Energy system planning and scheduling method based on machine learning
Through the division of energy areas and data analysis, the operation effect and impact values of renewable and non-renewable energy are calculated, and the problem of failure to fully analyze non-renewable energy data in the existing technology is solved, and more accurate energy scheduling and optimization are achieved , improves the management efficiency and sustainability of energy systems.
Patent Information
- Application Number
- CN202410880478.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-02
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing technology fails to fully analyze the relevant data of non-renewable energy, resulting in inaccuracy of energy scheduling results and instability of system operation, and is unable to improve energy utilization efficiency and optimize scheduling.
By dividing the energy area into a renewable energy sub-region and a non-renewable energy sub-region, the operation data of each area is collected and analyzed, the operation effect value and impact value are calculated, and the comparison and judgment are made to realize the planning and management of energy scheduling needs.
It realizes more accurate energy system state analysis and optimization, improves the intelligence and sustainability of energy scheduling, and enhances the reliability of scheduling management.
Smart Images

Figure CN119005554A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of energy scheduling, and in particular to an energy system planning and scheduling method based on machine learning. Background Art
[0002] Due to the increasing complexity of energy systems and the rapid changes in supply and demand, traditional energy dispatching systems are unable to efficiently manage existing energy dispatching situations. Therefore, it is necessary to reasonably dispatch relevant data of renewable and non-renewable energy in energy system dispatching to maximize energy benefits, effectively reduce energy costs, and improve energy reliability. Energy dispatching is not only a means of efficient management of energy systems, but also a way of efficient management to adapt to the needs of different fields for efficient, clean and sustainable energy. Therefore, real-time monitoring and collection of dispatching data of renewable and non-renewable energy sources is conducive to improving the accuracy of energy system dispatching and management efficiency.
[0003] For example, a patent application with announcement number: CN110807588B discloses a scheduling optimization method for a multi-energy coupling integrated energy system, including the following steps: S1: establishing a model of renewable energy, energy coupling equipment and energy storage equipment; S2: establishing a day-ahead scheduling model; S3: substituting the forecast results of the electricity, cooling, heating and gas loads for the next 24 hours into the model of step S2, and solving it in the Cplex optimizer; S4: establishing an intraday rolling optimization scheduling model; S5: substituting the real-time electricity, cooling, heating and gas loads into the model described in the above step S4, and solving it in the Cplex optimizer. The above application utilizes multi-time scale scheduling within the day-ahead to improve the accuracy of the results, thereby reducing the operating cost of the system and ensuring the stability and economic operation of the system.
[0004] For example, a patent application with announcement number: CN116562575B relates to the field of computer technology, and in particular to a method for optimizing the scheduling of an integrated energy system. According to the method of the above application, an initial model of the integrated energy system is established; historical photovoltaic data and historical wind speed data of the initial model of the integrated energy system are predicted respectively, and prediction error analysis is performed on the predicted data; based on historical photovoltaic data and its predicted data, historical wind speed data and its predicted data, with the lowest daily operating cost as the objective function, scheduling models of the integrated energy system in off-grid and grid-connected modes are established respectively; the scheduling models of the integrated energy system in off-grid and grid-connected modes are converted into mixed integer linear models by using a piecewise linearization method, and the mixed integer linear model is solved by using a CPLEX solver to achieve optimal scheduling of the integrated energy system.
[0005] However, in the process of implementing the embodiments of the present application, it was found that the above-mentioned technology has at least the following technical problems: the above-mentioned application did not conduct sufficient comparative analysis on the relevant data of non-renewable energy, such as fuel consumption, fuel cost, wastewater purification rate and waste recycling rate, which resulted in the inability to improve the accuracy of the energy scheduling results, thereby affecting the stable and timely adjustment and improvement of the system operation status; at the same time, when establishing the initial model of the comprehensive energy system, the above-mentioned application did not conduct real-time monitoring, collection and analysis of the energy scheduling data, such as the average solar radiation intensity, the average tidal height, and the average wind intensity, so that the final analysis results cannot improve the energy utilization efficiency, and cannot achieve the optimal scheduling of the energy system. Summary of the invention
[0006] In view of the shortcomings of the prior art, the present invention provides an energy system planning and scheduling method based on machine learning, which can effectively solve the problems involved in the above-mentioned background technology.
[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for planning and scheduling an energy system based on machine learning, comprising: S1. dividing an energy area into a renewable energy sub-area and a non-renewable energy sub-area; S2. collecting operation-related data of the renewable energy sub-area, and calculating the operation effect value of the renewable energy sub-area, wherein the operation effect value of the renewable energy sub-area is used to determine the operation-related data of the renewable energy sub-area; S3. collecting operation-related data of the non-renewable energy sub-area, and analyzing the operation impact value of the non-renewable energy sub-area, wherein the operation impact value of the non-renewable energy sub-area is used to determine the operation-related data of the non-renewable energy sub-area; S4. comparing the operation effect value of the renewable energy sub-area with a preset operation effect threshold, and at the same time comparing the operation impact value of the non-renewable energy sub-area with a preset operation impact value threshold, thereby planning and scheduling the energy scheduling demand area.
[0008] As a further method, the operation effect value of the renewable energy sub-area is compared with a preset operation effect threshold value, and the specific analysis process is as follows:
[0009] The operating effect value of the renewable energy sub-area is compared with the preset operating effect threshold. If the operating effect value of the renewable energy sub-area is lower than the operating effect threshold, the renewable energy sub-area is recorded as an energy scheduling demand area, and the energy scheduling demand area is planned and scheduled.
