different power sources, such as solar energy and wind energy, are integrated for comprehensive analysis. The integrated power generation data is clustered to identify different power generation modes. By classifying similar power generation data into one category, potential power generation modes can be discovered to optimize the scheduling strategy. The autoregressive integrated moving average model ARIMA is used to model the load demand data and identify the changing trend of load demand. The ARIMA model captures the dynamic characteristics of time series data and is suitable for predicting future load demand. The relationship between renewable energy generation and load demand is explored using polynomial regression analysis. By constructing a polynomial regression model, the key factors affecting the changes in power generation mode and load demand can be identified. These factors include: Meteorological factors: such as temperature, humidity, wind speed and sunshine duration. Load characteristics: such as load peak, load valley and load change rate. Power source characteristics: such as photovoltaic power generation capacity, wind power generation capacity and energy storage efficiency. Market factors: such as electricity price and demand response.

[0093] Preferably, in step S2, the independent variables specifically include the following:

[0094] The meteorological factors include temperature T, humidity H, wind speed W, and sunshine duration L; the load characteristics include load peak , load valley value , load change rate R; the power source characteristics include photovoltaic power generation capacity , Wind power generation capacity , energy storage efficiency E; the market factors include electricity price , demand response D;

[0095] Based on the above independent variables, a second-order polynomial regression model is constructed to predict the load demand Y, specifically:

[0096]

[0097] in, is the intercept term, and is the above independent variable, is the corresponding linear term regression coefficient, is the regression coefficient of the interaction term;

[0098] Utilizing indicators To evaluate the data fitting effect of the above model, The closer it is to 1, the better the effect of predicting load demand is; where RSS represents the sum of squares of the differences between the observed values ​​and the model predicted values, and TSS represents the total variation of the dependent variable.

[0099] Prior to this, the values ​​of the regression coefficients are given through data fitting and analysis. β0=10; β1=0.5 (temperature); β2=−0.3 (humidity); β3=0.2 (wind speed); β4=0.4 (sun duration); β5=1.5 (load peak); β6=−0.8 (load valley); β7=0.1 (load change rate); β8=1.2 (photovoltaic power generation capacity); β9=0.9 (wind power generation capacity); β10=0.05 (energy storage efficiency); β11=0.3 (electricity price); β12=0.2 (demand response). Similarly, the interaction coefficients βTT= 0.05; βHH= -0.02; βPP=0.01; βTH= -0.03; βTP=0.015; βHP=0.02. The values ​​of the remaining interaction terms are not given in detail, but they do not affect the quality of the prediction model.

[0100] Step S3: Based on the results of the above data analysis, the uncertainty of wind and solar power output and the joint constraints of information gap decision-making are introduced, and a two-layer energy scheduling model is constructed with maximum system reliability, optimal scheduling flexibility, and minimum carbon emissions as the main goals and minimum scheduling cost as the secondary goal.

[0101] Specifically, in step S3, the two-layer energy scheduling model is divided into an upper model and a lower model; the upper model solves the scheduling strategy to maximize the reliability of the system Scheduling flexibility and minimizing carbon emissions is the objective function Z of the upper model up , expressed as:

[0102]

[0103] in, is the availability of the energy dispatch system, represented by the probability of being available within a given time t, is the system available time, is the total time; is the load satisfaction rate, which is the proportion of actual load demand that is satisfied within a certain period of time t. To satisfy the total load, is the total load demand; is the dispatch response time, specifically the response time of the energy dispatch system to load changes. The scheduling adjustment capacity is the amount of power generation that can be adjusted within a certain period of time t, which is calculated by the sum of the differences between the maximum and minimum power generation capacities of each power generation unit; is the total carbon emissions of the dispatching system within a certain period of time t, through the unit carbon emissions of each power generation unit It is calculated by multiplying the power generation P by the sum of the products; , , , , , , is the preset weight of the corresponding parameter, which is used to balance the importance of each goal;

[0104] The goal of the lower model is to minimize the scheduling cost. down , expressed as:

[0105]

[0106] in, , , , , are the combination of photovoltaic power generation, load power, grid power, fuel-related costs of power generation facilities required to meet power demand, and the cost of flexible scheduling; N represents the duration of the forecast horizon given a specific time t, t+N is the cumulative sum of the duration, is the photovoltaic photoelectric quantity in time t, is the constant cost of photovoltaic power generation; is the cost of purchasing from and selling to the grid, and The amount of electricity purchased and sold to the grid, respectively; and are the efficiencies of the pumps and turbines in the grid, and The power generation capacity of the equipment pumps and turbines in the power grid respectively; is the maximum power generated by DG, F is the amount of diesel consumed per kilowatt-hour, is the cost per litre of fuel, and are the cost and quality of CO2 emissions, respectively; and are the flexibility and corresponding cost of scheduling energy in time period k, respectively.

