Virtual power plant interaction model construction method

Abstract

The invention discloses a virtual power plant interaction model construction method. The virtual power plant interaction model construction method comprises the following steps: 1, data collection and preprocessing; 2, constructing an interaction model; 3, model integration and interaction; 4, verifying and optimizing the model; and 5, deployment and real-time monitoring. According to the method, the new energy output model, the load model and the energy storage model are constructed, and historical data are used for training and verification, so that the prediction accuracy of future energy output, power demand and energy storage state can be improved; the virtual power plant can realize real-time interaction and response among the new energy, the load and the energy storage system.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twin model construction, and more specifically to a method for constructing a virtual power plant interactive model. Background Art

[0002] With the large-scale access of renewable energy and the gradual opening of the power market, the power system is facing more and more challenges. As a smart energy management system, virtual power plants can effectively aggregate and optimize various distributed energy resources and improve the flexibility, reliability and economy of the power system, so they have attracted much attention.

[0003] A virtual power plant is a power coordination and management system that uses advanced information and communication technologies and software systems to aggregate and coordinate distributed energy resources (DER) to participate in the electricity market and power grid operations as a special power plant.

[0004] The existing virtual power plant interaction model cannot accurately predict future energy output, power demand and energy storage status, which leads to deviations in energy allocation and optimization. Therefore, a new technical solution is needed to solve this problem. Summary of the invention

[0005] The purpose of the present invention is to provide a method for constructing a virtual power plant interaction model, which solves the problem that the existing virtual power plant interaction model cannot accurately predict future energy output, power demand and energy storage status, thereby causing deviations in energy allocation and optimization.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for constructing a virtual power plant interaction model, the method comprising the following steps:

[0007] Step 1: Data collection and preprocessing: Collect historical output data of new energy sources, historical load data, and relevant data of energy storage systems, clean and preprocess the collected data, remove outliers, process missing data, and ensure data quality and consistency;

[0008] Step 2: Construction of interactive model, which includes new energy output model, load model and energy storage model;

[0009] Step 3: Model integration and interaction: Integrate the constructed new energy output model, load model and energy storage model to form a unified virtual power plant interaction model. Set up an interaction mechanism in the model to enable real-time interaction and response between new energy output, load demand and energy storage status. When the new energy output increases, the model should be able to automatically adjust the charging and discharging strategy of the energy storage system to meet the load demand while achieving the optimal energy distribution.

[0010] Step 4: Model verification and optimization: Use a portion of data that is not involved in model construction to verify the virtual power plant interaction model, evaluate the model's prediction accuracy and performance, and optimize and adjust the model based on the verification results to improve the model's prediction accuracy and adaptability;

[0011] Step 5: Deployment and real-time monitoring: deploy the constructed virtual power plant interaction model to the actual system, realize real-time interaction with the actual power grid, monitor and maintain the model in real time, and ensure the stable operation of the model and real-time updating of data.

[0012] As a preferred embodiment of the present invention, the historical output data in step 1 includes solar energy, water source heat pumps and ground source heat pumps, including hourly or minute output, light intensity, water source temperature and soil temperature. The historical load data includes daily, weekly, monthly and annual load changes, as well as the time and regularity of load peaks and valleys. The relevant data of the energy storage system include battery charging and discharging rates, battery SOC. The preprocessing performs missing value processing on the data, uses interpolation or statistical-based methods to fill in missing values, performs data smoothing, eliminates noise and outliers, and can use sliding average or median filtering to perform data normalization, converting all data to the same dimension and range for model training and calculation.

[0013] As a preferred implementation mode of the present invention, the new energy output model in step 2 extracts features of the new energy output data, including time features and meteorological features, selects one of the random forest, gradient boosting regression tree or deep learning model for training, and obtains a new energy output prediction model. The load model extracts features of the load data, including time features, meteorological features and economic features, and constructs a load prediction model according to the importance and correlation of the features. The energy storage model analyzes the characteristics of the energy storage system, including battery capacity, charging and discharging efficiency, and charging and discharging rate. According to historical data and the characteristics of the energy storage system, an energy storage prediction model is constructed to predict the charging and discharging state and energy storage of the energy storage system.

[0014] As a preferred implementation mode of the present invention, the model integration and interaction in step 3 designs the interaction rules and strategies between the new energy output, load demand and energy storage status, and automatically adjusts the charging and discharging strategy of the energy storage system according to the predicted new energy output, load demand and energy storage status to achieve the optimal distribution of energy.

[0015] As a preferred embodiment of the present invention, the model verification and optimization in step 4 uses an independent data set that does not participate in the model construction to perform model verification, calculate the error index of the prediction result, analyze the gap and cause between the prediction result and the actual data, evaluate the prediction accuracy and performance of the model, and adjust the model parameters or optimize the algorithm according to the verification results to improve the prediction accuracy and stability of the model.

[0016] As a preferred implementation mode of the present invention, the deployment and real-time monitoring in step 5 deploys the optimized virtual power plant interaction model into the actual system, integrates it with the power grid dispatching system and control system, configures data transmission and communication interfaces, and ensures that the model can receive and process power grid operation data in real time; monitors the operating status and prediction results of the virtual power plant interaction model in real time to ensure the stability and accuracy of the model, and regularly collects and analyzes actual operation data for updating and improving the model.