[0010] As a further method, the operation impact value of the non-renewable energy sub-area is compared with a preset operation impact value threshold, and the specific analysis process is as follows:
[0011] The operation impact value of the non-renewable energy sub-area is compared with the preset operation impact value threshold. If the operation impact value of the non-renewable energy sub-area is higher than the operation impact threshold, the non-renewable energy sub-area is recorded as an energy scheduling demand area, and the energy scheduling demand area is planned and scheduled.
[0012] As a further method, the operating effect value of the renewable energy sub-area represents data evaluating the degree of operating effect of the renewable energy sub-area through comprehensive analysis of operating-related data of the renewable energy sub-area, wherein the operating-related data of the renewable energy sub-area includes energy supply and demand data, meteorological data, and energy equipment operating data.
[0013] As a further method, the operation impact value of the non-renewable energy sub-region represents data evaluating the operation impact degree of the non-renewable energy sub-region through comprehensive analysis of operation-related data of the non-renewable energy sub-region, wherein the operation-related data of the non-renewable energy sub-region includes fuel consumption data and environmental data.
[0014] As a further method, the operation effect value of the renewable energy sub-area, the specific data acquisition and analysis process is:
[0015] The types of renewable energy in the statistical renewable energy sub-region are recorded as various types of renewable energy. At the same time, based on the energy supply and demand data of the renewable energy sub-region, the average daily output and average daily consumption of various types of renewable energy in the set energy scheduling cycle are obtained; the average daily output surplus of various types of renewable energy in the energy scheduling cycle is obtained through difference processing, and the energy consumption proportion of various types of renewable energy in the energy scheduling cycle is obtained, thereby evaluating the energy supply and demand efficiency index corresponding to various types of renewable energy in the renewable energy sub-region; according to the meteorological data of the renewable energy sub-region, based on the average solar radiation intensity, average tidal height, average wind intensity, average temperature and average humidity in the renewable energy sub-region during the energy scheduling cycle, The meteorological benefit index of the renewable energy sub-area is evaluated thereby; statistics are collected on various energy equipment in the renewable energy sub-area, and the performance parameters of each energy equipment within the energy scheduling cycle are obtained based on the operation data of the energy equipment in the renewable energy sub-area, wherein the performance parameters include the average power factor of the generator, the no-load loss of the generator, the average speed of the turbine, and the average energy storage efficiency of the battery, thereby evaluating the performance benefit index corresponding to each energy equipment in the renewable energy sub-area; the energy supply and demand benefit index corresponding to various types of renewable energy in the renewable energy sub-area, the meteorological benefit index of the renewable energy sub-area, and the performance benefit index corresponding to each energy equipment in the renewable energy sub-area are comprehensively compared to determine the operation effect value of the renewable energy sub-area.
[0016] As a further method, the operation impact value of the non-renewable energy sub-area, the specific data acquisition and analysis process is:
[0017] The fuel types in the non-renewable energy sub-area are recorded as various types of fuels. At the same time, based on the fuel consumption data of the non-renewable energy sub-area, the power generation and consumption of each type of fuel in the energy scheduling cycle are obtained, and the fuel utilization rate of each type of fuel in the energy scheduling cycle is obtained through numerical analysis and processing; at the same time, the fuel loss degree and fuel cost of each type of fuel in the energy scheduling cycle are obtained, and the fuel consumption assessment factors corresponding to each type of fuel in the non-renewable energy sub-area are analyzed; based on the environmental data of the non-renewable energy sub-area, the wastewater purification rate and waste recycling rate of the non-renewable energy sub-area in the energy scheduling cycle are obtained, and the environmental treatment assessment factors of the non-renewable energy sub-area are obtained through data analysis and processing; the fuel consumption assessment factors corresponding to each type of fuel in the non-renewable energy sub-area and the environmental treatment assessment factors of the non-renewable energy sub-area are comprehensively compared, and the operation impact value of the non-renewable energy sub-area is analyzed.
[0018] As a further method, the operating effect value of the renewable energy sub-area is calculated by the following formula: Where Ψ is the operating effect value of the renewable energy sub-area, ε j , and τ r are the energy supply and demand benefit index and meteorological benefit index corresponding to the j-th renewable energy in the renewable energy sub-area, and the performance benefit index corresponding to the r-th energy equipment. c1, c2 and c3 are the weight factors corresponding to the set energy supply and demand benefit index, meteorological benefit index and performance benefit index, respectively. j is the number of each type of renewable energy, j=1,2,...,n, n is the type of renewable energy. r is the number of each type of energy equipment, r=1,2,...,q, q is the number of energy equipment. e is a natural constant.
[0019] As a further method, the operation impact value of the non-renewable energy sub-area is calculated by the following formula: Among them, ο is the operation impact value of the non-renewable energy sub-area, β1 and β2 are the weights corresponding to the set fuel consumption assessment factor and environmental treatment assessment factor, respectively, and σ i and κ are the fuel consumption assessment factor and environmental treatment assessment factor corresponding to the i-th fuel in the non-renewable energy sub-area, respectively. i is the number of each type of fuel, i = 1, 2, ..., u, and u represents the fuel type.
[0020] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:
[0021] (1) The present invention provides an energy system planning and scheduling method based on machine learning. First, the energy area is divided into renewable energy sub-areas and non-renewable energy sub-areas, and the energy supply and demand data, meteorological data and energy equipment operation data of the renewable energy sub-areas are analyzed, and then the operation effect value of the renewable energy sub-areas is analyzed. At the same time, the fuel consumption data and environmental data of the non-renewable energy sub-areas are analyzed, and then the operation impact value of the non-renewable energy sub-areas is analyzed. In this way, the energy scheduling demand area is planned and scheduled for management, which is not only conducive to more accurate analysis and optimization of the operation status of the energy system, but also conducive to achieving more intelligent and sustainable energy scheduling management.
