Comprehensive energy management system based on big data

Abstract

The invention discloses a comprehensive energy management system based on big data. The comprehensive energy management system comprises a data acquisition and processing module, an intelligent scheduling and analysis module and an efficiency evaluation module, the data acquisition and processing module is connected with the intelligent scheduling and analysis module, and the data acquisition and processing module is used for comprehensive and real-time collection of data and processing and analysis of the collected data through cloud computing. And the intelligent scheduling and analysis module is also connected with the efficiency evaluation module and is used for performing deep analysis on the comprehensively processed energy data. And the efficiency evaluation module is used for multi-dimensional analysis of photovoltaic power generation efficiency, energy storage system efficiency and energy consumption management. The core design idea of the invention is to break the barrier, ensure seamless data exchange among various energy subsystems and end the information island phenomenon through a unified data standard and an advanced communication protocol, and form a comprehensive management ecology with data sharing and resource coordination.

Description

Technical Field

[0001] The present invention belongs to the technical field of big data analysis, and specifically relates to a comprehensive energy management system based on big data. Background Art

[0002] Against the backdrop of dramatic changes in the global energy landscape, the booming development of solar energy and other green energy sources is leading a profound energy revolution. This process has not only greatly promoted the sustainable development agenda, but also posed unprecedented challenges to the modern energy management system. Traditional energy monitoring systems were not designed to fully anticipate today's needs, especially when dealing with the refined operation of solar photovoltaic power stations, the efficient coordination of energy storage systems, and the complex and changeable energy consumption management on the user side.

[0003] The market is in urgent need of an innovative integrated energy management system that has a high degree of integration and can transcend the boundaries of a single energy type or application field to achieve seamless integration of multiple data such as photovoltaic power generation output, energy storage resource allocation, and user-side energy consumption. The platform should integrate advanced Internet of Things technology, big data analysis, and artificial intelligence algorithms, which can not only collect and process massive amounts of data in real time, but also deeply explore the correlations and trends behind the data, providing a strong data-driven foundation for decision support.

[0004] The integrated energy management platform will not only be limited to data aggregation and display, but will further evolve into an incubator for smart energy solutions. It can tailor energy scheduling strategies for users based on accurate energy supply and demand forecasting models to maximize the self-sufficiency rate of renewable energy; at the same time, through detailed energy consumption analysis and optimization suggestions, it can help users effectively reduce energy costs and improve the overall efficiency of energy use. In addition, the platform should also have flexible scalability and compatibility, and be able to continuously upgrade with technological advances and changes in market demand, ensuring that users are always in a favorable position in the evolving energy ecosystem.

[0005] In existing technologies, traditional energy management often revolves around a single energy type or independent system, which makes it difficult to exchange data between systems and creates high barriers to information exchange. This "information island" phenomenon not only limits the comprehensive integration and in-depth mining of data, but also prevents managers from obtaining a unified and comprehensive view of energy supply and demand, which is not conducive to the efficient allocation and optimization of resources. Summary of the invention

[0006] The purpose of the present invention is to provide an integrated energy management system based on big data to solve the problems of insufficient integration, low intelligence level, slow response and weak decision support in the integrated energy management in the prior art mentioned in the background technology.

[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0008] A comprehensive energy management system based on big data, including a data acquisition and processing module, an intelligent scheduling and analysis module and an efficiency evaluation module; the data acquisition and processing module is connected with the intelligent scheduling and analysis module, the data acquisition and processing module is used for comprehensive and real-time data collection, and the collected data is processed and analyzed through cloud computing;

[0009] The intelligent scheduling and analysis module is also connected to the efficiency evaluation module. The intelligent scheduling and analysis module is used for monitoring and scheduling of photovoltaic power generation and energy storage, as well as in-depth analysis of the comprehensively processed energy data, accurate prediction and optimization of intelligent scheduling of energy storage resources, and fine monitoring of energy consumption behavior;

[0010] The performance evaluation module is used to conduct multi-dimensional analysis of photovoltaic power generation efficiency, energy storage system efficiency and energy management, and generate performance indicator reports.

[0011] According to the above technical solution, the data collected in the data acquisition and processing module include: data from the power generation side, the energy storage side and the energy consumption side. The power generation side includes solar panels, photovoltaic inverters, solar charging controllers, monitoring systems and meteorological data; the energy storage side includes energy storage batteries, battery management systems, energy storage inverters and energy management systems; the energy consumption side includes air conditioning, lighting, charging piles and data centers.

[0012] According to the above technical solution, data processing in the data acquisition and processing module includes: eliminating invalid and erroneous information through data cleaning to ensure data quality; and deeply integrating and standardizing the cleaned data to unify its format for easy analysis.

[0013] According to the above technical solution, the intelligent scheduling and analysis module includes photovoltaic monitoring and optimization module, intelligent scheduling of energy storage system, energy consumption monitoring and energy-saving suggestions, and centralized control and trading of virtual power plants.

[0014] According to the above technical solution, the photovoltaic monitoring and optimization module is used to monitor the light intensity and temperature of the photovoltaic power station in real time, combine historical data and real-time data analysis, predict photovoltaic power output, optimize the tilt angle and cleaning maintenance plan to maximize photovoltaic power generation efficiency;

[0015] The photovoltaic monitoring and optimization module includes real-time tracking and prediction models as well as power regulation and matching algorithms.

[0016] According to the above technical solution, the real-time tracking and prediction model includes an electric energy output prediction model, specifically:

[0017] P predicted (t) = h(L predicted (t), Tpredicted (t), H(t); Θ P )

[0018] P prcdicted (t): predicted power output (kW); h is the prediction function based on light intensity, temperature and historical power output, Θ p is the model parameter.