[0107] Optionally, the constraints of the two-layer energy scheduling model include active power balance constraints:

[0108]

[0109] in, , , , They are the photovoltaic power generation, the power generation of the pumps and turbines in the grid equipment, and the power generated by the DG; and They are load power and grid power sales power respectively;

[0110] Power generation capacity constraints:

[0111]

[0112] Among them, and represent the actual output of wind and photovoltaic power generation respectively; and denote the predicted output power of wind and photovoltaic respectively;

[0113] Energy conversion equipment constraints:

[0114]

[0115] in, represents the input power of the type i energy conversion device; Indicates the maximum capacity limit of type i equipment; represents the climbing rate constraint of the i-type device;

[0116] Energy storage constraints:

[0117]

[0118] in, is the energy storage state at time t; and are the charging and discharging efficiencies, respectively; and are the charging and discharging power of the energy storage system, respectively;

[0119] Carbon emission constraints:

[0120]

[0121] in, is the carbon emission coefficient of the ith energy source; is the power generation of the i-th energy source; is the maximum carbon emission allowed;

[0122] Information gap parameter constraints:

[0123]

[0124] in, Represents the actual objective function value Estimation of objective function value based on uncertainty information The information gap between It is the set information gap threshold, which indicates the risk range that decision makers can tolerate;

[0125] Uncertainty constraints on wind and solar power output:

[0126]

[0127] in, and are wind and solar power generation respectively; To meet the load requirements, is the safety factor; For maximum power generation capacity.

[0128] Step S4: The non-dominated sorting genetic algorithm NSGA-II is used in combination with the optimization method of crowding distance and elite strategy to solve the above energy scheduling model, output the solution set on the Pareto frontier, and obtain the optimal scheduling strategy.

[0129] Specifically, Figure 2 As shown,

[0130] S41. Initialize the population size: Set the population size N=100. Individual encoding: Each individual represents a scheduling strategy, and the encoding can be a vector containing the scheduling amount of each energy source (such as solar energy, wind energy, and energy storage). Random generation: Randomly generate individuals in the initial population to ensure that the entire feasible solution space is covered.

[0131] S42. Fitness evaluation objective function: Calculate the fitness of each individual according to the objective defined in step S3. The objective function includes: system reliability (such as availability, failure rate, etc.), scheduling flexibility (such as response time, scheduling adjustment capability, etc.), carbon emissions (such as carbon emissions per unit of electricity), and scheduling costs (such as operating costs, maintenance costs, etc.).

[0132] S43. Non-dominated sorting Non-dominated sorting: Perform non-dominated sorting on the population to determine the dominance level of each individual. The dominance relationship is defined as: if individual A is superior to individual B in all objectives, then A dominates B. Level assignment: Assign individuals to different levels, the lower the level, the higher the superiority of the individual.

[0133] S44. Crowding distance calculation Crowding distance: For each individual in each level, calculate its crowding distance. Crowding distance is evaluated by comparing the distance of individuals in the target space. The larger the distance, the sparser the individuals are in the target space. Sorting: When selecting individuals, first sort them according to non-dominated level, and then sort them according to crowding distance within the same level.

[0134] S45. Selection operation selection strategy: Use tournament selection or roulette selection to give priority to individuals with low non-dominated levels and large crowding distances to ensure the retention of excellent individuals.

[0135] S46. Crossover and mutation Crossover operation: Use single-point crossover or uniform crossover to generate new individuals. The crossover probability is set to 0.7. Mutation operation: Mutate the newly generated individuals, and the mutation probability is set to 0.1 to ensure the diversity of the population.

[0136] S47. Elite Strategy Elite Retention: In each generation, retain the top 10% of outstanding individuals and directly enter the next generation to ensure the inheritance of excellent genes.

[0137] S48. Termination condition Iteration number: Set the maximum number of iterations to 200. Convergence: Monitor the change of population fitness, and terminate the algorithm when the fitness change is less than the set threshold.

[0138] S49. Output the optimal solution Pareto front: The final output is the solution set on the Pareto front, which represents the best compromise between different objectives.