[0022] (2) The present invention analyzes the operating effect values of renewable energy sub-regions, and sequentially analyzes energy supply and demand data, meteorological data, and energy equipment operating data to obtain corresponding benefit indexes. The corresponding benefit index can be used to determine the energy operating effect of the item, which is beneficial for providing a reliable data basis for subsequent planning and scheduling. At the same time, it can comprehensively evaluate the equipment performance of renewable energy sub-regions, which is helpful for optimizing the scheduling of renewable energy in energy planning and scheduling.
[0023] (3) The present invention analyzes the operational impact values of non-renewable energy sub-areas, acquires and analyzes the fuel consumption data and environmental data respectively, and obtains corresponding evaluation factors, thereby providing data efficient evaluation for the operational scheduling of non-renewable energy sub-areas, providing a scientific basis for formulating more reasonable energy planning and scheduling strategies, and is conducive to achieving significant scheduling management in energy supply and demand scheduling, while also being able to improve the reliability of energy scheduling. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.
[0025] Figure 1 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION
[0026] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0027] Reference Figure 1As shown, the present invention provides an energy system planning and scheduling method based on machine learning, comprising:
[0028] S1. Divide the energy region into renewable energy sub-regions and non-renewable energy sub-regions.
[0029] S2. Collecting operation-related data of the renewable energy sub-region and calculating the operation effect value of the renewable energy sub-region, wherein the operation effect value of the renewable energy sub-region is used to determine the operation-related data of the renewable energy sub-region.
[0030] Specifically, the operating effect value of the renewable energy sub-region, in this embodiment, the operating effect value of the renewable energy sub-region can not only be obtained through the neural network model, wherein the specific acquisition process of the neural network model is as follows: by establishing a series of key indicator models, including but not limited to energy supply and demand benefits, meteorological benefits and equipment performance benefits, analyzing to obtain various benefit indexes, and then using the hierarchical analysis method to determine the relative weight of each key indicator, thereby comprehensively analyzing to obtain the operating effect value of the renewable energy sub-region; it can also be obtained through a more accurate calculation method, the specific calculation method is as follows: Among them, Ψ is the operating effect value of the renewable energy sub-area, the low energy supply and demand benefit index caused by insufficient energy supply, the low meteorological benefit index caused by the decline in solar and wind power generation efficiency due to bad weather, and the low performance benefit index caused by failure or performance degradation of energy equipment. These factors will affect the operating effect of the renewable energy sub-area. In order to reduce these negative impacts, energy supply and demand information, meteorological data and energy equipment-related data should be monitored and managed in real time to achieve more effective energy distribution, ensure the stability of energy supply, and promptly and efficiently solve potential failures of energy equipment to improve the performance and efficiency of energy equipment.
[0031] ε j , and τ rThey are respectively the energy supply and demand benefit index and meteorological benefit index corresponding to the j-th renewable energy in the renewable energy sub-region, and the performance benefit index corresponding to the r-th energy equipment, wherein the energy supply and demand benefit index represents the data value corresponding to the energy supply and demand benefit degree corresponding to the j-th renewable energy in the renewable energy sub-region evaluated by comprehensive analysis of the energy supply and demand data of the renewable energy sub-region; the meteorological benefit index represents the value of the meteorological benefit degree of the renewable energy sub-region obtained by comprehensive evaluation of the average solar radiation intensity, average tidal height, average wind intensity and other values in the meteorological data of the renewable energy sub-region; the performance benefit index represents the data of the performance benefit degree corresponding to the r-th energy equipment in the renewable energy sub-region obtained by comprehensive evaluation of the generator-related data, turbine-related data and battery-related data in the energy equipment operation data of the renewable energy sub-region.
[0032] c1, c2 and c3 are the weight factors corresponding to the set energy supply and demand benefit index, meteorological benefit index and performance benefit index respectively, j is the number of each type of renewable energy, j = 1, 2, ..., n, n is the type of renewable energy, r represents the number of each type of energy equipment, r = 1, 2, ..., q, q is the number of energy equipment, and e is a natural constant.
[0033] Furthermore, the operation-related data of the renewable energy sub-area include energy supply and demand data, meteorological data and energy equipment operation data, wherein the energy supply and demand data include the average daily output, the average daily consumption and the proportion of energy consumption; the meteorological data include the average solar radiation intensity, the average tidal height, the average wind intensity, the average temperature and the average humidity; the energy equipment operation data include the average power factor of the generator, the no-load loss of the generator, the average speed of the turbine and the average battery energy storage efficiency.
