Park energy storage intelligent scheduling system and method based on big data

Abstract

The invention discloses a park energy storage intelligent scheduling system and method based on big data, and relates to the technical field of big data park energy storage intelligent scheduling, and the method comprises the following steps: S100, collecting the real-time operation data of multiple types of energy systems in a park, constructing a time sequence database, and carrying out the preprocessing; s200, predicting load change based on an energy demand scene, and constructing a polymorphic energy storage model in combination with dynamic characteristics of energy storage equipment; s300, defining a multi-objective optimization function taking energy loss, renewable energy utilization rate and energy storage efficiency as cores, establishing a total energy balance model, and setting constraint conditions of charging and discharging and heat and cold storage power of energy storage equipment; and S400, updating the load demand and the energy storage state by using the real-time data, calculating the load deviation and correcting the target function, and readjusting the charging and discharging of the energy storage equipment and the heat storage and cold storage power distribution according to the corrected target function.

Description

Technical Field

[0001] The present invention relates to the technical field of big data park energy storage intelligent scheduling, and specifically to a park energy storage intelligent scheduling system and method based on big data. Background Art

[0002] In recent years, with the advancement of the "dual carbon" goals and the need for large-scale development of renewable energy, the importance of park energy storage systems has become increasingly prominent. Long-term energy storage technology, as the key to large-scale application of renewable energy, can not only effectively solve the intermittent and volatility problems of new energy, but also smooth energy output and improve the energy utilization efficiency of the park.

[0003] However, there are many problems in the design and operation of existing park energy storage systems. First, the uncertainty of the park energy system increases. The operation of the park's comprehensive energy system is affected by multiple uncertain factors such as load fluctuations, changes in energy prices and natural conditions. The existing planning and design methods and operation strategies are difficult to deal with effectively. In addition, the energy demand of the park is diverse and dynamic, and the coordinated scheduling of cooling and heating, multiple energy storage and renewable energy is extremely complex. Secondly, the dynamic characteristics of the energy storage system are insufficient. The deep coupling of energy storage and renewable energy requires grasping the dynamic operating characteristics of different energy systems. However, the current technical level has little research on the coupling mechanism of multi-energy systems and the dynamic characteristics of all operating conditions, resulting in low efficiency in the coordinated operation of the energy storage system and other energy systems. Therefore, technological innovation is urgently needed to realize the intelligent scheduling and management of park energy storage. Summary of the invention

[0004] The purpose of the present invention is to provide a park energy storage intelligent scheduling system and method based on big data to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: a method for intelligent scheduling of energy storage in a park based on big data, the method comprising the following steps: S100, collect real-time operation data of various types of energy systems in the park, including load data, renewable energy data and equipment operation status data, build a time series database, and clean, model and extract features of the data; S200, predict short-term and long-term load changes based on energy demand scenarios, analyze the dynamic characteristics of energy storage equipment, build thermal, electrical, and cold multi-modal energy storage models, and evaluate load uncertainty and coupling relationships; S300. Based on the energy demand scenario and energy storage model, define a multi-objective optimization function with energy loss, renewable energy utilization rate and energy storage efficiency as the targets, build a total energy balance model, and set constraints on the charging and discharging of energy storage equipment and the heat storage and cold storage power; S400, using real-time data to update load demand and energy storage status, calculate load deviation and correct the objective function, and readjust the energy storage equipment charging and discharging and heat storage and cold storage power distribution according to the corrected objective function.

[0006] In step S100, the specific steps include: S101. Collect energy demand data from the end users of the park, build a time series database, and monitor changes in energy demand, including load data, renewable energy data, and equipment status data. The load data includes the power demand, heating load, and cooling load of the park, taking into account the impact of seasonality, diurnal periodicity, and meteorological conditions on load changes. Renewable energy data includes real-time output and historical data of photovoltaic power generation and wind power generation, combined with meteorological data related to solar radiation intensity, wind speed, and temperature and humidity. Equipment status data includes the real-time status and charging and discharging efficiency of electric energy storage, thermal energy storage, and cold energy storage equipment. S102. Clean and model the original data, complete the data through outlier detection combined with regression analysis based on historical data and adaptive interpolation algorithm, and use big data fusion technology to build a unified data model of electricity-heat-cold multi-energy flow based on the characteristics of electricity, heat, cooling load and renewable energy. Then, use the Latin hypercube sampling method, combine historical data and environmental parameters, establish a random uncertainty mathematical model, and analyze the randomness and volatility of load demand and renewable energy output; S103. Use principal component analysis and machine learning algorithms to extract features of energy demand and energy storage equipment operation characteristics, and output energy demand scenarios under typical load and meteorological conditions and load change characteristics based on time series.