[0139] For example, power generation capacity: the maximum power generation capacity of solar and wind power (such as kW). Load demand: real-time load demand (such as kW). Energy storage capacity: the maximum charge and discharge capacity of the energy storage system (such as kWh). Carbon emission factor: the carbon emission factor of different energy sources (such as gCO2 / kWh). Dispatch cost: the operating cost of each energy source (such as yuan / kWh).

[0140] Exemplarily, the dispatch strategy includes but is not limited to the distribution of power generation of various renewable energy sources (such as solar energy, wind energy, and energy storage) within a specific time period; load response, the response strategy under different load demand conditions, including how to adjust power generation to meet load demand; and the dispatch schedule, a specific dispatch schedule that indicates when to enable or disable specific power generation units.

[0141] For example, 1. Allocation of renewable energy generation Solar power generation allocation: Based on real-time meteorological data (such as light intensity, temperature, etc.) and historical power generation data, predict the solar power generation capacity within a specific time period. In periods with good lighting conditions (such as noon), prioritize solar power generation to maximize its power generation. In periods with insufficient light (such as cloudy days or at night), reduce the dispatch of solar power generation and increase the power generation of other energy sources. Wind power generation allocation: Based on real-time wind speed data and the characteristics of wind power generators, predict wind power generation capacity. In periods with moderate and stable wind speeds, prioritize wind power generation. When the wind speed is too high or too low, adjust the dispatch of wind power generation to avoid equipment damage or inefficient operation. Energy storage system power generation allocation: When there is excess renewable energy generation (such as peak solar power generation during the day), store excess electricity in the energy storage system. When there is insufficient renewable energy generation (such as at night or when the wind speed is insufficient), prioritize the energy storage system to release electricity to meet load demand. 2. Load response strategy Load forecasting and response: Use the autoregressive integrated moving average model (ARIMA) to predict load demand and identify load peaks and valleys. During peak load periods, prioritize renewable energy generation and release electricity in combination with energy storage systems to meet demand. During low load periods, reduce power generation, especially non-renewable energy generation, to avoid resource waste. Demand response mechanism: Through interaction with users, implement demand response strategies to encourage users to reduce electricity consumption during peak load periods (such as providing electricity price discounts). During peak load periods, adjust power generation strategies and give priority to the use of low-carbon and renewable energy to reduce carbon emissions. 3. Scheduling schedule Daily scheduling schedule: Morning (6:00-9:00): According to the morning load demand and the gradual increase in solar power generation, moderately dispatch solar power generation and release electricity in combination with energy storage systems. Noon (9:00-15:00): During the period of strongest sunlight, prioritize solar power generation to maximize its power generation, while monitoring wind power generation. Afternoon (15:00-18:00): According to the changes in load demand, timely adjust the power generation of solar and wind energy to ensure that the load is met. Evening (18:00-21:00): During the peak load period, dispatch the energy storage system to release electricity to ensure that user needs are met. Nighttime (21:00-6:00): Reduce power generation, mainly rely on the energy storage system to ensure the economy and stability of the system. Special situation dispatching: Under extreme weather conditions (such as storms, extreme high temperatures, etc.), dynamically adjust the power generation strategy based on real-time data to ensure the safety and stable operation of the system. During equipment failure or maintenance, timely adjust the dispatch strategy to ensure that other power generation units can supplement the missing power generation capacity.

[0142] Preferably, a sharing mechanism is introduced in the fitness calculation process to reduce competition between similar individuals and encourage the population to explore different solutions; at the same time, parallel computing is used to reduce the time complexity of the algorithm to O(nlogn).

[0143] Step S5: Deploy an energy scheduling plan based on the above optimal scheduling strategy; and dynamically adjust the scheduling strategy based on real-time data and forecast information.

[0144] Optionally, the optimization scheduling method further includes:

[0145] Step S6: Evaluate the deployed energy scheduling scheme, compare the actual operation data with the predicted data, and verify the effectiveness and reliability of the scheduling strategy;

[0146] Step S7: Establish a feedback mechanism to feed back the actual operation results to the data analysis system;

[0147] Step S8: Conduct multi-scenario simulation, consider different load demands, meteorological conditions and equipment failures, and formulate corresponding emergency plans;

[0148] Step S9: Provide visualized scheduling results and real-time data, allowing users to interact and make manual adjustments.