[0034] Specifically, the energy supply and demand benefit index, meteorological benefit index corresponding to the j-th renewable energy in the renewable energy sub-region, and the performance benefit index corresponding to the r-th energy equipment, the specific data acquisition and analysis process is as follows:
[0035] The types of renewable energy in the statistical renewable energy sub-region are recorded as various types of renewable energy. At the same time, according to the energy supply and demand data of the renewable energy sub-region, the average daily output and average daily consumption of various types of renewable energy in the set energy scheduling cycle are obtained. The average daily output surplus of various types of renewable energy in the energy scheduling cycle is obtained through difference processing. At the same time, the energy consumption proportion of various types of renewable energy in the energy scheduling cycle is obtained. Through the analysis formula
[0036] Get the energy supply and demand efficiency index ε corresponding to the jth type of renewable energy in the renewable energy sub-region jIn this embodiment, the energy supply and demand efficiency index ε corresponding to the j-th renewable energy in the above renewable energy sub-region is j , which can be obtained not only by the above formula, but also by the deep learning model. The historical energy supply and demand benefits are used to train the model to learn the complex relationship between various types of renewable energy and energy supply and demand benefits. The trained model is used to predict new data, so as to obtain the energy supply and demand benefit index corresponding to the jth type of renewable energy in the renewable energy sub-area. During the energy dispatch cycle, if the average daily output of renewable energy is insufficient, it may lead to insufficient energy supply and affect the stability and reliability of the energy system. If the energy consumption of renewable energy accounts for too high a proportion, it means that a large amount of renewable energy is consumed, which may cause the system to be unable to meet demand and affect the balance of energy supply and demand. If the average daily output surplus of various types of renewable energy is insufficient, it means that the energy storage demand of the system cannot be effectively met, which may lead to inflexible energy dispatch and inability to cope with peak demand. In order to reduce these negative effects, it is recommended to improve the average daily output of renewable energy by testing equipment performance and energy utilization efficiency to ensure that the energy dispatch system can meet demand, improve the energy utilization rate of the energy dispatch system, ensure sufficient energy output surplus, and improve the flexibility of the energy dispatch system.
[0037] in and Δχ j h They are respectively the average daily output, average daily consumption and average daily output surplus of the jth renewable energy in the set energy scheduling cycle, wherein the average daily output, average daily consumption and average daily output surplus are obtained by obtaining the energy output and energy consumption in the energy scheduling cycle through the energy data acquisition platform, and then performing data analysis to obtain the average daily output and average daily consumption, while the average daily output surplus can be obtained by calculating the average daily output and average daily consumption.
[0038] Y j is the energy consumption proportion of the jth type of renewable energy in the energy dispatch cycle, where the energy consumption proportion indicates the proportion of energy consumed by renewable energy in the energy dispatch cycle to the total output, which can be calculated by monitoring and recording the actual output and consumption of the energy system.
[0039] a1 and a3 are the benefit evaluation factors corresponding to the preset energy output surplus ratio and energy consumption ratio, respectively. a2 is the consumption impact factor corresponding to the predefined daily average consumption of a single renewable energy source. Y′ j is the preset reference proportion of the j-th energy consumption, where j is the number of each type of renewable energy, j=1,2,...,n, n is the type of renewable energy, and e is a natural constant.
[0040] According to the meteorological data of the renewable energy sub-region, the average solar radiation intensity, average tidal height, average wind intensity, average temperature and average humidity of the renewable energy sub-region during the energy scheduling period are analyzed by formula
[0041] Get the meteorological benefit index of renewable energy sub-region In this embodiment, the meteorological benefit index of the above-mentioned renewable energy sub-area can be obtained not only by the above-mentioned formula, but also by a deep learning model. The model is trained using historical meteorological benefits to learn the complex relationship between the energy supply and demand benefit index and the meteorological benefit corresponding to the j-th type of renewable energy in the renewable energy sub-area. The trained model is used to predict new data to obtain the meteorological benefit index of the renewable energy sub-area. Solar radiation is affected by season, weather and time. Cloudy days, nights or winter may cause the power generation efficiency of solar panels to decrease, thereby reducing solar energy production. Low tidal heights may limit the efficiency of tidal power generation equipment. Low wind intensity may cause the power generation efficiency of wind turbines to decrease. Extreme high temperatures may cause the efficiency of photovoltaic panels to decrease. Excessive humidity may cause corrosion and failure of electrical equipment. These negative effects need to be considered in the planning and scheduling of the energy system. Therefore, it is necessary to analyze the average solar radiation intensity, the average tidal height, the average wind intensity, the average temperature and the average humidity, so that the energy system can better adapt to diverse meteorological conditions, effectively reduce the adverse effects of meteorological factors, and ultimately achieve positive meteorological benefits.
[0042] S y and S f They are the average solar radiation intensity and the average wind intensity of the renewable energy sub-area during the energy scheduling cycle, respectively. The methods for obtaining the values of the average solar radiation intensity and the average wind intensity are as follows: using a solar radiation sensor to continuously measure the solar radiation intensity, and then performing time-weighted averaging of these measured values during the energy scheduling cycle to obtain the average solar radiation intensity; using a wind speed measuring device to record continuous wind speed data during the energy scheduling cycle, and then performing time-weighted averaging of these data to obtain the average wind intensity.
[0043] H, T and W are the average tidal height, average temperature and average humidity of the renewable energy sub-area during the energy scheduling cycle, respectively. The methods for obtaining the values of the average tidal height, average temperature and average humidity are as follows: using a tidal sensor to regularly measure the tidal height, measuring the tidal height at regular time intervals during the energy scheduling cycle, and then performing time-weighted averaging to obtain the average tidal height; using a temperature sensor to perform continuous temperature measurement, measuring the temperature regularly during the energy scheduling cycle, and performing time-weighted averaging to obtain the temperature average value; using a humidity sensor to perform continuous humidity measurement, measuring the humidity regularly during the energy scheduling cycle, and performing time-weighted averaging to obtain the humidity average value.
[0044] d1, d3d4 and d5 represent the benefit evaluation factors corresponding to the predefined average solar radiation intensity, average wind intensity, average temperature and average humidity respectively. d2 represents the benefit impact factor corresponding to the predefined average unit tidal height. S y ′、S f ′, T′ and W′ represent the preset solar radiation intensity limit value, wind intensity limit value, temperature reference value and humidity reference value.
[0045] The energy devices in the renewable energy sub-area are counted, wherein the energy devices include but are not limited to generators, turbines, and batteries.