[0019] According to the above technical solution, the power regulation and matching algorithm includes a power regulation algorithm and a grid matching optimization algorithm, specifically:

[0020] Power regulation algorithm:

[0021] P inverter (t) = f regulate (P predicted (t), P grid-demand (t); θ R )

[0022] Among them, P grid-demand (t): grid load demand at time t (kW); P inverter (t): inverter output power at time t (kW); f regulate : Power regulation function, f cguiate Consider predicting power output and grid demand, and adjust inverter output to ensure matching with grid frequency and voltage; R To adjust the function parameters;

[0023] Grid matching optimization algorithm:

[0024]

[0025] Where P grid-demand (t): grid load demand at time t (kW); P inverter (t): inverter output power at time t (kW).

[0026] According to the above technical solution, the performance evaluation module includes photovoltaic power generation performance evaluation and energy storage system efficiency review; wherein, photovoltaic power generation performance evaluation includes a capacity efficiency analysis algorithm and an environmental impact correction algorithm;

[0027] Capacity efficiency analysis algorithm:

[0028] P theoretical (t) = A area ×I light (t) = η nominal ×α×(1-β×(T cell (t)-T ref ))

[0029]

[0030] Where, T ref is the reference temperature, P theoretical (t) represents the photovoltaic power generation efficiency; P gen (t): actual electrical power generated at time t (kW); A area :Total area of ​​photovoltaic panels (m 2 ), I light (t): light intensity at time t (kW / m 2 ), T cell (t): time t battery temperature (℃), α: light intensity influence coefficient, β: temperature coefficient, η nominal : Theoretical maximum conversion efficiency (%), η eff (t): actual conversion efficiency at time t (%);

[0031] The environmental impact correction algorithm includes a meteorological calibration model algorithm and a seasonal correction algorithm; the meteorological calibration model algorithm is specifically:

[0032] M corr (t) = f(I light_corrected (t), T cell_corrected (t), Wind_speed(t), Humidity(t))

[0033] In the formula, I light_correted (t) represents the light intensity, T cell_corrected (t) represents the battery temperature, Wind_speed(t) represents the wind speed, and Humidity(t) represents the humidity;

[0034] The specific seasonal correction algorithm is:

[0035] F season =g(Month(t), DayOfYear(t))

[0036] Where Month(t) represents the seasonal correction factor; DayOfYear(t) represents the month and the specific day of time t.

[0037] According to the above technical solution, the efficiency review of the energy storage system includes charging and discharging cycle efficiency tracking; the charging and discharging cycle efficiency tracking includes the charging and discharging efficiency calculation algorithm and the battery life prediction algorithm response speed evaluation algorithm;

[0038] The specific calculation algorithm of charge and discharge efficiency is as follows: the charge efficiency (CE) and discharge efficiency (DE) can be calculated by measuring the energy difference during the charge and discharge process; assuming the rated capacity of the battery is C nominal , the energy actually charged in this cycle is E charegd, the actual energy released during discharge is E discharged :

[0039]

[0040] Among them, V charge and V discharge are the average voltages during charging and discharging, respectively;

[0041] Battery life prediction algorithm:

[0042] Assuming that the battery life has an exponential decay relationship with its cycle number Ngde, the battery state of health (SOH) can be expressed in a simplified form of the Arrhenius equation:

[0043]

[0044] Where k is the degradation rate constant that depends on the battery type and usage conditions.

[0045] According to the above technical solution, the charge-discharge cycle efficiency tracking also includes a response speed evaluation algorithm;

[0046] Response speed evaluation algorithm:

[0047] T response，start =τ sys +τ inv

[0048] Among them, T sys Represents the processing time of the system control layer, including communication and instruction parsing.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] The core design idea of ​​this invention is to break down barriers, ensure seamless data exchange between various energy subsystems through unified data standards and advanced communication protocols, end the phenomenon of information islands, and form a comprehensive management ecology of data sharing and resource coordination. On this basis, big data analysis technology is used to deeply explore energy supply and demand patterns, identify potential efficiency bottlenecks and waste points, and lay a solid foundation for refined management.

[0051] What is particularly important is that the present invention introduces the powerful force of artificial intelligence and big models. Through machine learning algorithms, it conducts deep learning of historical data and real-time information. It can not only automatically predict energy demand trends and optimize scheduling strategies, but also play an important role in anomaly detection and fault warning, significantly improving emergency response speed and processing efficiency, thereby enhancing the flexibility and reliability of the entire energy system. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a system structure block diagram of the present invention;

[0053] Figure 2 A flow chart is provided for use with the system of the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] Embodiment 1

[0056] like Figure 1 As shown, a comprehensive energy management system based on big data includes a data acquisition and processing module, an intelligent scheduling and analysis module, and an efficiency evaluation module; the data acquisition and processing module is connected to the intelligent scheduling and analysis module, the data acquisition and processing module is used for comprehensive and real-time data collection, and the collected data is processed and analyzed through cloud computing.

[0057] The intelligent scheduling and analysis module is also connected to the performance evaluation module. The intelligent scheduling and analysis module is used to conduct in-depth analysis of the comprehensively processed energy data, accurately predict and optimize the intelligent scheduling of energy storage resources, and finely monitor energy consumption behavior.

[0058] The performance evaluation module is used to conduct multi-dimensional analysis of photovoltaic power generation efficiency, energy storage system efficiency and energy management, and generate performance indicator reports.

[0059] The core design idea of ​​this invention is to break down barriers, ensure seamless data exchange between various energy subsystems through unified data standards and advanced communication protocols, end the phenomenon of information islands, and form a comprehensive management ecology of data sharing and resource coordination. On this basis, big data analysis technology is used to deeply explore energy supply and demand patterns, identify potential efficiency bottlenecks and waste points, and lay a solid foundation for refined management.