[0149] In one embodiment of the present application, Figure 3 As shown, an energy optimization scheduling system based on data analysis is also provided, and the system includes:

[0150] Data acquisition module, used to collect real-time power generation data, load demand data and meteorological data of solar energy, wind energy and energy storage systems;

[0151] A data analysis module, used to process the collected data using data analysis technology to identify the power generation mode of renewable energy and the change trend of load demand;

[0152] The dispatch model building module is used to introduce uncertainty considerations of wind and solar power output and information gap decision theory based on data analysis results to build a two-layer energy dispatch model with the main goals of achieving maximum system reliability, optimal dispatch flexibility, and minimum carbon emissions, and the secondary goal of minimizing dispatch costs;

[0153] The optimization solution module uses the non-dominated sorting genetic algorithm NSGA-II combined with the optimization method of crowding distance and elite strategy to solve the above energy scheduling model, output the solution set on the Pareto frontier, and obtain the optimal scheduling strategy;

[0154] The scheduling execution module deploys the energy scheduling plan according to the optimal scheduling strategy and dynamically adjusts the scheduling strategy according to real-time data and forecast information.

[0155] Based on the same inventive concept, according to another aspect of the present invention, an embodiment of the present invention further provides a computer device, including:

[0156] at least one processor; and

[0157] The memory stores a computer program that can be run on the processor, and the processor executes the steps of any of the above methods for testing database performance when executing the program.

[0158] Based on the same inventive concept, according to another aspect of the present invention, an embodiment of the present invention further provides a computer-readable storage medium, which stores computer program instructions. When the computer program instructions are executed by a processor, the steps of any one of the above management methods are performed.

[0159] Finally, it should be noted that a person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only storage memory (ROM) or a random access memory (RAM), etc. The above-mentioned computer program embodiments can achieve the same or similar effects as the corresponding above-mentioned arbitrary method embodiments.

[0160] In addition, typically, the devices and equipment disclosed in the embodiments of the present invention may be various electronic terminal devices, such as mobile phones, personal digital assistants (PDAs), tablet computers (PADs), smart TVs, etc., or large terminal devices, such as servers, etc. Therefore, the protection scope disclosed in the embodiments of the present invention should not be limited to a certain type of device or equipment. The client disclosed in the embodiments of the present invention may be applied to any of the above electronic terminal devices in the form of electronic hardware, computer software, or a combination of the two.

[0161] In addition, the method disclosed in the embodiment of the present invention can also be implemented as a computer program executed by a CPU, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the CPU, the above functions defined in the method disclosed in the embodiment of the present invention are performed.

[0162] In addition, the above method steps and system units may also be implemented using a controller and a computer-readable storage medium for storing a computer program that enables the controller to implement the above steps or unit functions.

[0163] In addition, it should be understood that the computer-readable storage medium (e.g., memory) herein can be a volatile memory or a non-volatile memory, or can include both a volatile memory and a non-volatile memory. As an example and not by way of limitation, a non-volatile memory can include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), or a flash memory. A volatile memory can include a random access memory (RAM), which can act as an external cache memory. As an example and not by way of limitation, RAM can be obtained in a variety of forms, such as synchronous RAM (DRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct Rambus RAM (DRRAM). The storage devices of the disclosed aspects are intended to include, but are not limited to, these and other suitable types of memory.

[0164] The above are exemplary embodiments disclosed in the present invention, but it should be noted that various changes and modifications may be made without departing from the scope disclosed in the embodiments of the present invention as defined in the claims. The functions, steps and / or actions of the method claims according to the disclosed embodiments described herein do not need to be performed in any marked order. In addition, although the elements disclosed in the embodiments of the present invention may be described or required in individual form, they may also be understood as multiple unless explicitly limited to the singular.

[0165] In this specification, the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the embodiments described later, the description is relatively simple, and the relevant parts can be referred to the partial description of the previous embodiments.