[0046] At the same time, according to the energy equipment operation data of the renewable energy sub-area, the performance parameters of each energy equipment in the energy dispatch cycle are obtained, where the performance parameters include the average power factor of the generator, the no-load loss of the generator, the average speed of the turbine and the average energy storage efficiency of the battery.
[0047] Get the performance benefit index τ corresponding to the rth energy device in the renewable energy sub-area rIn this embodiment, the performance benefit index corresponding to the rth energy device in the above renewable energy sub-area can be obtained not only by the above formula, but also by a deep learning model. The model is trained using the average power factor of the generator, the no-load loss of the generator, the average speed of the turbine, and the average energy storage efficiency of the battery to learn the complex relationship between the performance parameters of each energy device in the energy scheduling cycle and the performance benefit index. The trained model is used to predict new data to obtain the performance benefit index corresponding to the rth energy device in the renewable energy sub-area. A low average power factor may cause the energy device to lose more electricity when generating electricity, affecting the system efficiency; a high no-load loss may cause the energy device to still consume energy in a non-working state, reducing the overall efficiency; an excessively high or low turbine speed may cause unstable energy output, affecting the reliability of the system; low energy storage efficiency may cause more energy loss during energy storage and release. Therefore, it is necessary to ensure the appropriate average power factor of the generator, the no-load loss of the generator, the average speed of the turbine, and the average energy storage efficiency of the battery to reduce the negative impact of the performance parameters of the energy device on the performance benefit of the entire system and improve the overall performance and operation efficiency of the energy system.
[0048] Where P r g is the average power factor corresponding to the generator of the rth energy device within the energy scheduling period, where the average power factor of the generator refers to the proportion of the average useful power in the average apparent power, and the average apparent power refers to the average total power output by the generator.
[0049] P r s is the no-load loss of the generator of the rth energy device during the energy scheduling cycle. The no-load loss refers to the power consumed by the generator when it is running without generating useful power. When the generator is not connected to any load, it still requires a certain amount of energy to maintain its operation, including overcoming internal electromagnetic resistance and friction resistance, etc. This part of the consumed power is the no-load loss.
[0050] V r is the average speed of the turbine of the rth energy device during the energy scheduling cycle, where the average speed of the turbine can be obtained by real-time monitoring using a sensor. The sensor is installed on the turbine to measure the speed of the rotor, and the average speed is obtained through continuous measurement during the scheduling cycle.
[0051] η rIt is the average battery energy storage efficiency of the rth energy device during the energy scheduling cycle, where the energy storage efficiency refers to the efficiency of the battery's electric energy conversion during charging and discharging. It represents the ratio of the actual available electric energy extracted from the battery to the electric energy stored in the battery. The value of the energy storage efficiency is usually expressed as a percentage, which indicates the degree of loss of the battery during the energy conversion process.
[0052] Φ1, Φ2, Φ3 and Φ4 are the benefit evaluation factors corresponding to the predefined average battery energy storage efficiency, the average power factor of the generator, the no-load loss of the generator and the average speed of the turbine, respectively. η′, P g ′、P s ′ and V′ are respectively the preset battery energy storage efficiency limit, generator power factor limit, generator no-load loss limit and turbine reference speed, r is the number of each energy device, r=1,2,...,q, q is the number of energy devices.
[0053] In a specific embodiment, the present invention analyzes the operating effect value of the renewable energy sub-area, and analyzes the energy supply and demand data, meteorological data, and energy equipment operation data in turn to obtain the corresponding benefit index. The energy operation effect corresponding to the item can be determined by the corresponding benefit index, which is beneficial to provide a reliable data basis for subsequent planning and scheduling. At the same time, it can comprehensively evaluate the equipment performance of the renewable energy sub-area, which is helpful to optimize the scheduling of renewable energy in energy planning and scheduling.
[0054] S3. Collecting operation-related data of the non-renewable energy sub-region and analyzing the operation impact value of the non-renewable energy sub-region, wherein the operation impact value of the non-renewable energy sub-region is used to determine the operation-related data of the non-renewable energy sub-region.
[0055] Specifically, the operation impact value of the non-renewable energy sub-region, in this embodiment, the operation impact value of the non-renewable energy sub-region can not only be obtained through the principal component analysis model, wherein the specific acquisition process of the principal component analysis model is as follows: the weights corresponding to the fuel consumption assessment factor and the environmental treatment assessment factor are reduced and features are extracted by the PCA technology, and the fuel consumption assessment factor and the environmental treatment assessment factor corresponding to the i-th fuel in the non-renewable energy sub-region are obtained by life cycle assessment, and the operation impact value of the non-renewable energy sub-region is obtained by comprehensive analysis; it can also be obtained by a more accurate calculation method, and the specific calculation method is as follows:
[0056] Among them, ο is the operational impact value of the non-renewable energy sub-region, which refers to the measurement or evaluation value of the negative impact of non-renewable energy projects on the environment and society during operation in a specific region. An unfavorable evaluation factor value may indicate higher fuel consumption and environmental burden, thereby generating an unfavorable operational impact value; therefore, by analyzing the weight distribution of the fuel consumption evaluation factor and the environmental treatment evaluation factor, the system can better adapt to changes in the external environment, improve its ability to cope with complex situations and reduce its excessive sensitivity to uncertain data.
[0057] σ i and κ are respectively the fuel consumption assessment factor and the environmental treatment assessment factor corresponding to the i-th fuel in the non-renewable energy sub-region, wherein the fuel consumption assessment factor represents the data value corresponding to the fuel consumption assessment degree corresponding to the i-th fuel in the non-renewable energy sub-region evaluated by comprehensive analysis of the fuel utilization rate, fuel loss rate and fuel cost in the fuel consumption data of the non-renewable energy sub-region; the environmental treatment assessment factor represents the data value corresponding to the environmental treatment assessment degree corresponding to the i-th fuel in the non-renewable energy sub-region evaluated by comprehensive analysis of the wastewater purification rate and waste recycling rate in the environmental data of the non-renewable energy sub-region.