[0007] In step S200, the specific steps include: S201. Based on the energy demand scenario outputted in step S100, combined with the user's historical energy consumption behavior and load characteristics, load forecasting is performed using short-term and long-term forecasting models. A long short-term memory network model is used to predict short-term load fluctuations during peak power consumption periods. At the same time, the long-term trend of seasonal load demand is predicted based on typical meteorological conditions and energy demand scenarios. Through a deep learning model of multi-task learning, the coordinated forecasting of electricity, heat and cooling load demand is analyzed and the coupling relationship between loads is evaluated. S202. Based on the constructed unified data model of electric-heat-cold multi-energy flow and the mathematical model of random uncertainty, study the operating characteristics of heating, cooling and electric energy storage equipment and their adaptability in different energy scenarios, and quantify the volatility and uncertainty of rigid loads in different scenarios through load forecasting results; S203. Construct a multi-modal model of the energy storage system, including: analyzing the energy conversion efficiency of energy storage based on the Brayton cycle, studying the thermodynamic model of the electrical, thermal and cold coupling characteristics of the energy storage equipment, and establishing an economic model based on the round-trip efficiency, energy storage density and power density of the energy storage equipment. At the same time, analyze the dynamic characteristic model of the energy storage efficiency and output power change law of the energy storage equipment under different load rates, environmental conditions and power inputs. Finally, based on the energy demand scenario and load forecast results, combined with the operating characteristics of the energy storage equipment, generate a charging and discharging plan for the energy storage equipment.

[0008] In step S300, the specific steps include: S301. Based on the energy demand scenario generated by S100 and the energy storage model constructed by S200, a multi-objective optimization function with energy loss, renewable energy utilization rate and energy storage efficiency as the core is defined. The specific formula is defined as follows: The energy loss calculation formula is as follows: ; Among them, P total (t) represents the difference between the total power demand and the actual output in time period t, η overall Indicates the comprehensive loss rate including grid transmission and charging and discharging of energy storage equipment; The formula for calculating the utilization rate of renewable energy is as follows: ; Among them, P renewable (t) represents the total power generation of renewable energy in period t, represents the renewable energy power actually utilized during time period t; The energy storage efficiency calculation formula is as follows: ; Among them, E in (t), E out (t) represents the input and output electrical energy of the energy storage device in time period t, Q heat,actual (t), Q heat,predicted (t) represents the heat storage and release power of the thermal energy storage device in time period t, Q cold,actual (t), Q cold,predicted (t) represents the cold storage and release power of the cold energy storage device in time period t; S302, calculate the optimization objective function according to energy loss, renewable energy utilization rate and energy storage efficiency, and the calculation formula is as follows: total =J renewable / J loss +ln(1+J efficiency ); and set energy balance and energy storage equipment charging and discharging constraints. The total energy balance calculation formula is as follows: ; Among them, P demand (t) represents the total load demand in time period t, P grid (t) represents the power purchased from the power grid during period t, P storage Indicates the charging and discharging power of the energy storage device; Energy balance formula for thermal energy storage equipment: Q heat (t) = Q heat,demand (t)+Q heat,charge (t)-Q heat,discharge (t); Among them, Q heat,demand (t) represents the heat load demand during time period t; Energy balance formula for cold energy storage equipment: Q cold (t) = Q cold,demand (t)+Q cold,charge (t)-Q cold,discharge (t); Among them, Q cold,demand (t) represents the cooling load demand in period t; The state of the energy storage device is defined as follows: E(t)=E(t-1)+η charge *P charge (t)*Δt-P discharge (t)*Δt / η discharge ; Where E(t) and E(t-1) represent the remaining energy of the energy storage device in time periods t and t-1, P charge (t), P discharge (t) represents the charging power and discharging power of the energy storage device, η charge , η discharge It represents the discharge efficiency and discharge efficiency of the energy storage device, and Δt represents the time interval.

[0009] In step S400, the specific steps include: S401. In each optimization window, the load demand, renewable energy generation power and energy storage equipment status are updated using the latest real-time data, the electric, thermal and cooling load deviations are calculated and the objective function is corrected. The electric load deviation calculation formula is as follows: ΔP(t)=P actual (t)-P predicted (t); where P actual (t) represents the actual electric load demand during time period t, P predicted (t) represents the predicted electric load demand in time period t, and the thermal load deviation is calculated as follows: ΔQ heat (t) = Q heat,actual (t)-Q heat,predicted (t); where Q heat,actual (t) represents the actual heat load demand during time period t, Q heat,predicted(t) represents the predicted heating load demand in time period t, and the cooling load deviation is calculated as follows: ΔQ cold (t) = Q cold,actual (t)-Q cold,predicted (t); where Q cold,actual (t) represents the actual cooling load demand during period t, Q cold,predicted (t) represents the predicted cooling load demand in period t; S402, when ΔP(t) is greater than 0 or ΔQ heat (t) greater than 0 or ΔQ cold (t) is greater than 0, indicating that the load is insufficient. Increase the discharge power of the energy storage device or the heat release power of the thermal energy storage or the cooling power of the cold energy storage. When ΔP(t) is less than or equal to 0 or ΔQ heat (t) is less than or equal to 0 or ΔQ cold (t) is less than or equal to 0, indicating overload. The charging power of the energy storage equipment or the heat storage power of the hot energy storage or the cold storage power of the cold energy storage is increased. The objective function is corrected according to the load deviation. The calculation formula is as follows: J adjusted =J total +β P *|ΔP(t)|+β Q *(|ΔQ heat (t)|+|ΔQ cold (t)|); where β P and β Q Represents the adjustment coefficient, according to the modified objective function J adjusted , recalculate the energy storage equipment charging and discharging plan and the heat and cold storage power distribution.