[0166] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. An energy optimization scheduling method based on data analysis, characterized in that: The method comprises the following steps: Step S1: Real-time collection of power generation data, load demand data and meteorological data of solar energy, wind energy and energy storage systems; Step S2: using data analysis technology to process the collected data, identifying the power generation mode of renewable energy and the change trend of load demand; using the integrated management system of the virtual power plant to integrate the real-time power generation data of different power generation sources, using the K-means algorithm to identify different power generation modes; using the ARIMA model to identify the change trend of load demand; Step S3: Based on the results of the above data analysis, the uncertainty of wind and solar power output and the joint constraints of information gap decision are introduced to take system reliability R el Maximum,scheduling flexibility F le The upper model takes the best and the least carbon emission E as the main goal, and the lower model takes the least scheduling cost as the secondary goal to construct a two-layer energy scheduling model; the objective function Z of the upper model up , expressed as: Among them, U t / (U t +Δt) is the availability of the energy dispatch system, U t is the system available time, U t +Δt is the total time; S D / T D is the load satisfaction rate, S D is the total load to be satisfied, T D is the total load demand; R eact (t) is the scheduling response time, C apc max -C apc min Adjusting capabilities for scheduling; is the total carbon emissions, and P are the unit carbon emissions of each power generation unit and its corresponding power generation P; γ, γ1, γ2, σ is the preset weight of the corresponding parameter, which is used to balance the importance of each goal; Step S4: using the non-dominated sorting genetic algorithm NSGA-II combined with the optimization method of crowding distance and elite strategy to solve the above energy scheduling model, output the solution set on the Pareto frontier, and obtain the optimal scheduling strategy; Step S5: Deploy an energy scheduling plan based on the above optimal scheduling strategy; and dynamically adjust the scheduling strategy based on real-time data and forecast information.

2. The energy optimization scheduling method based on data analysis according to claim 1 is characterized in that: In step S1, High-precision photovoltaic power generation monitoring sensors, wind speed sensors and battery management system BMS sensors are used to collect corresponding data at a frequency of once per minute; 5G is used for real-time data transmission.

3. The energy optimization scheduling method based on data analysis according to claim 1 is characterized in that: In step S2, The collected data is preprocessed, and the partial autocorrelation function PACF is used to extract the trend, seasonality and periodicity characteristics of the data; at the same time, the characteristics of the meteorological data are extracted, including temperature, humidity and wind speed; polynomial regression analysis is used to explore the relationship between renewable energy power generation and load demand, and the key factors affecting the power generation mode and load demand changes are identified, including meteorological factors, load characteristics, power source characteristics and market factors.

4. The energy optimization scheduling method based on data analysis according to claim 3 is characterized in that: In step S2, the independent variables specifically include the following: The meteorological factors include temperature T, humidity H, wind speed W, and sunshine duration L; the load characteristics include load peak P peak , load valley value P valley , load change rate R; the power source characteristics include photovoltaic power generation capacity P PV , Wind power generation capacity P Wind , energy storage efficiency E; the market factors include electricity price P price , demand response D; Based on the above independent variables, a second-order polynomial regression model is constructed to predict the load demand Y, specifically: Among them, β0 is the intercept term, X i and X j is the above independent variable, β i is the corresponding linear regression coefficient, β ij is the regression coefficient of the interaction term; Using the indicator R 2 =1-RSS / TSS to evaluate the data fitting effect of the above model, R 2 The closer it is to 1, the better the effect of predicting load demand is; where RSS represents the sum of squares of the differences between the observed values ​​and the model predicted values, and TSS represents the total variation of the dependent variable.

5. The energy optimization scheduling method based on data analysis according to claim 4 is characterized in that: In step S3, the target Z of the lower model down , expressed as: Among them, C PV , C load , C grid , C DG , C flex are the combination of photovoltaic power generation, load power, grid power, fuel costs of power generation facilities to meet power demand, and the cost of flexible scheduling; N represents the duration of the forecast horizon at a given time t, t+N is the cumulative sum of the duration, and P pv (t) is the photovoltaic power in time t, c pv is the constant cost of photovoltaic power generation; c grid (t) is the cost of purchasing from the grid and selling to the grid, P purch (t) and P sold (t) are the electricity purchased and sold by the power grid; η p and η t are the efficiencies of the pump and turbine in the power grid, P p (t) and P t (t) The power generation capacity of the pumps and turbines in the power grid respectively; is the maximum power generated by DG, F is the amount of diesel consumed per kilowatt-hour, c f is the cost per litre of fuel, c m and are the cost and quality of CO2 emission respectively; Δe k (x) and are the flexibility and corresponding cost of scheduling energy in time period k, respectively.