[0060] What is particularly important is that the present invention introduces the powerful force of artificial intelligence and big models. Through machine learning algorithms, it conducts deep learning of historical data and real-time information. It can not only automatically predict energy demand trends and optimize scheduling strategies, but also play an important role in anomaly detection and fault warning, significantly improving emergency response speed and processing efficiency, thereby enhancing the flexibility and reliability of the entire energy system.

[0061] Embodiment 2

[0062] This embodiment is a further refinement of the first embodiment.

[0063] The data acquisition and processing module includes: Comprehensive data acquisition: The acquisition covers the power generation side (solar panels, photovoltaic inverters, solar charging controllers, monitoring systems, meteorological data), energy storage side (energy storage batteries, battery management systems, energy storage inverters, energy management systems), and energy consumption side (air conditioning, lighting, charging piles, data centers, etc.). This module integrates the mainstream sensor technology and multiple communication protocols on the market (such as MQTT, Modbus, IEC103 / 104, SunSpec, Can Bus and other protocols), and can achieve seamless automatic acquisition of multi-source energy data across different energy types. Whether it is the real-time power generation data of distributed photovoltaic power stations, or the industrial and commercial energy storage data and the power consumption data of industrial energy equipment, they can all be accurately captured and uploaded to the cloud big data platform in real time, laying a solid data foundation for subsequent analysis and decision-making.

[0064] High-performance big data processing platform: Built on a powerful cloud computing architecture, this module has excellent capabilities for processing large-scale concurrent data and is the heart of the integrated energy management system. It can not only efficiently process the massive amount of raw data received from the data acquisition module, but also eliminate invalid and erroneous information through advanced data cleaning technology to ensure data quality; it can also deeply integrate and standardize these purified data to make them in a unified format for easy analysis. The platform uses advanced data storage and management systems to ensure data security and access efficiency, and provides a solid infrastructure for the long-term accumulation and in-depth mining of energy data. On this basis, the platform further empowers the data analysis module, and through advanced analysis tools such as machine learning algorithms, it reveals energy consumption patterns, predicts future trends, and provides a scientific basis for intelligent decision-making in integrated energy management.

[0065] The intelligent scheduling and analysis module includes: Intelligent photovoltaic monitoring and optimization module: Based on historical meteorological data, light intensity, temperature and other information, the model is trained to accurately predict photovoltaic power generation and guide the daily operation and maintenance and cleaning timing of the photovoltaic array. At the same time, combined with the intelligent optimization algorithm, the optimal tilt angle and azimuth angle of the photovoltaic module are adjusted to improve power generation efficiency.

[0066] Energy storage system intelligent scheduling module: Establish a charging and discharging strategy model for the energy storage system, realize the optimal charging and discharging operation of the energy storage equipment in different time periods, and cooperate with photovoltaic power generation to perform peak shaving and valley filling and energy time shifting. At the same time, the AI ​​model is used to identify battery attenuation in advance, dynamically adjust the charging and discharging depth, and extend the service life of the energy storage system.

[0067] Energy consumption monitoring and energy-saving suggestion module: By analyzing the user's electricity consumption habits, load characteristics and external environmental factors, it provides accurate energy consumption forecasts and energy-saving strategy suggestions. It monitors the energy consumption of each link in real time, identifies potential energy waste through anomaly detection algorithms, and proposes improvement measures to achieve energy conservation and carbon reduction.

[0068] Centralized control and trading of virtual power plants: Using reinforcement learning and game theory to build complex decision-making models, virtual power plants can conduct flexible and efficient energy dispatch and power trading based on real-time electricity price signals, grid dispatch instructions, and the status of distributed energy within themselves. Simulate profit strategies under various market scenarios to maximize asset value.

[0069] Embodiment 3

[0070] This embodiment is a further refinement of the second embodiment.

[0071] The intelligent photovoltaic monitoring and optimization module uses advanced meteorological forecasting models and Internet of Things technology to monitor the light intensity, temperature and other environmental parameters of the photovoltaic power station in real time. It combines historical data and real-time data analysis to predict photovoltaic power output, optimize the tilt angle and cleaning and maintenance plan to maximize photovoltaic power generation efficiency.

[0072] The intelligent photovoltaic monitoring and optimization module includes two sub-modules: real-time tracking and prediction, and power regulation and matching. Among them, real-time tracking and prediction includes environmental parameter prediction model, power output prediction model and cleaning maintenance plan formulation model; power regulation and matching includes power regulation algorithm and grid matching optimization algorithm.

[0073] Environmental parameter prediction model: In photovoltaic power generation, environmental parameters such as light intensity and temperature directly affect the efficiency of power output. By predicting these parameters, power output can be planned in advance, system operation can be optimized, and power losses caused by sudden environmental changes can be reduced. In addition, accurate predictions can help better integrate into grid dispatching and improve the overall stability of the power system.

[0074] L predicted (t) = f poredict (L history , T history , H history , ...; Θ L )

[0075] T predicctrd (t) = g predict (T history , L history , H history , ...; Θ T )

[0076] In the light intensity prediction model, Lhistory, Thistory, and Hhistory refer to historical light intensity data, historical temperature data, and historical humidity data, respectively.

[0077] In the temperature prediction model, Thistory, Lhistory, and Hhistory represent historical temperature data, historical light intensity data, and historical humidity data, respectively.

[0078] Among them, L predicted (t): light intensity at time t (lux); T predicted (t): ambient temperature at time t (℃); f predict and g predict is the prediction model, Θ L and θ T Represents model parameters.

[0079] Power output forecasting model: Power output forecasting is crucial to the economic operation of photovoltaic systems. It helps operators understand future power output and facilitates the arrangement of power transactions, energy storage system scheduling, and interaction with the power grid in advance. Accurate forecasting can reduce the demand for backup power and unnecessary energy waste, while also providing users with a more stable power supply.