[0010] A big data-based park energy storage intelligent dispatching system, the system includes: a data acquisition module, a prediction and analysis module, an optimization dispatching module and an adaptive adjustment module, the data acquisition module is used to collect real-time operation data of multiple types of energy systems in the park, including load data, renewable energy data and equipment operation status data, build a time series database, and clean, model and extract features of the data; the prediction and analysis module predicts short-term and long-term load changes based on energy demand scenarios, analyzes the dynamic characteristics of energy storage equipment, constructs heat, electricity and cold multi-form energy storage models, and evaluates the uncertainty and coupling relationship of loads; the optimization dispatching module is based on energy demand scenarios and The energy storage model defines a multi-objective optimization function with energy loss, renewable energy utilization rate and energy storage efficiency as targets, constructs a total energy balance model, and sets constraints on energy storage device charging and discharging and heat storage and cold storage power; the adaptive adjustment module uses real-time data to update load demand and energy storage status, calculates load deviation and corrects the objective function, and readjusts energy storage device charging and discharging and heat storage and cold storage power distribution according to the corrected objective function, the output end of the data acquisition module is connected to the input end of the prediction and analysis module, the output end of the prediction and analysis module is connected to the input end of the optimization scheduling module, and the output end of the optimization scheduling module is connected to the input end of the adaptive adjustment module.

[0011] The data acquisition module includes a load data acquisition unit, a renewable energy data acquisition unit, an equipment operation status acquisition unit and a data processing unit. The load data acquisition unit is used to collect the power demand, heating load and cooling load of the park, taking into account the impact of seasonality, diurnal periodicity and meteorological conditions on load changes. The renewable energy data acquisition unit is used to collect the real-time output and historical data of photovoltaic power generation and wind power generation, combined with solar radiation intensity, wind speed and temperature and humidity related meteorological data. The equipment operation status acquisition unit is used to collect the real-time status and charging and discharging efficiency of electric energy storage, thermal energy storage and cold energy storage equipment. The data processing unit The element is used to build a time series database based on the energy demand data collected from the terminal users of the park. The data is completed through outlier detection combined with regression analysis based on historical data and adaptive interpolation algorithm. Big data fusion technology is used to build a unified data model of electricity-heat-cold multi-energy flow according to the characteristics of electricity, heat, cooling loads and renewable energy. Subsequently, the Latin hypercube sampling method is used to combine historical data and environmental parameters to establish a mathematical model of random uncertainty. The randomness and volatility of load demand and renewable energy output are analyzed, and principal component analysis and machine learning algorithms are used to extract features of energy demand and energy storage equipment operation characteristics.

[0012] The prediction and analysis module includes a load prediction unit, an energy storage feature analysis unit and a load evaluation unit. The load prediction unit is used to predict short-term and long-term load changes based on energy demand scenarios, users' historical energy consumption behaviors and load characteristics. The energy storage feature analysis unit is used to analyze the operating characteristics of energy storage equipment and construct a thermal, electrical and cold multi-modal energy storage model. The load evaluation unit is used to evaluate the uncertainty of the load and the coupling relationship between electrical, thermal and cold loads.

[0013] The optimization scheduling module includes an objective function definition unit, an energy balance modeling unit and a constraint setting unit. The objective function definition unit is used to define a multi-objective optimization function with energy loss, renewable energy utilization rate and energy storage efficiency as the core. The energy balance calculation unit is used to construct a total energy balance model, a heat energy storage device balance model, a cold energy storage device balance model and an energy storage device state model. The constraint setting unit is used to set constraints on energy storage device charging and discharging and heat storage and cold storage power.

[0014] The adaptive adjustment module includes a target correction unit and a scheduling adjustment unit. The target correction unit is used to calculate the load deviation and correct the optimization objective function. The scheduling adjustment unit re-plans the energy storage equipment charging and discharging and heat storage and cold storage power allocation plan based on the corrected objective function.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention realizes multi-dimensional comprehensive energy scheduling analysis by constructing a unified data model of electricity-heat-cold multi-energy flow, combining the load demand data of electricity, heat energy and cold energy and the operating status of energy storage equipment. Different from the scheduling mode of a single energy form in the prior art, the present invention captures the complex coupling relationship between electricity, heat and cold, and formulates a collaborative optimization scheduling strategy.

[0016] 2. The present invention introduces a multi-objective optimization algorithm to calculate the charging and discharging plan of the energy storage equipment, the heat storage and cold storage power allocation plan, and corrects the scheduling target in real time according to the deviation calculation formula. Different from the static resource allocation method in the prior art, the present invention adaptively adjusts the allocation plan of energy storage resources according to load changes and the status of energy storage equipment, avoids energy waste or resource shortage problems, and realizes intelligent energy storage scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of a process of a park energy storage intelligent scheduling method based on big data according to the present invention; Figure 2 This is a structural schematic diagram of a park energy storage intelligent scheduling system based on big data in the present invention. DETAILED DESCRIPTION

[0018] 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.