6. The energy optimization scheduling method based on data analysis according to claim 5 is characterized in that: The constraints of the two-layer energy dispatch model include active power balance constraints: P PV +P P +P t +P DG =P L +P V Among them, P PV , P P , P t , P DG are the photovoltaic power generation, the power generation of the pump and turbine in the power grid, and the power generated by DG; P L and P V They are load power and grid power sales power respectively; Power generation capacity constraints: 0≤P WG (t)≤P WG,max ;0≤P VG (t)≤P VG,max Among them, P WG (t) and P VG (t) represent the actual output of wind power and photovoltaic power generation respectively; P WG,max and P VG,max denote the predicted output power of wind and photovoltaic respectively; Energy conversion equipment constraints: 0≤P i (t)≤P i max ;|(P i (t+1)-P i (t))|≤ΔP i Among them, P i (t) represents the input power of the i-type energy conversion device; P i max Indicates the maximum capacity limit of type i equipment; ΔP i represents the climbing rate constraint of the i-type device; Energy storage constraints: E t+1 =E t +n charge ·P charge -P discharge / or discharge Among them, E t is the energy storage state at time t; η charge and η discharge are charging and discharging efficiency respectively; P charge and P discharge are the charging and discharging power of the energy storage system, respectively; Carbon emission constraints: Among them, E i is the carbon emission coefficient of the i-th energy source; P i is the power generation of the ith energy source; is the maximum carbon emission allowed; Information gap parameter constraints: in, Represents the actual objective function value Z up Estimation of objective function value based on uncertainty information ξ is the information gap between the two; ξ is the set information gap threshold, which indicates the risk range that the decision maker can tolerate; Uncertainty constraints on wind and solar power output: S D -ι≤P wind +P solar ≤P max Among them, P wind and P solar are wind power and solar power generation respectively; S D is the load requirement to be met, ι is the safety factor; P max For maximum power generation capacity.

7. The energy optimization scheduling method based on data analysis according to claim 1 is characterized in that: In step S4, the specific process of optimization solution is as follows: S41. Set the size N of the initial population; each individual represents a scheduling strategy, encode the individual, and the individual includes the scheduling amount of solar energy, wind energy, and energy storage; randomly generate individuals in the initial population to ensure that the entire feasible solution space is covered; S42. Calculate the fitness of each individual according to the objective function defined in step S3; S43. Perform non-dominated sorting on the population and determine the dominance level of each individual; S44. For each individual in each level, calculate its crowding distance; when selecting individuals, first sort them according to the non-dominated level, and then sort them according to the crowding distance in the same level; S45. Roulette selection is used to give priority to individuals with low non-dominated levels and large crowding distances to ensure the retention of excellent individuals; S46. Use uniform crossover to generate new individuals; mutate the newly generated individuals; S47. In each generation, the top 10% of excellent individuals are retained and directly passed on to the next generation to ensure the inheritance of excellent genes; S48. Set the maximum number of iterations; monitor the change of population fitness, and terminate the algorithm when the fitness change is less than the set threshold; S49. Finally, the solution set on the Pareto frontier is output, representing the optimal scheduling strategy.

8. The energy optimization scheduling method based on data analysis according to claim 7 is characterized in that: In step S4, A sharing mechanism is introduced in the fitness calculation process to reduce competition between similar individuals and encourage the population to explore different solutions. At the same time, parallel computing is used to reduce the time complexity of the algorithm to O(nlogn).

9. The energy optimization scheduling method based on data analysis according to claim 1 is characterized in that: The optimization scheduling method also includes: Step S6: Evaluate the deployed energy scheduling scheme, compare the actual operation data with the predicted data, and verify the effectiveness and reliability of the scheduling strategy; Step S7: Establish a feedback mechanism to feed back the actual operation results to the data analysis system; Step S8: Conduct multi-scenario simulation, consider different load demands, meteorological conditions and equipment failures, and formulate corresponding emergency plans; Step S9: Provide visualized scheduling results and real-time data, allowing users to interact and make manual adjustments.

10. An energy optimization scheduling system based on data analysis, used to implement an energy optimization scheduling method based on data analysis as claimed in any one of claims 1 to 9, characterized in that: The system comprises: Data acquisition module, used to collect real-time power generation data, load demand data and meteorological data of solar energy, wind energy and energy storage systems; A data analysis module, used to process the collected data using data analysis technology to identify the power generation mode of renewable energy and the change trend of load demand; The dispatch model building module is used to build a two-layer energy dispatch model based on data analysis results, introduce uncertainty considerations of wind and solar power output and information gap decision theory, and achieve maximum system reliability, optimal dispatch flexibility, and minimum carbon emissions as the main goals, and minimum dispatch cost as the secondary goal; The optimization solution module uses the non-dominated sorting genetic algorithm NSGA-II combined with the optimization method of crowding distance and elite strategy to solve the above energy scheduling model, output the solution set on the Pareto frontier, and obtain the optimal scheduling strategy; The scheduling execution module deploys the energy scheduling plan according to the optimal scheduling strategy and dynamically adjusts the scheduling strategy according to real-time data and forecast information.

Images