[0080] P predicted (t) = h(L predicted (t), T predicted (t), H(t); Θ P )

[0081] P predicted (t): predicted power output (kW); h is the prediction function based on light intensity, temperature and historical power output, Θ p is the model parameter, H history (t): historical electricity production at time t (kWh).

[0082] Cleaning and maintenance plan formulation model: The cleanliness of photovoltaic panels directly affects their photoelectric conversion efficiency. By analyzing the panel cleaning cycle through a predictive model, maintenance can be effectively arranged to avoid reducing power generation efficiency due to dust accumulation. A reasonable cleaning and maintenance plan can extend the service life of the photovoltaic system, reduce maintenance costs, and maintain a high level of energy output.

[0083] C schedule = k(L history , T history , H history , P history ; Θ C )

[0084] C schedule : Cleaning maintenance plan; K is the cleaning plan formulation function based on environmental factors and the decreasing trend of power output, Θ c as a parameter.

[0085] Power regulation algorithm: Photovoltaic power output is intermittent and unstable. The power regulation algorithm dynamically adjusts the inverter output power to ensure that power is smoothly injected into the grid and avoid impact on the grid. This is crucial to maintaining grid frequency and voltage stability.

[0086] P inverter (t) = f regulate (P predicted (t), P grid-demand (t); θ R )

[0087] Among them, P grid-demand (t): grid load demand at time t (kW); P inverter (t): inverter output power at time t (kW); f regulate : Power regulation function, f cguiate Consider predicting power output and grid demand, and adjust inverter output to ensure matching with grid frequency and voltage; R To adjust the function parameters;

[0088] Grid matching optimization algorithm:

[0089]

[0090] Where P grid-demand (t): grid load demand at time t (kW); P inverter (t): inverter output power at time t (kW).

[0091] The intelligent dispatching module of the energy storage system includes two sub-modules: energy storage optimization configuration and collaborative control strategy.

[0092] Energy storage optimization configuration: In combination with the intermittent characteristics of photovoltaic power generation, advanced algorithm models are used to optimize the discharge capacity (not the physical capacity of the energy storage battery) and charging and discharging strategies of energy storage equipment (such as lithium battery energy storage systems), ensuring that excess electricity is stored when there is sufficient sunlight, and that electricity is released during non-sunny periods or during peak periods of the power grid, smoothing the electricity consumption curve and improving the stability of the power grid.

[0093]

[0094] Among them, E grid (t) = E load (t)-E pv (t)+P dis (t)-P charge (t), is the net grid demand at time t; α and β represent weight factors, reflecting the importance of smoothing grid demand and energy storage level fluctuations, respectively.

[0095] Constraints include:

[0096] 1. Energy storage charging and discharging power limit: 0≤P charge (t)≤P max,charge,0 ≤P dis (t)≤P max,dis .

[0097] 2. Energy storage capacity limitation: C min ≤S(t)≤C max .

[0098] 3. Energy conservation:

[0099] Collaborative control strategy: The central controller uses machine learning algorithms to automatically adjust the energy storage charging and discharging strategy based on the real-time monitoring of the grid status, user demand and energy storage system status, to achieve a dynamic balance between energy supply and demand and improve overall energy utilization efficiency.

[0100] Define states t =[E pv (t),E lood (t),S(t)], action a t :=[P charge (t),P dis (t)], reward r t It can be designed based on grid balance, energy storage efficiency and cost-effectiveness.

[0101] Use Q-leamning algorithm to update Q value:

[0102]

[0103] Among them, α is the learning rate, γ is the discount factor, and r t It is an instant reward, obtained through simulation or actual operation.

[0104] The algorithm uses an enhanced machine learning framework, which continuously learns the relationship between the environment (grid status, user demand and energy storage status) and actions (charging and discharging strategies), and automatically adjusts the strategy to achieve better grid stability and energy utilization efficiency.

[0105] The energy consumption monitoring and energy-saving suggestion module includes three sub-modules: multi-dimensional energy consumption monitoring, energy efficiency analysis and suggestions, and demand-side response.

[0106] Multi-dimensional energy consumption monitoring deploys smart meters and sensors to conduct fine-grained monitoring of energy consumption on the user side, covering multiple types of energy such as electricity and thermal energy, to achieve transparent management of energy flows.

[0107]

[0108] Among them, Ns and N m are the number of sensors and smart meters respectively; E electric (t): power consumption at time t (kWh); E thermal (t): heat energy consumption at time t (kWh); Sensor i (t): the reading of the i-th sensor at time t; Meters j (t): The record of the jth smart meter at time t. total (t) represents the total amount of electricity used in a certain period of time. Energy efficiency analysis and suggestions: Analyze user energy consumption data, identify energy consumption peaks, inefficient equipment and potential energy saving space, and provide users with customized energy-saving optimization suggestions and strategies, such as load transfer and equipment upgrade suggestions. Energy efficiency analysis and suggestions include energy consumption peak identification algorithms, inefficient equipment identification algorithms and energy-saving suggestion generation algorithms.

[0109] Energy consumption peak identification algorithm: By analyzing historical electricity consumption data and identifying peak energy consumption periods, it can help companies understand their electricity consumption patterns, assist grid operators in load management, avoid power supply shortages and overload risks, and improve the safety and reliability of grid operations.

[0110]

[0111] Among them, E peak (t k ): Peak energy consumption in time period k, E total (t) represents the maximum total power consumption.

[0112] Inefficient equipment identification algorithm: Identifying which equipment or systems consume too much energy during operation is crucial to improving overall energy efficiency. Through data analysis, it can be determined which equipment needs maintenance, upgrades or replacements, thereby reducing energy waste and lowering operating costs.