[0019] like Figure 1-Figure 2 As shown, the present invention provides a technical solution, a park energy storage intelligent scheduling method based on big data, the method comprising the following steps: S100, collect real-time operation data of various types of energy systems in the park, including load data, renewable energy data and equipment operation status data, build a time series database, and clean, model and extract features of the data; S200, predict short-term and long-term load changes based on energy demand scenarios, analyze the dynamic characteristics of energy storage equipment, build thermal, electrical, and cold multi-modal energy storage models, and evaluate load uncertainty and coupling relationships; S300. Based on the energy demand scenario and energy storage model, define a multi-objective optimization function with energy loss, renewable energy utilization rate and energy storage efficiency as the targets, build a total energy balance model, and set constraints on the charging and discharging of energy storage equipment and the heat storage and cold storage power; S400, using real-time data to update load demand and energy storage status, calculate load deviation and correct the objective function, and readjust the energy storage equipment charging and discharging and heat storage and cold storage power distribution according to the corrected objective function.

[0020] In step S100, the specific steps include: S101. Collect energy demand data from the end users of the park, build a time series database, and monitor changes in energy demand, including load data, renewable energy data, and equipment status data. The load data includes the power demand, heating load, and cooling load of the park, taking into account the impact of seasonality, diurnal periodicity, and meteorological conditions on load changes. Renewable energy data includes real-time output and historical data of photovoltaic power generation and wind power generation, combined with meteorological data related to solar radiation intensity, wind speed, and temperature and humidity. Equipment status data includes the real-time status and charging and discharging efficiency of electric energy storage, thermal energy storage, and cold energy storage equipment. S102. Clean and model the original data, complete the data through outlier detection combined with regression analysis based on historical data and adaptive interpolation algorithm, and use big data fusion technology to build a unified data model of electricity-heat-cold multi-energy flow based on the characteristics of electricity, heat, cooling load and renewable energy. Then, use the Latin hypercube sampling method, combine historical data and environmental parameters, establish a random uncertainty mathematical model, and analyze the randomness and volatility of load demand and renewable energy output; S103. Use principal component analysis and machine learning algorithms to extract features of energy demand and energy storage equipment operation characteristics, and output energy demand scenarios under typical load and meteorological conditions and load change characteristics based on time series.

[0021] In step S200, the specific steps include: S201. Based on the energy demand scenario outputted in step S100, combined with the user's historical energy consumption behavior and load characteristics, load forecasting is performed using short-term and long-term forecasting models. A long short-term memory network model is used to predict short-term load fluctuations during peak power consumption periods. At the same time, the long-term trend of seasonal load demand is predicted based on typical meteorological conditions and energy demand scenarios. Through a deep learning model of multi-task learning, the coordinated forecasting of electricity, heat and cooling load demand is analyzed and the coupling relationship between loads is evaluated. S202. Based on the constructed unified data model of electric-heat-cold multi-energy flow and the mathematical model of random uncertainty, study the operating characteristics of heating, cooling and electric energy storage equipment and their adaptability in different energy scenarios, and quantify the volatility and uncertainty of rigid loads in different scenarios through load forecasting results; S203. Construct a multi-modal model of the energy storage system, including: analyzing the energy conversion efficiency of energy storage based on the Brayton cycle, studying the thermodynamic model of the electrical, thermal and cold coupling characteristics of the energy storage equipment, and establishing an economic model based on the round-trip efficiency, energy storage density and power density of the energy storage equipment. At the same time, analyze the dynamic characteristic model of the energy storage efficiency and output power change law of the energy storage equipment under different load rates, environmental conditions and power inputs. Finally, based on the energy demand scenario and load forecast results, combined with the operating characteristics of the energy storage equipment, generate a charging and discharging plan for the energy storage equipment.

[0022] In step S300, the specific steps include: S301. Based on the energy demand scenario generated by S100 and the energy storage model constructed by S200, a multi-objective optimization function with energy loss, renewable energy utilization rate and energy storage efficiency as the core is defined. The specific formula is defined as follows: The energy loss calculation formula is as follows: ; Among them, P total (t) represents the difference between the total power demand and the actual output in time period t, η overall Indicates the comprehensive loss rate including grid transmission and charging and discharging of energy storage equipment; The formula for calculating the utilization rate of renewable energy is as follows: ; Among them, P renewable (t) represents the total power generation of renewable energy in period t, represents the renewable energy power actually utilized during time period t; The energy storage efficiency calculation formula is as follows: ; Among them, E in (t), E out (t) represents the input and output electrical energy of the energy storage device in time period t, Q heat,actual (t), Q heat,predicted (t) represents the heat storage and release power of the thermal energy storage device in time period t, Q cold,actual (t), Q cold,predicted (t) represents the cold storage and release power of the cold energy storage device in time period t; S302, calculate the optimization objective function according to energy loss, renewable energy utilization rate and energy storage efficiency, and the calculation formula is as follows: total =J renewable / J loss +ln(1+J efficiency ); and set energy balance and energy storage equipment charging and discharging constraints. The total energy balance calculation formula is as follows: ; Among them, P demand (t) represents the total load demand in time period t, P grid (t) represents the power purchased from the power grid during period t, P storage Indicates the charging and discharging power of the energy storage device; Energy balance formula for thermal energy storage equipment: Q heat (t) = Q heat,demand (t)+Q heat,charge (t)-Q heat,discharge (t); Among them, Q heat,demand (t) represents the heat load demand during time period t; Energy balance formula for cold energy storage equipment: Q cold (t) = Q cold,demand (t)+Q cold,charge (t)-Q cold,discharge (t); Among them, Q cold,demand (t) represents the cooling load demand in period t; The state of the energy storage device is defined as follows: E(t)=E(t-1)+η charge *P charge (t)*Δt-P discharge (t)*Δt / η discharge ; Where E(t) and E(t-1) represent the remaining energy of the energy storage device in time periods t and t-1, P charge (t), P discharge(t) represents the charging power and discharging power of the energy storage device, η charge , η discharge It represents the discharge efficiency and discharge efficiency of the energy storage device, and Δt represents the time interval.