[0058] β1 and β2 are weights corresponding to the set fuel consumption assessment factor and environmental treatment assessment factor, respectively. i is the number of each type of fuel, i=1,2,...,u, where u represents the fuel type.
[0059] Furthermore, the operation-related data of the non-renewable energy sub-area include fuel consumption data and environmental data, wherein the fuel consumption data includes power generation, consumption, fuel loss and fuel cost, and the environmental data includes wastewater purification rate and waste recycling rate.
[0060] Specifically, the fuel consumption assessment factor and environmental treatment assessment factor corresponding to the i-th fuel in the non-renewable energy sub-area, the specific data acquisition and analysis process is:
[0061] The fuel types in the statistical non-renewable energy sub-area are recorded as various fuels. It should be explained that various fuels include but are not limited to oil, natural gas, coal and biomass waste.
[0062] At the same time, according to the fuel consumption data of non-renewable energy sub-areas, the power generation and consumption of various fuels in the energy scheduling cycle are obtained, and the fuel utilization rate of various fuels in the energy scheduling cycle is obtained through numerical analysis and processing.
[0063] At the same time, the fuel loss degree and fuel cost of each type of fuel in the energy scheduling cycle are obtained, thereby calculating the fuel consumption assessment factors corresponding to each type of fuel in the non-renewable energy sub-area. In this embodiment, the fuel consumption assessment factors corresponding to each type of fuel in the non-renewable energy sub-area can be obtained not only through the life cycle assessment method, wherein the specific process of the life cycle assessment method is as follows: clearly define the system boundary, determine the functional unit, and collect key life cycle data, including fuel utilization, fuel loss degree and fuel cost, etc., by establishing a comprehensive life cycle model, calculating the energy consumption of each stage, and applying standardization and normalization technology to obtain the fuel consumption assessment factor corresponding to the i-th type of fuel in the non-renewable energy sub-area; it can also be obtained through a more accurate calculation method, the specific calculation method is as follows: where σ i is the fuel consumption assessment factor corresponding to the i-th fuel in the non-renewable energy sub-area. Low effective utilization of fuel during the energy scheduling cycle may lead to low fuel utilization; higher loss may lead to lower efficiency of the energy system; exceeding the fuel loss limit may lead to unsatisfactory system performance; exceeding the fuel limit cost may lead to poor economic benefits. Therefore, in order to reduce these negative impacts, it is necessary to analyze the fuel utilization, fuel loss and fuel cost to keep them within a reasonable range to extend the service life of energy resources and reduce the total cost of energy production.
[0064] ω i is the fuel utilization rate of the i-th fuel in the energy scheduling cycle, where the fuel utilization rate refers to the efficiency of the energy system in utilizing a certain type of fuel in the energy scheduling cycle. It means that in the energy conversion process, the input fuel is effectively converted into usable energy output.
[0065] θ i s and θ i c They are respectively the fuel loss rate and fuel cost of the i-th fuel in the energy scheduling cycle, and the methods of obtaining the fuel loss rate and fuel cost are: monitor and record the input and output of fuel in the energy system, calculate the energy loss during the conversion process, and obtain the fuel loss rate through the ratio of the loss to the input fuel; record the relevant costs of fuel in the energy system, including procurement cost, transportation cost, storage cost, etc., and calculate the total fuel cost to obtain the fuel cost.
[0066] θ s ′ and θ c′ is the set fuel consumption limit value and fuel limit cost respectively, α1 is the utilization impact value corresponding to the set unit fuel utilization rate, α2 and α3 are the correction factors corresponding to the set fuel loss degree and fuel cost respectively, i is the number of each type of fuel, i=1,2,...,u, u is the fuel type.
[0067] Based on the environmental data of the non-renewable energy sub-region, the wastewater purification rate and waste material recycling rate of the non-renewable energy sub-region within the energy scheduling cycle are obtained, and the environmental treatment assessment factor of the non-renewable energy sub-region is obtained through data analysis and processing. In this embodiment, the environmental treatment assessment factor of the non-renewable energy sub-region can be obtained not only through the material flow analysis model, wherein the specific acquisition process of the material flow analysis model is as follows: a system dynamic model of wastewater purification and waste recycling is established, different technologies and efficiencies are considered, parameter sensitivity analysis is performed, and finally a comprehensive environmental treatment assessment factor of the non-renewable energy sub-region is obtained; it can also be obtained through a more accurate calculation method, and the specific calculation method is as follows: Among them, κ is the environmental treatment assessment factor of the non-renewable energy sub-area. If the wastewater purification rate and waste recycling rate are low, it may lead to environmental pollution and resource waste, and have a negative impact on the ecosystem; if the environmental treatment impact factor is low, it means that the environmental treatment effect has not reached the expected level, which may lead to an increase in the environmental burden. In order to reduce these negative impacts, it is necessary to ensure the appropriate unit wastewater purification rate and the environmental treatment impact factor corresponding to the unit waste recycling rate, so that it is within a reasonable range, so as to effectively utilize waste resources, reduce resource waste, improve the overall resource utilization efficiency of the energy system, and protect ecological diversity.