[0113]

[0114] LowEfficiency devices ={device|Efficiency score (device)<θ

[0115] In the formula, Efficiency device: device energy efficiency ratio; θ is the energy efficiency threshold; Efficiencyscore(device) represents the efficiency score of the device; Einput(device) represents the input energy of the device; Eoutput(device) represents the output energy of the device; LowEfficiencydevices represents the set of inefficient devices.

[0116] Energy-saving suggestion generation algorithm: Based on energy consumption analysis, the system can automatically generate personalized energy-saving suggestions for users, including but not limited to adjustments to device usage habits, suggestions for device upgrades, and more efficient energy management strategies.

[0117]

[0118] Strategy custom (user) = load transfer suggestion + equipment upgrade suggestion

[0119] Where, Savingspotential: potential energy saving space; Savingspotential: represents the total energy saving potential; represents the set of inefficient devices; Einput(device) represents the input energy of the device; Eoutput(device) represents the output energy of the device; η optimal Represents the optimal efficiency coefficient; Strategycustom(user) generates a personalized energy-saving strategy based on the data passed in by the user.

[0120] Demand-side response algorithm: Combines user behavior models and real-time electricity price information to encourage users to use electricity during off-peak hours or participate in demand response programs, balances grid loads through economic incentive mechanisms, and reduces energy costs.

[0121] Establishing a user behavior model helps to understand the electricity consumption habits and preferences of different user groups. It can predict and guide users to adjust their electricity consumption within a specific time period to respond to changes in electricity supply and demand, reduce peak electricity demand, and smooth the grid load curve:

[0122] U utility (t)=α·Load(t)-β·Price(t)·Load(t)

[0123] Where α and β represent the user's sensitivity to electricity consumption and price, Uutility(t) is the energy utility value of the user in time t; Price(t) is the low electricity price period; Load(t) is the power load level.

[0124] In order to encourage users to participate in demand-side response, it is necessary to design a reasonable incentive mechanism, such as electricity price discounts, reward points or other forms of compensation. The incentive mechanism model can quantify the effects of different incentives, evaluate user participation and response effects, and then optimize the incentive strategy to improve the efficiency and participation of demand-side response:

[0125] Incentive(t)=γ·Load shifted (t)·(Price peak -Price(t))

[0126] In the formula, γ is the incentive coefficient, Price peak is the peak electricity price; lncentive(t) is the incentive amount the user receives at time t due to load shifting; Loadshifted(t) is the load reduction caused by the user changing his electricity usage habits at time t; Pricepeak-Price(t) is the difference between the peak price Pricepeak and the price Price(t) at the current time period t.

[0127] Through intelligent scheduling and control, load transfer strategies can shift part of the electricity demand during off-peak hours to other hours, reducing the power pressure during peak hours while taking advantage of the lower electricity prices during off-peak hours. This can not only reduce costs, but also improve the overall efficiency and stability of the power system and promote the rational allocation of power resources:

[0128]

[0129] Where Loadoptimal(t) represents the optimal load level at a given time t; Utility(t) represents the total utility at time t; Incentive(t) represents the incentive amount at time t.

[0130] Determine the optimal load transfer strategy based on the principles of economic incentives and user utility maximization.

[0131] Through the above technical integration, the present invention builds a closed-loop energy management ecosystem, which not only improves the utilization rate of photovoltaic power generation and the economy of energy storage systems, but also promotes the intelligent and efficient use of energy on the user side, and significantly improves the flexibility, reliability and sustainability of the entire energy supply chain. This not only helps to reduce dependence on fossil fuels, but also provides strong technical support for achieving carbon reduction goals and promoting green energy transformation.

[0132] Embodiment 4

[0133] This embodiment is a further refinement of the second embodiment.

[0134] Performance evaluation: To ensure continuous improvement and optimization of the overall performance of the energy system, the present invention has built-in an advanced periodic evaluation system that focuses on comprehensive, detailed and quantitative review of energy operation results. This mechanism not only includes multi-dimensional analysis of photovoltaic power generation efficiency, energy storage system efficiency and energy management, but also automatically generates detailed performance indicator reports through highly integrated data analysis tools, providing solid data support and insights for management's strategic decisions.

[0135] The performance evaluation includes four sub-modules: photovoltaic power generation performance evaluation, energy storage system efficiency review, energy consumption monitoring and energy-saving suggestions, and comprehensive performance indicator report.

[0136] Photovoltaic power generation efficiency evaluation includes capacity efficiency analysis algorithm and environmental impact correction algorithm.

[0137] Capacity efficiency analysis algorithm: Use advanced algorithms to deeply analyze the power generation of the photovoltaic array, consider the influence of factors such as light intensity, temperature, tilt angle, etc., quantitatively evaluate the power generation efficiency per unit area, identify performance bottlenecks and propose improvement measures.

[0138] P theoretical (t) = A area ×I lighi (t)×η nominal ×α×(1-β×(T cell (t)-T ref ))

[0139]

[0140] Where, T ref is the reference temperature, P tedoretical (t) represents the photovoltaic power generation efficiency; P gen (t): actual electrical power generated at time t (kW); A area :Total area of ​​photovoltaic panels (m 2 ), I light (t): light intensity at time t (kW / m 2 ), T cell (t): time t battery temperature (℃), α: light intensity influence coefficient, β: temperature coefficient, η nominal : Theoretical maximum conversion efficiency (%), η eff (t): actual conversion efficiency at time t (%).

[0141] Environmental impact correction algorithm: Considering the potential impact of seasonal changes and climate anomalies on power generation, the accuracy and reliability of the evaluation results are ensured through meteorological data calibration algorithm, seasonal correction algorithm and performance evaluation algorithm. The environmental impact correction algorithm includes meteorological calibration model algorithm and seasonal correction algorithm.