[0023] In step S400, the specific steps include: S401. In each optimization window, the load demand, renewable energy generation power and energy storage equipment status are updated using the latest real-time data, the electric, thermal and cooling load deviations are calculated and the objective function is corrected. The electric load deviation calculation formula is as follows: ΔP(t)=P actual (t)-P predicted (t); where P actual (t) represents the actual electric load demand during time period t, P predicted (t) represents the predicted electric load demand in time period t, and the thermal load deviation is calculated as follows: ΔQ heat (t) = Q heat,actual (t)-Q heat,predicted (t); where Q heat,actual (t) represents the actual heat load demand during time period t, Q heat,predicted (t) represents the predicted heating load demand in time period t, and the cooling load deviation is calculated as follows: ΔQ cold (t) = Q cold,actual (t)-Q cold,predicted (t); where Q cold,actual (t) represents the actual cooling load demand during period t, Q cold,predicted (t) represents the predicted cooling load demand in period t; S402, when ΔP(t) is greater than 0 or ΔQ heat (t) greater than 0 or ΔQ cold (t) is greater than 0, indicating that the load is insufficient. Increase the discharge power of the energy storage device or the heat release power of the thermal energy storage or the cooling power of the cold energy storage. When ΔP(t) is less than or equal to 0 or ΔQ heat (t) is less than or equal to 0 or ΔQ cold (t) is less than or equal to 0, indicating overload. The charging power of the energy storage equipment or the heat storage power of the hot energy storage or the cold storage power of the cold energy storage is increased. The objective function is corrected according to the load deviation. The calculation formula is as follows: J adjusted =J total +β P *|ΔP(t)|+β Q *(|ΔQ heat (t)|+|ΔQ cold (t)|); where β P and β Q Represents the adjustment coefficient, according to the modified objective function J adjusted, recalculate the energy storage equipment charging and discharging plan and the heat and cold storage power distribution.

[0024] A big data-based park energy storage intelligent dispatching system, the system includes: a data acquisition module, a prediction and analysis module, an optimization dispatching module and an adaptive adjustment module, the data acquisition module is used to collect real-time operation data of multiple types of energy systems in the park, including load data, renewable energy data and equipment operation status data, build a time series database, and clean, model and extract features of the data; the prediction and analysis module predicts short-term and long-term load changes based on energy demand scenarios, analyzes the dynamic characteristics of energy storage equipment, constructs heat, electricity and cold multi-form energy storage models, and evaluates the uncertainty and coupling relationship of loads; the optimization dispatching module is based on energy demand scenarios and The energy storage model defines a multi-objective optimization function with energy loss, renewable energy utilization rate and energy storage efficiency as targets, constructs a total energy balance model, and sets constraints on energy storage device charging and discharging and heat storage and cold storage power; the adaptive adjustment module uses real-time data to update load demand and energy storage status, calculates load deviation and corrects the objective function, and readjusts energy storage device charging and discharging and heat storage and cold storage power distribution according to the corrected objective function, the output end of the data acquisition module is connected to the input end of the prediction and analysis module, the output end of the prediction and analysis module is connected to the input end of the optimization scheduling module, and the output end of the optimization scheduling module is connected to the input end of the adaptive adjustment module.

[0025] The data acquisition module includes a load data acquisition unit, a renewable energy data acquisition unit, an equipment operation status acquisition unit and a data processing unit. The load data acquisition unit is used to collect the power demand, heating load and cooling load of the park, taking into account the impact of seasonality, diurnal periodicity and meteorological conditions on load changes. The renewable energy data acquisition unit is used to collect the real-time output and historical data of photovoltaic power generation and wind power generation, combined with solar radiation intensity, wind speed and temperature and humidity related meteorological data. The equipment operation status acquisition unit is used to collect the real-time status and charging and discharging efficiency of electric energy storage, thermal energy storage and cold energy storage equipment. The data processing unit It is used to build a time series database based on the energy demand data collected from the terminal users of the park, complete the data through outlier detection combined with regression analysis based on historical data and adaptive interpolation algorithm, and use big data fusion technology to build a unified data model of electricity-heat-cold multi-energy flow according to the characteristics of electricity, heat, cooling loads and renewable energy. Subsequently, the Latin hypercube sampling method is used to combine historical data and environmental parameters to establish a random uncertainty mathematical model to analyze the randomness and volatility of load demand and renewable energy output, and use principal component analysis and machine learning algorithms to extract features of energy demand and energy storage equipment operation characteristics.

[0026] The prediction and analysis module includes a load prediction unit, an energy storage feature analysis unit and a load evaluation unit. The load prediction unit is used to predict short-term and long-term load changes based on energy demand scenarios, users' historical energy consumption behaviors and load characteristics. The energy storage feature analysis unit is used to analyze the operating characteristics of energy storage equipment and build a thermal, electrical and cold multi-modal energy storage model. The load evaluation unit is used to evaluate the uncertainty of the load and the coupling relationship between electrical, thermal and cold loads.