[0068] as well as They are the wastewater purification rate and waste recycling rate of the non-renewable energy sub-area during the energy scheduling cycle, and the methods of obtaining data for the wastewater purification rate and the waste recycling rate are: continuously monitoring the water quality parameters in the wastewater treatment process through an online monitoring system, analyzing the data in real time, and obtaining the wastewater purification rate; using the Internet of Things technology to track the entire process from waste generation to recycling, analyze and collect data, and obtain the waste recycling rate.
[0069] s1 and s2 are environmental treatment impact factors corresponding to the preset unit wastewater purification rate and unit waste recycling rate, where e is represented by a natural constant.
[0070] In a specific embodiment, the present invention analyzes the operating impact values of non-renewable energy sub-areas, acquires and analyzes fuel consumption data and environmental data respectively, and obtains corresponding evaluation factors, thereby providing data efficient evaluation for the operation scheduling of non-renewable energy sub-areas, providing a scientific basis for formulating more reasonable energy planning and scheduling strategies, and is conducive to achieving significant scheduling management in energy supply and demand scheduling, while being able to improve the reliability of energy scheduling.
[0071] S4. Compare the operation effect values of the renewable energy sub-areas and the operation impact values of the non-renewable energy sub-areas respectively, and thereby plan and dispatch the energy dispatch demand areas.
[0072] Specifically, the comparison of the operating effect values of the renewable energy sub-areas is performed, and the specific analysis process is as follows:
[0073] The operating effect value of the renewable energy sub-area is compared with the preset operating effect threshold. If the operating effect value of the renewable energy sub-area is lower than the operating effect threshold, in this embodiment, the operating effect value of the above-mentioned renewable energy sub-area is lower than the operating effect threshold, indicating that the energy supply is unstable and the environmental benefit is low. Insufficient renewable energy output may have an unstable impact on energy equipment, resulting in fluctuations and instability of the energy equipment system. If the stability of the resources and the feasibility of the technology in the renewable energy sub-area are not judged, the operating effect of the renewable energy will be reduced.
[0074] The renewable energy sub-area is recorded as the energy scheduling demand area, and the energy scheduling demand area is planned, scheduled and managed. In this embodiment, the above-mentioned planning, scheduling and management of the energy scheduling demand area introduces an intelligent scheduling system, uses real-time data and prediction models for precise scheduling, establishes smart grids and microgrids, implements multi-energy coordinated scheduling, and realizes efficient integration and distribution of renewable energy.
[0075] The operation impact values of the non-renewable energy sub-areas are compared, and the specific analysis process is as follows:
[0076] The operating impact value of the non-renewable energy sub-area is compared with the preset operating impact value threshold. If the operating impact value of the non-renewable energy sub-area is higher than the operating impact threshold, in this embodiment, the operating impact value of the above-mentioned non-renewable energy sub-area is higher than the operating impact threshold, indicating that the exploitation and utilization of non-renewable energy will have negative impacts on the surrounding ecosystem such as destruction of vegetation and loss of biodiversity, or may cause air pollution, water pollution or soil pollution. If a detailed environmental assessment is not carried out and it is not determined whether the exploitation of non-renewable energy complies with relevant environmental protection laws and standards, it will cause environmental pollution and ecological damage.
[0077] The non-renewable energy sub-area is recorded as an energy scheduling demand area, and the energy scheduling demand area is planned and scheduled. In this embodiment, the above-mentioned planning and scheduling management of the energy scheduling demand area adopts high-efficiency combustion equipment, such as supercritical and ultra-supercritical thermal power technology, to improve energy conversion efficiency, accurately evaluate the reserves of non-renewable energy, and formulate supply and demand balance strategies to prevent resource depletion.
[0078] In a specific embodiment, the present invention provides an energy system planning and scheduling method based on machine learning. First, the energy area is divided into renewable energy sub-areas and non-renewable energy sub-areas, and the energy supply and demand data, meteorological data and energy equipment operation data of the renewable energy sub-areas are analyzed, and then the operation effect value of the renewable energy sub-area is analyzed. At the same time, the fuel consumption data and environmental data of the non-renewable energy sub-area are analyzed, and then the operation impact value of the non-renewable energy sub-area is analyzed. In this way, the energy scheduling demand area is planned and scheduled for management, which is not only conducive to more accurate analysis and optimization of the operating status of the energy system, but also conducive to achieving more intelligent and sustainable energy scheduling management.
[0079] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.
Claims
1. A method for energy system planning and scheduling based on machine learning, characterized in that: include: S1. Divide the energy region into renewable energy sub-regions and non-renewable energy sub-regions; S2. Collecting operation-related data of renewable energy sub-regions, analyzing the operation effect values of renewable energy sub-regions, wherein the operation effect values of renewable energy sub-regions are used to determine the operation-related data of renewable energy sub-regions; S3. Collecting operation-related data of non-renewable energy sub-regions, analyzing the operation impact values of non-renewable energy sub-regions, wherein the operation impact values of non-renewable energy sub-regions are used to determine the operation-related data of non-renewable energy sub-regions; S4. Compare the operating effect value of the renewable energy sub-area with the preset operating effect threshold, and compare the operating impact value of the non-renewable energy sub-area with the preset operating impact value threshold, thereby planning and scheduling management of the energy scheduling demand area.
2. The energy system planning and scheduling method based on machine learning according to claim 1 is characterized in that: The operation effect value of the renewable energy sub-area is compared with the preset operation effect threshold value, and the specific analysis process is as follows: The operating effect value of the renewable energy sub-area is compared with the preset operating effect threshold. If the operating effect value of the renewable energy sub-area is lower than the operating effect threshold, the renewable energy sub-area is recorded as an energy scheduling demand area, and the energy scheduling demand area is planned and scheduled.