[0142] Meteorological calibration model algorithm: The power generation of photovoltaic systems is affected by many meteorological factors, such as sunshine intensity, temperature, cloud cover, etc. The meteorological calibration model collects and analyzes real-time and historical meteorological data to correct power generation fluctuations caused by weather changes. This helps to more accurately predict and evaluate the power generation capacity of photovoltaic systems and provide reliable data support for power dispatch and energy storage planning.

[0143] M corr (t) = f(I light_corrected (t), T cell_corrected (t), Wind_speed(t), Humidity(t))

[0144] I light_correted (t) represents the light intensity, T cell_corrected (t) represents the battery temperature, Wind_speed(t) represents the wind speed, and Humidity(t) represents the humidity;

[0145] Seasonal correction algorithm: Due to seasonal changes in sunshine duration and angle, the power generation of the photovoltaic system will fluctuate with the season. The seasonal correction algorithm corrects the capacity of the photovoltaic system by considering the solar radiation characteristics of different seasons, ensuring that the evaluation results reflect the true seasonal impact and helping managers formulate reasonable annual operation and maintenance plans and financial budgets.

[0146] F season =g(Month(t), DayOfYear(t))

[0147] Where Month(t) represents the seasonal correction factor; DayOfYear(t) represents the month and the specific day of time t.

[0148] Efficiency evaluation algorithm: After comprehensively considering production efficiency and environmental factors, the efficiency evaluation algorithm is used to calculate the overall efficiency indicators of the photovoltaic system, such as performance ratio, energy yield index, etc. These indicators not only reflect the power generation efficiency of the system, but also consider the impact of external environmental conditions on system performance, providing a scientific basis for the optimization and upgrading of photovoltaic systems.

[0149] P corrected (t) = P gen (t)×M corr (t)×F season

[0150]

[0151] Where P gen (t): The actual electric power generated at time t (kW), P corrected(t) Corrected power generation; η final (t) Final efficiency; M cor r(t) is the meteorological calibration factor; F season is the seasonal correction factor; A area is the effective area of ​​the photovoltaic array; I light_corrected (t) Corrected time t.

[0152] The energy storage system efficiency review includes two sub-modules: charging and discharging cycle efficiency tracking and economic and environmental contribution analysis.

[0153] The charge and discharge cycle efficiency tracking includes the charge and discharge efficiency calculation algorithm, the battery life prediction algorithm and the response speed evaluation algorithm.

[0154] Charge and discharge efficiency calculation algorithm: It is crucial to evaluate the energy conversion efficiency of the energy storage system during the charge and discharge process. This not only affects the overall performance of the system, but is also directly related to its economy and sustainability. By calculating the charge and discharge efficiency, the loss of the system during the energy conversion process can be determined, providing data support for optimizing the design and operation strategy.

[0155] The charge efficiency (CE) and discharge efficiency (DE) can be calculated by measuring the energy difference during the charge and discharge process. Assume the rated capacity of the battery is C nominal , the energy actually charged in a cycle is E charged , the actual energy released during discharge is E discharged .

[0156]

[0157] Among them, V charge and V discharge are the average voltage during charging and discharging, respectively.

[0158] Battery life prediction algorithm: Battery life is a key factor in determining the cost and replacement cycle of energy storage systems. The prediction algorithm can estimate the remaining service life of the battery by analyzing the battery's charge and discharge history, temperature, depth of discharge and other conditions, helping operation and maintenance personnel plan maintenance and replacement in advance, reduce unexpected downtime, and optimize the total cost of the system during its life cycle.

[0159] Assuming that the battery life has an exponential decay relationship with its cycle number Ngde, the battery state of health (SOH) can be expressed in a simplified form of the Arrhenius equation:

[0160]

[0161] Where k is the degradation rate constant, which depends on the battery type and usage conditions, N cyclos Indicates the number of complete charge and discharge cycles the battery has undergone.

[0162] Response speed evaluation algorithm: Response speed is a key performance indicator for energy storage systems to play frequency regulation and rapid response roles in the power grid. By measuring the time it takes for the system to fully respond from receiving instructions, the evaluation algorithm can ensure that the energy storage system can act quickly in an emergency, meet the instantaneous needs of the power grid, and improve the stability and reliability of the entire power system.

[0163] Based on the change time from when the battery management system (BMS) receives the command to when the inverter starts to output power, plus the inverter's own response time ((τ_{inv}), in seconds). If it is assumed that the response of the battery and inverter is approximately linear, the algorithm formula is:

[0164] T response，start =τ sys +τ inv

[0165] Among them, T response，start It represents the time required for the energy storage system to respond from receiving the command; τ sys It represents the time from when the energy storage system receives the command to when the system is ready to respond; τ inv Indicates the time from receiving a command to the inverter starting to output current or voltage.

[0166] In addition, it is also necessary to consider the impact of battery status (such as SOC, temperature), historical charge and discharge status on response time, and environmental factors (such as the impact of temperature on battery performance).

[0167] The economic and environmental contribution analysis includes an economic analysis algorithm, an environmental contribution analysis algorithm, and a demand response efficiency feedback algorithm.

[0168] Economic analysis algorithm:

[0169] The economic efficiency of energy storage systems is directly related to their commercial feasibility and promotion potential. By analyzing economic indicators such as the return on investment, operating costs, and power saving benefits of energy storage systems, the economic benefits of the system can be comprehensively evaluated, providing a basis for decision-making for investors and decision makers.

[0170] The economic value of balancing grid supply and demand (E balance ): The cost savings and increased revenue of the energy storage system through peak load shaving and valley filling can be expressed as follows:

[0171] E balance =∑(peak power price - valley power price)×energy storage capacity×utilization rate

[0172] Among them, the utilization rate reflects the proportion of the energy storage system actually used for peak shaving and valley filling.

[0173] Environmental contribution analysis algorithm: It is becoming increasingly important to evaluate the environmental benefits of energy storage systems in terms of reducing carbon emissions and improving energy efficiency. By quantifying the positive impact of energy storage systems on the environment, the environmental contribution analysis algorithm can prove their role in building a green and low-carbon society and promote energy companies and the public to support clean energy technologies.