[0027] The optimization scheduling module includes an objective function definition unit, an energy balance modeling unit and a constraint setting unit. The objective function definition unit is used to define a multi-objective optimization function with energy loss, renewable energy utilization rate and energy storage efficiency as the core. The energy balance calculation unit is used to construct a total energy balance model, a heat energy storage device balance model, a cold energy storage device balance model and an energy storage device state model. The constraint setting unit is used to set the constraints on the charging and discharging of the energy storage device and the heat and cold storage power.

[0028] The adaptive adjustment module includes a target correction unit and a scheduling adjustment unit. The target correction unit is used to calculate the load deviation and correct the optimization objective function. The scheduling adjustment unit re-plans the energy storage equipment charging and discharging and heat storage and cold storage power allocation plan based on the corrected objective function.

[0029] In the embodiment: the energy equipment information and load conditions in the park are collected, and the total installed capacity of the photovoltaic power generation system is 2MW, the total installed capacity of the wind power generation system is 1MW, the capacity of the electric energy storage equipment is 1200kMW, the charging and discharging efficiency is 92%, the maximum charging and discharging power is 500kW, the heat storage capacity of the thermal energy storage equipment is 200MWh, the maximum charging / releasing power is 5MW, the cold storage capacity of the cold energy storage equipment is 150MWh, the maximum charging / releasing power is 3MW, the maximum power purchase power of the power grid is 6MW, the peak value of the electric load is 4MW, the average value is 2.5MW, the peak value of the thermal load is 1.8MW, and the average value is 1.2M W, the peak value of cooling load is 1.5MW, the average value is 1MW, the time range is 24 hours, and each hour is a period. The energy demand data is obtained, the missing data is corrected by outlier detection, and a unified data model of electric-heat-cold multi-energy flow is constructed. Feature extraction is performed to obtain energy demand scenarios. Peak scenario: electric load 4MW, heat load 1.8MW, cooling load 1.5MW; normal scenario: electric load 2.5MW, heat load 1.2MW, cooling load 1MW. The long short-term memory network LSTM is used to predict short-term fluctuations and long-term trends of electric, heat, and cooling loads. The electric load prediction results, when time t=12, predict the electric load P predicted (t) is 3.5MW, and at time t=12, the predicted heat load Q heat,predicted (t) is 1.5MW, when time t=12, the heat load prediction Qcold,predicted (t) is 1MW, and the energy storage characteristics are analyzed. The initial value of the electric energy storage state is 60%, the charging and discharging efficiency is 92%, the initial heat storage of the thermal energy storage is 50MWh, the heat release efficiency is 95%, and the initial heat storage of the cold energy storage is 60MWh, and the heat release efficiency is 90%. The multi-objective optimization function with energy loss, renewable energy utilization rate and energy storage efficiency as the core is calculated, and the total energy balance model is constructed. The constraints of the charging and discharging and the heat and cold storage power of the energy storage equipment are calculated. By calculating the deviation, it is found that the electric load deviation is 0.1MW, the heat load deviation is 0.1MW, and the cold load deviation is 0.2MW. At this time, the electric, heat and cold loads are all greater than 0, indicating that the load is insufficient, and the discharge power, heat release power and cold release power need to be increased. The objective function is corrected according to the load deviation, and the energy storage scheduling is adjusted according to the corrected objective function.

[0030] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for intelligent dispatching of park energy storage based on big data, characterized by: The method comprises the following steps: S100, collect real-time operation data of various types of energy systems in the park, including load data, renewable energy data and equipment operation status data, build a time series database, and clean, model and extract features of the data; S200, predict short-term and long-term load changes based on energy demand scenarios, analyze the dynamic characteristics of energy storage equipment, build thermal, electrical, and cold multi-modal energy storage models, and evaluate load uncertainty and coupling relationships; S300. Based on the energy demand scenario and energy storage model, define a multi-objective optimization function with energy loss, renewable energy utilization rate and energy storage efficiency as the targets, build a total energy balance model, and set constraints on the charging and discharging of energy storage equipment and the heat and cold storage power; S400, using real-time data to update load demand and energy storage status, calculate load deviation and correct the objective function, and readjust the energy storage equipment charging and discharging and heat storage and cold storage power distribution according to the corrected objective function.

2. According to the big data-based intelligent scheduling method for park energy storage according to claim 1, it is characterized by: In step S100, the specific steps include: S101. Collect energy demand data from the end users of the park, build a time series database, and monitor changes in energy demand, including load data, renewable energy data, and equipment status data. The load data includes the power demand, heating load, and cooling load of the park, taking into account the impact of seasonality, diurnal periodicity, and meteorological conditions on load changes. Renewable energy data includes real-time output and historical data of photovoltaic power generation and wind power generation, combined with meteorological data related to solar radiation intensity, wind speed, and temperature and humidity. Equipment status data includes the real-time status and charging and discharging efficiency of electric energy storage, thermal energy storage, and cold energy storage equipment. S102. Clean and model the original data, complete the data through outlier detection combined with regression analysis based on historical data and adaptive interpolation algorithm, and use big data fusion technology to build a unified data model of electricity-heat-cold multi-energy flow based on the characteristics of electricity, heat, cooling load and renewable energy. Then, use the Latin hypercube sampling method, combine historical data and environmental parameters, establish a random uncertainty mathematical model, and analyze the randomness and volatility of load demand and renewable energy output; S103. Use principal component analysis and machine learning algorithms to extract features of energy demand and energy storage equipment operation characteristics, and output energy demand scenarios under typical load and meteorological conditions and load change characteristics based on time series.