3. The energy system planning and scheduling method based on machine learning according to claim 1 is characterized in that: The operation impact value of the non-renewable energy sub-area is compared with the preset operation impact value threshold, and the specific analysis process is as follows: The operation impact value of the non-renewable energy sub-area is compared with the preset operation impact value threshold. If the operation impact value of the non-renewable energy sub-area is higher than the operation impact threshold, the non-renewable energy sub-area is recorded as an energy scheduling demand area, and the energy scheduling demand area is planned and scheduled.
4. The energy system planning and scheduling method based on machine learning according to claim 2 is characterized in that: The operation effect value of the renewable energy sub-region represents data evaluating the degree of operation effect of the renewable energy sub-region through comprehensive analysis of operation-related data of the renewable energy sub-region, wherein the operation-related data of the renewable energy sub-region includes energy supply and demand data, meteorological data and energy equipment operation data.
5. The energy system planning and scheduling method based on machine learning according to claim 3 is characterized in that: The operation impact value of the non-renewable energy sub-region represents data evaluating the operation impact degree of the non-renewable energy sub-region through comprehensive analysis of operation-related data of the non-renewable energy sub-region, wherein the operation-related data of the non-renewable energy sub-region includes fuel consumption data and environmental data.
6. The energy system planning and scheduling method based on machine learning according to claim 4 is characterized in that: The specific data acquisition and analysis process of the operating effect value of the renewable energy sub-area is as follows: The types of renewable energy in the statistical renewable energy sub-region are recorded as various types of renewable energy. At the same time, according to the energy supply and demand data of the renewable energy sub-region, the average daily output and average daily consumption of various types of renewable energy in the set energy scheduling cycle are obtained; The average daily output surplus of each type of renewable energy in the energy dispatch cycle is obtained through difference processing, and the energy consumption proportion of each type of renewable energy in the energy dispatch cycle is obtained at the same time, thereby evaluating the energy supply and demand efficiency index corresponding to each type of renewable energy in the renewable energy sub-region; According to the meteorological data of the renewable energy sub-region, the meteorological benefit index of the renewable energy sub-region is evaluated based on the average solar radiation intensity, average tidal height, average wind intensity, average temperature and average humidity of the renewable energy sub-region during the energy dispatch period; Count the energy equipment in the renewable energy sub-area, and obtain the performance parameters of each energy equipment in the energy dispatch cycle based on the energy equipment operation data in the renewable energy sub-area, where the performance parameters include the average power factor of the generator, the no-load loss of the generator, the average speed of the turbine, and the average energy storage efficiency of the battery, thereby evaluating the performance benefit index corresponding to each energy equipment in the renewable energy sub-area; By comprehensively comparing the energy supply and demand benefit indexes corresponding to various types of renewable energy in the renewable energy sub-area, the meteorological benefit index of the renewable energy sub-area, and the performance benefit index corresponding to each energy equipment in the renewable energy sub-area, the operating effect value of the renewable energy sub-area is determined.
7. The energy system planning and scheduling method based on machine learning according to claim 5 is characterized in that: The specific data acquisition and analysis process of the operation impact value of the non-renewable energy sub-area is as follows: The fuel types in the non-renewable energy sub-area are recorded as various fuels. At the same time, according to the fuel consumption data of the non-renewable energy sub-area, the power generation and consumption of various fuels in the energy scheduling cycle are obtained, and the fuel utilization rate of various fuels in the energy scheduling cycle is obtained through numerical analysis and processing; At the same time, the fuel loss and fuel cost of each type of fuel in the energy scheduling cycle are obtained, and the fuel consumption assessment factors corresponding to each type of fuel in the non-renewable energy sub-area are analyzed; Based on the environmental data of the non-renewable energy sub-region, the wastewater purification rate and waste recycling rate of the non-renewable energy sub-region within the energy dispatch cycle are obtained, and the environmental treatment assessment factor of the non-renewable energy sub-region is obtained through data analysis and processing; By comprehensively comparing the fuel consumption assessment factors corresponding to various types of fuels in the non-renewable energy sub-area and the environmental treatment assessment factors of the non-renewable energy sub-area, the operation impact value of the non-renewable energy sub-area is analyzed.
8. The energy system planning and scheduling method based on machine learning as described in claim 6 is characterized by: The specific calculation formula for the operating effect value of the renewable energy sub-area is: Where Ψ is the operating effect value of the renewable energy sub-area, εj, and τr are the energy supply and demand benefit index, meteorological benefit index and performance benefit index of the j-th renewable energy in the renewable energy sub-area, respectively; c1, c2 and c3 are the weight factors corresponding to the set energy supply and demand benefit index, meteorological benefit index and performance benefit index, respectively; j is the number of each type of renewable energy, j = 1, 2, ..., n, n is the type of renewable energy; r is the number of each type of energy equipment, r = 1, 2, ..., q, q is the number of energy equipment; and e is a natural constant.
9. The energy system planning and scheduling method based on machine learning according to claim 7 is characterized in that: The specific calculation formula for the operation impact value of the non-renewable energy sub-area is: Among them, ο is the operation impact value of the non-renewable energy sub-area, β1 and β2 are the weights corresponding to the set fuel consumption assessment factor and environmental treatment assessment factor, σi and κ are the fuel consumption assessment factor and environmental treatment assessment factor corresponding to the i-th fuel in the non-renewable energy sub-area, respectively, and i is the number of each type of fuel, i=1,2,...,u, where u represents the fuel type.
Citation Information
Cited By
Industrial energy saving management method and system based on big data and medium
CN119358971A
Informatization interconnection method and system of solar water purification terminal
CN119940889A
Solid waste recycling full life cycle environmental impact evaluation method
CN120541427A