[0174] Carbon emission reduction (Creduced): Based on the energy storage system promoting more renewable energy access and reducing the use time of fossil fuel power plants, the carbon emission reduction can be estimated by comparing the total emission changes before and after energy storage:

[0175] C reduced = (emissions without energy storage - emissions with energy storage) × number of years of operation

[0176] Energy system resilience improvement (Rresilience): The resilience of the energy system is difficult to quantify directly as a monetary value, but it can be indirectly reflected by reducing power outage losses or improving disaster recovery speed. Here it is simplified to a hypothetical ratio of resilience improvement coefficient to avoided power outage cost:

[0177] R resilience = Toughness improvement factor × average power outage loss avoidance

[0178] Energy consumption monitoring and energy-saving suggestions: Energy consumption segmentation and optimization suggestions: Deeply mine the energy consumption data on the user side, segment it into specific time periods, regions and even individual devices, identify high-energy consumption links, and propose personalized strategies for energy conservation and emission reduction.

[0179] Energy consumption monitoring and energy-saving suggestions include two steps: energy consumption segmentation and optimization suggestion generation.

[0180] Energy consumption segmentation is as follows: The energy consumption segmentation algorithm can break down the overall energy consumption into specific time periods, regions, and even individual devices, which is crucial for identifying energy consumption patterns and high-energy consumption links. Through refined energy consumption data, the source of energy waste can be accurately located, providing data support for the next step of energy-saving transformation:

[0181] Summarize the energy consumption of each region r at time t:

[0182]

[0183] By comparing the total energy consumption Etotal(t, r) under each time-region combination with the set energy consumption threshold Cthreshold, all time periods and regions above the threshold are found and marked, thus obtaining the high energy consumption time period set Hperiods and the high energy consumption region set Hregions. This information is crucial for subsequent analysis of energy consumption patterns, identification of problem sources, and formulation of energy-saving measures.

[0184] The optimization suggestion generation is as follows: Based on the results of energy consumption segmentation, the optimization suggestion algorithm can generate targeted energy-saving and emission reduction strategies. Whether it is equipment upgrades, operation process optimization or time scheduling adjustments, accurate energy-saving suggestions can help enterprises or individuals reduce energy consumption and reduce operating costs.

[0185] For each high energy consumption period p∈H periods and region r∈H regions , based on historical data and equipment characteristics analysis, propose strategies:

[0186] S segment (r,p)=Personalized energy saving strategy (E device (t∈p,r,n),n)

[0187] Combining all the suggestions to form the final strategy set:

[0188] S final =U r,p S segment (r,p)

[0189] Demand response effectiveness feedback: Evaluate the degree of response and actual effect of users participating in the demand-side management plan, quantify the power saving results, and provide feedback information for subsequent demand response strategy optimization.

[0190] Demand response effectiveness feedback algorithm: evaluates the response level and actual effect of users participating in the demand-side management plan, quantifies the power saving effect, and provides feedback information for subsequent demand response strategy optimization.

[0191] The demand response performance feedback algorithm includes two steps: participation quantification and power saving effectiveness quantification.

[0192] Participation quantification: The success of a demand response program depends largely on user participation. By quantifying participation, we can evaluate how users respond to demand response programs, understand which incentives are most effective, and which time periods or types of users are more willing to participate.

[0193]

[0194] Among them, T total is the total time period length.

[0195] Quantification of power saving results: Quantification of power saving results is the key to evaluating the actual effect of demand response programs. By comparing energy consumption data before and after the implementation of demand response, the power saving results brought by the program can be intuitively demonstrated, including power saving, cost savings and environmental benefits.

[0196] Calculate the power saving ratio:

[0197]

[0198] In the formula, E ratio Indicates the proportion of power saving; E base (t) represents the reference electric energy consumption at time t; E saved (t) represents the electric energy saved by energy-saving measures at time t.

[0199] Calculate the effectiveness index:

[0200] Effectiveness = DR score ×E ratio

[0201] In the formula, Effectiveness represents the effectiveness index of power saving effect; DR score represents the demand response score; E ratio Indicates the percentage of power saving.

[0202] Embodiment 5

[0203] This embodiment provides a comprehensive energy management method based on big data, and the management method includes the following steps:

[0204] Step S1, equipment deployment and debugging: installing and debugging the corresponding equipment;

[0205] Step S2: IoT edge gateway deployment, device interconnection and intercommunication; local area network technology is used to deploy IoT edge gateway as a bridge connecting devices and the cloud;

[0206] Step S3, big data engine, intelligent data aggregation; the big data engine collects IoT edge gateway data in all sites in real time through a secure and efficient message queue to ensure data integrity, security and stability;

[0207] Step S4, intelligent engine, intelligent insight and decision support: Using artificial intelligence and machine learning algorithms, the intelligent engine conducts in-depth analysis of the managed data, not only generating real-time alarms and accurate prediction models, but also providing a scientific basis for intelligent control of equipment and enterprise operational decisions;

[0208] Step S5, business application, comprehensive coverage of diversified business scenarios; according to the specific needs of energy management, the system provides a wealth of business applications, covering real-time monitoring and in-depth analysis of energy supply, intelligent scheduling and health management of energy storage resources, and efficiency monitoring and fault diagnosis of energy-consuming equipment.