3. The method for intelligent dispatching of energy storage in a park based on big data according to claim 1, characterized in that: In step S200, the specific steps include: S201. Based on the energy demand scenario outputted in step S100, combined with the user's historical energy consumption behavior and load characteristics, load forecasting is performed using short-term and long-term forecasting models. A long short-term memory network model is used to predict short-term load fluctuations during peak power consumption periods. At the same time, the long-term trend of seasonal load demand is predicted based on typical meteorological conditions and energy demand scenarios. Through a deep learning model of multi-task learning, the coordinated forecasting of electricity, heat and cooling load demand is analyzed and the coupling relationship between loads is evaluated. S202. Based on the constructed unified data model of electric-heat-cold multi-energy flow and the mathematical model of random uncertainty, study the operating characteristics of heating, cooling and electric energy storage equipment and their adaptability in different energy scenarios, and quantify the volatility and uncertainty of rigid loads in different scenarios through load forecasting results; S203. Construct a multi-modal model of the energy storage system, including: analyzing the energy conversion efficiency of energy storage based on the Brayton cycle, studying the thermodynamic model of the electrical, thermal and cold coupling characteristics of the energy storage equipment, and establishing an economic model based on the round-trip efficiency, energy storage density and power density of the energy storage equipment. At the same time, analyze the dynamic characteristic model of the energy storage efficiency and output power change law of the energy storage equipment under different load rates, environmental conditions and power inputs. Finally, based on the energy demand scenario and load forecast results, combined with the operating characteristics of the energy storage equipment, generate a charging and discharging plan for the energy storage equipment.

4. The method for intelligent dispatching of energy storage in a park based on big data according to claim 1, characterized in that: In step S300, the specific steps include: S301. Based on the energy demand scenario generated by S100 and the energy storage model constructed by S200, a multi-objective optimization function with energy loss, renewable energy utilization rate and energy storage efficiency as the core is defined. The specific formula is defined as follows: The energy loss calculation formula is as follows: ； Among them, P total (t) represents the difference between the total power demand and the actual output in time period t, η overall Indicates the comprehensive loss rate including grid transmission and charging and discharging of energy storage equipment; The formula for calculating the utilization rate of renewable energy is as follows: ； Among them, P renewable (t) represents the total power generation of renewable energy in period t, represents the renewable energy power actually utilized during time period t; The energy storage efficiency calculation formula is as follows: ； Among them, E in (t), E out (t) represents the input and output electrical energy of the energy storage device in time period t, Q heat,actual (t), Q heat,predicted (t) represents the heat storage and release power of the thermal energy storage device in time period t, Q cold,actual (t), Q cold,predicted (t) represents the cold storage and release power of the cold energy storage device in time period t; S302, calculate the optimization objective function according to energy loss, renewable energy utilization rate and energy storage efficiency, and the calculation formula is as follows: total =J renewable / J loss +ln(1+J efficiency ); and set energy balance and energy storage equipment charging and discharging constraints. The total energy balance calculation formula is as follows: ； Among them, P demand (t) represents the total load demand in time period t, P grid (t) represents the power purchased from the power grid during period t, P storage Indicates the charging and discharging power of the energy storage device; Energy balance formula for thermal energy storage equipment: Q heat (t) = Q heat,demand (t)+Q heat,charge (t)-Q heat,discharge (t); Among them, Q heat,demand (t) represents the heat load demand during time period t; Energy balance formula for cold energy storage equipment: Q cold (t) = Q cold,demand (t)+Q cold,charge (t)-Q cold,discharge (t); Among them, Q cold,demand (t) represents the cooling load demand in period t; The state of the energy storage device is defined as follows: E(t)=E(t-1)+η charge *P charge (t)*Δt-P discharge (t)*Δt / η discharge ; Where E(t) and E(t-1) represent the remaining energy of the energy storage device in time periods t and t-1, P charge (t), P discharge (t) represents the charging power and discharging power of the energy storage device, η charge , η discharge It represents the discharge efficiency and discharge efficiency of the energy storage device, and Δt represents the time interval.

5. The method for intelligent dispatching of energy storage in a park based on big data according to claim 1, characterized in that: In step S400, the specific steps include: S401. In each optimization window, the load demand, renewable energy generation power and energy storage equipment status are updated using the latest real-time data, the electric, thermal and cooling load deviations are calculated and the objective function is corrected. The electric load deviation calculation formula is as follows: ΔP(t)=P actual (t)-P predicted (t); where P actual (t) represents the actual electric load demand during time period t, P predicted (t) represents the predicted electric load demand in time period t, and the thermal load deviation is calculated as follows: ΔQ heat (t) = Q heat,actual (t)-Q heat,predicted (t); where Q heat,actual (t) represents the actual heat load demand during time period t, Q heat,predicted (t) represents the predicted heating load demand in time period t, and the cooling load deviation is calculated as follows: ΔQ cold (t) = Q cold,actual (t)-Q cold,predicted (t); where Q cold,actual (t) represents the actual cooling load demand during period t, Q cold,predicted (t) represents the predicted cooling load demand in period t; S402, when ΔP(t) is greater than 0 or ΔQ heat (t) greater than 0 or ΔQ cold (t) is greater than 0, indicating that the load is insufficient. Increase the discharge power of the energy storage device or the heat release power of the thermal energy storage or the cooling power of the cold energy storage. When ΔP(t) is less than or equal to 0 or ΔQ heat (t) is less than or equal to 0 or ΔQ cold (t) is less than or equal to 0, indicating overload. The charging power of the energy storage equipment or the heat storage power of the hot energy storage or the cold storage power of the cold energy storage is increased. The objective function is corrected according to the load deviation. The calculation formula is as follows: J adjusted =J total +β P *|ΔP(t)|+β Q *(|ΔQ heat (t)|+|ΔQ cold (t)|); where β P and β Q Represents the adjustment coefficient, according to the modified objective function J adjusted , recalculate the energy storage equipment charging and discharging plan and the heat and cold storage power distribution.

6. A park energy storage intelligent dispatching system based on big data, characterized by: The system includes: a data acquisition module, a prediction and analysis module, an optimization and scheduling module and an adaptive adjustment module. The data acquisition module is used to collect real-time operation data of various types of energy systems in the park, including load data, renewable energy data and equipment operation status data, build a time series database, and clean, model and extract features of the data; the prediction and analysis module predicts short-term and long-term load changes based on energy demand scenarios, analyzes the dynamic characteristics of energy storage equipment, builds a multi-form energy storage model of heat, electricity and cold, and evaluates the uncertainty and coupling relationship of loads; the optimization and scheduling module defines energy loss based on energy demand scenarios and energy storage models. A multi-objective optimization function with energy consumption, renewable energy utilization rate and energy storage efficiency as targets is used to construct a total energy balance model, and constraints on energy storage device charging and discharging and heat storage and cold storage power are set; the adaptive adjustment module uses real-time data to update load demand and energy storage status, calculates load deviation and corrects the objective function, and readjusts energy storage device charging and discharging and heat storage and cold storage power distribution according to the corrected objective function, the output end of the data acquisition module is connected to the input end of the prediction and analysis module, the output end of the prediction and analysis module is connected to the input end of the optimization scheduling module, and the output end of the optimization scheduling module is connected to the input end of the adaptive adjustment module.

7. The big data-based park energy storage intelligent dispatching system according to claim 6 is characterized by: The data acquisition module includes a load data acquisition unit, a renewable energy data acquisition unit, an equipment operation status acquisition unit and a data processing unit. The load data acquisition unit is used to collect the power demand, heating load and cooling load of the park, taking into account the impact of seasonality, diurnal periodicity and meteorological conditions on load changes. The renewable energy data acquisition unit is used to collect the real-time output and historical data of photovoltaic power generation and wind power generation, combined with solar radiation intensity, wind speed and temperature and humidity related meteorological data. The equipment operation status acquisition unit is used to collect the real-time status and charging and discharging efficiency of electric energy storage, thermal energy storage and cold energy storage equipment. The data processing unit The element is used to build a time series database based on the energy demand data collected from the terminal users of the park. The data is completed through outlier detection combined with regression analysis based on historical data and adaptive interpolation algorithm. Big data fusion technology is used to build a unified data model of electricity-heat-cold multi-energy flow according to the characteristics of electricity, heat, cooling loads and renewable energy. Subsequently, the Latin hypercube sampling method is used to combine historical data and environmental parameters to establish a mathematical model of random uncertainty. The randomness and volatility of load demand and renewable energy output are analyzed, and principal component analysis and machine learning algorithms are used to extract features of energy demand and energy storage equipment operation characteristics.

8. The park energy storage intelligent dispatching system based on big data according to claim 6 is characterized by: The prediction and analysis module includes a load prediction unit, an energy storage feature analysis unit and a load evaluation unit. The load prediction unit is used to predict short-term and long-term load changes based on energy demand scenarios, users' historical energy consumption behaviors and load characteristics. The energy storage feature analysis unit is used to analyze the operating characteristics of energy storage equipment and construct a thermal, electrical and cold multi-modal energy storage model. The load evaluation unit is used to evaluate the uncertainty of the load and the coupling relationship between electrical, thermal and cold loads.

9. The park energy storage intelligent dispatching system based on big data according to claim 6 is characterized by: The optimization scheduling module includes an objective function definition unit, an energy balance modeling unit and a constraint setting unit. The objective function definition unit is used to define a multi-objective optimization function with energy loss, renewable energy utilization rate and energy storage efficiency as the core. The energy balance calculation unit is used to construct a total energy balance model, a heat energy storage device balance model, a cold energy storage device balance model and an energy storage device state model. The constraint setting unit is used to set constraints on energy storage device charging and discharging and heat storage and cold storage power.

10. The park energy storage intelligent dispatching system based on big data according to claim 6, characterized in that: The adaptive adjustment module includes a target correction unit and a scheduling adjustment unit. The target correction unit is used to calculate the load deviation and correct the optimization objective function. The scheduling adjustment unit is used to re-plan the energy storage equipment charging and discharging and heat storage and cold storage power allocation plan based on the corrected objective function.

Images