[0209] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0210] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A comprehensive energy management system based on big data, characterized by: It includes a data collection and processing module, an intelligent scheduling and analysis module, and an efficiency evaluation module. The data collection and processing module is connected to the intelligent scheduling and analysis module. The data collection and processing module is used for comprehensive and real-time data collection, and the collected data is processed and analyzed through cloud computing. The intelligent scheduling and analysis module is also connected to the efficiency evaluation module. The intelligent scheduling and analysis module is used for monitoring and scheduling of photovoltaic power generation and energy storage, as well as in-depth analysis of the comprehensively processed energy data, accurate prediction and optimization of intelligent scheduling of energy storage resources, and fine monitoring of energy consumption behavior; The performance evaluation module is used to conduct multi-dimensional analysis of photovoltaic power generation efficiency, energy storage system efficiency and energy management, and generate performance indicator reports.

2. The comprehensive energy management system based on big data according to claim 1, characterized in that: The data collected in the data acquisition and processing module include: data from the power generation side, energy storage side and energy consumption side. The power generation side includes solar panels, photovoltaic inverters, solar charging controllers, monitoring systems and meteorological data; the energy storage side includes energy storage batteries, battery management systems, energy storage inverters and energy management systems; the energy consumption side includes air conditioning, lighting, charging piles and data centers.

3. The comprehensive energy management system based on big data according to claim 1, characterized in that: Data processing in the data acquisition and processing module includes: eliminating invalid and erroneous information through data cleaning to ensure data quality; it can also deeply integrate and standardize the cleaned data to make it in a unified format for easy analysis.

4. The comprehensive energy management system based on big data according to claim 1, characterized in that: The intelligent scheduling and analysis module includes photovoltaic monitoring and optimization module, intelligent scheduling of energy storage system, energy consumption monitoring and energy-saving suggestions, and centralized control and trading of virtual power plants.

5. The comprehensive energy management system based on big data according to claim 4 is characterized in that: The photovoltaic monitoring and optimization module is used to monitor the light intensity and temperature of the photovoltaic power station in real time, combine historical data and real-time data analysis to predict photovoltaic power output, optimize the tilt angle and cleaning and maintenance plan to maximize photovoltaic power generation efficiency; The photovoltaic monitoring and optimization module includes real-time tracking and prediction models as well as power regulation and matching algorithms.

6. The comprehensive energy management system based on big data according to claim 5 is characterized by: The real-time tracking and prediction model includes the power output prediction model, specifically: P predicted (t)=h(L predicted (t),T predicted (t),H(t);Θ P ) Pprcdicte d (t): predicted power output (kW); h is the prediction function based on light intensity, temperature and historical power output, and Θp is the model parameter.

7. The comprehensive energy management system based on big data according to claim 1 is characterized by: The power regulation and matching algorithm includes the power regulation algorithm and the grid matching optimization algorithm, which are as follows: Power regulation algorithm: P inverter (t)=f regulate (P predicted (t),P grid-demand (t);Θ R ) Among them, P grid-demand (t): grid load demand at time t (kW); P inverter (t): inverter output power at time t (kW); f regulate : Power regulation function, f cguiate Consider predicting power output and grid demand, and adjust inverter output to ensure matching with grid frequency and voltage; R To adjust the function parameters; Grid matching optimization algorithm: Where P grid-demand (t): grid load demand at time t (kW); P inverter (t): inverter output power at time t (kW).

8. The comprehensive energy management system based on big data according to claim 1 is characterized by: The performance evaluation module includes photovoltaic power generation performance evaluation and energy storage system efficiency review; the photovoltaic power generation performance evaluation includes a capacity efficiency analysis algorithm and an environmental impact correction algorithm; Capacity efficiency analysis algorithm: P heoretical (t)=A area ×I light (t)×η nominal ×α×(1-β×(T cell (t)-T ref )) Where, T ref is the reference temperature, P theoretical (t) represents the photovoltaic power generation efficiency; P gen (t): the actual electrical power generated at time t (kW); A area :Total area of ​​photovoltaic panels (m), I light (t): light intensity at time t (kW / m 2 ), T cell (t): time t battery temperature (℃), α: light intensity influence coefficient, β: temperature coefficient, η nominal : Theoretical maximum conversion efficiency (%), η eff (t): actual conversion efficiency at time t (%); The environmental impact correction algorithm includes a meteorological calibration model algorithm and a seasonal correction algorithm; the meteorological calibration model algorithm is specifically: M corr (t)=f(I light_corrected (t),T cell_corrected (t),Wind_speed(t),Humidity(t)) In the formula, I light_correted (t) represents the light intensity, T cell_corrected (t) represents the battery temperature, Wind_speed(t) represents the wind speed, and Humidity(t) represents the humidity; The specific seasonal correction algorithm is: F season =g(Month(t),DayOfYear(t)) Where Month(t) represents the seasonal correction factor; DayOfYear(t) represents the month and the specific day of time t.

9. The comprehensive energy management system based on big data according to claim 8, characterized in that: Energy storage system efficiency review includes charge and discharge cycle efficiency tracking; charge and discharge cycle efficiency tracking includes charge and discharge efficiency calculation algorithm and battery life prediction algorithm response speed evaluation algorithm; The specific calculation algorithm of charge and discharge efficiency is as follows: the charge efficiency (CE) and discharge efficiency (DE) can be calculated by measuring the energy difference during the charge and discharge process; assuming the rated capacity of the battery is C nominal , the energy actually charged in this cycle is E charged , the actual energy released during discharge is E discharged : Among them, V charge and V discharge are the average voltages during charging and discharging, respectively; Battery life prediction algorithm: Assuming that the battery life has an exponential decay relationship with its cycle number Ngde, the battery state of health (SOH) can be expressed in a simplified form of the Arrhenius equation: Where k is the degradation rate constant that depends on the battery type and usage conditions.

10. The comprehensive energy management system based on big data according to claim 9, characterized in that: The charge-discharge cycle efficiency tracking also includes a response speed assessment algorithm; Response speed evaluation algorithm: T response,start =t sys +t inv Among them, τ sys Represents the processing time of the system control layer, including communication and instruction parsing.

Images