[0017] As a preferred embodiment of the present invention, the new energy output model includes a photovoltaic module output model, a water source heat pump output model and a ground source heat pump output model;

[0018] The algorithm formula of the photovoltaic module output model is as follows:

[0019] Calculation of photovoltaic module output power:

[0020] P=S×η cell ×η sys ×G

[0021] P is the output power of the photovoltaic module (W),

[0022] S is the area of ​​the photovoltaic module (m 2 ),

[0023] η cell is the conversion efficiency of the photovoltaic cell (dimensionless),

[0024] η sys is the efficiency of the photovoltaic system, taking into account the overall efficiency of the system including factors such as inverter efficiency and line losses (dimensionless),

[0025] G is the solar radiation intensity (W / m 2 ), in actual applications, it will vary according to factors such as geographical location, season, and weather;

[0026] Calculation of daily power generation of photovoltaic modules:

[0027] E=Ppeak×H

[0028] E is the daily electricity generation (kWh),

[0029] P peakis the peak power of the PV module (kWp),

[0030] H is the peak sunshine hours (h / day), which represents the equivalent number of hours in a day that the standard test conditions of the photovoltaic modules are met;

[0031] Calculation of annual power generation of photovoltaic modules:

[0032] Eyear=E×365

[0033] E year is the annual electricity generation (kWh / year), assuming the same sunshine conditions every day;

[0034] Calculation of photovoltaic system power generation efficiency:

[0035]

[0036] η s is the photovoltaic system power generation efficiency,

[0037] P is the electricity generated by the photovoltaic system (kWh),

[0038] Ap is the surface area of ​​the PV module (m2)

[0039] The algorithm formula of the water source heat pump output model is as follows:

[0040]

[0041] Q represents the cooling or heating capacity of the heat pump, usually in kilowatts (kW) or kilocalories per hour (kcal / h), C op is the energy efficiency ratio of the heat pump, that is, the ratio of heating or cooling capacity to input electrical energy, dimensionless,

[0042] is the mass flow rate of water in kilograms per second (kg / s) or kilograms per hour (kg / h),

[0043] ΔT is the temperature difference between the inlet and outlet of water, in degrees Celsius (℃);

[0044] The algorithm formula of the ground source heat pump output model is as follows:

[0045] Assuming that the cooling load is large, taking the cooling mode as an example, the heat exchange capacity Q of the buried pipe is calculated:

[0046]

[0047] Among them, COP cool It is the coefficient of performance under refrigeration conditions, indicating the cooling capacity (kW) that can be provided without consuming 1 kilowatt of electrical energy;

[0048] If the heat load is large, take the heating mode as an example to calculate the heat exchange capacity Q of the buried pipe:

[0049]

[0050] Among them, COP heat It is the performance coefficient under heating conditions, indicating the amount of heat (kW) that can be provided without consuming 1 kilowatt of electrical energy;

[0051] As a preferred embodiment of the present invention, the load model includes an overall load forecasting model, a renewable energy generation model and a load elasticity model;

[0052] The algorithm formula of the overall load forecasting model is as follows:

[0053]

[0054] L total (t) is the predicted total load of the virtual power plant (MW) at time t,

[0055] L i (t) is the predicted load of the ith distributed energy source or load at time t,

[0056] L flex (t) is the flexibility load provided by the adjustable load at time t;

[0057] The algorithm formula of the renewable energy power generation model is as follows:

[0058] P rencwable (t) = C capacoty ×R availability (t)

[0059] P rencwable (t) is the power generated by renewable energy at time t (MW),

[0060] C capacoty is the installed capacity (MW),

[0061] R availability (t) is the resource availability factor, which takes into account weather conditions, geographical location, etc., and is usually a value between 0 and 1;

[0062] The algorithm formula of the load elasticity model is as follows:

[0063] L flex (t) = B × (L base (t)-L setpoint (t))

[0064] L flex (t) is the load flexibility achieved through DR,

[0065] B is the baseline load.

[0066] L base (t) is the expected load without DR intervention,

[0067] L setpoint (t) is the load target set by the DR program;

[0068] As a preferred implementation of the present invention, the algorithm formula of the energy storage model is as follows:

[0069]

[0070] dSOC(t) is the state of the energy storage system at time t, usually expressed as a percentage,

[0071] P charge (t) the power charged to the energy storage system at time t (positive value),

[0072] P discharge (t) the power discharged from the energy storage system at time t (negative value),

[0073] E rated The rated energy capacity of the energy storage system, used to normalize the rate of change of SOC.

[0074] Compared with the prior art, the present invention has the following beneficial effects:

[0075] The present invention can improve the accuracy of prediction of future energy output, power demand and energy storage status by constructing new energy output model, load model and energy storage model, and using historical data for training and verification. Through model integration and interaction, the virtual power plant can realize real-time interaction and response between new energy, load and energy storage system. This enables the virtual power plant to adjust energy allocation strategy more flexibly, optimize energy use, improve energy utilization and stability of power grid operation, and adopt appropriate data encryption, access control and privacy protection technology to ensure the data security of the virtual power plant and prevent unauthorized access and data leakage. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 A schematic diagram of the process of constructing the virtual power plant interaction model of the present invention. DETAILED DESCRIPTION

[0077] 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.

[0078] The present invention provides a technical solution: a method for constructing a virtual power plant interaction model, the method comprising the following steps:

[0079] Step 1: Data collection and preprocessing: Collect historical output data of new energy sources, historical load data, and relevant data of energy storage systems, clean and preprocess the collected data, remove outliers, process missing data, and ensure data quality and consistency;

[0080] Step 2: Construction of interactive model, which includes new energy output model, load model and energy storage model;

[0081] Step 3: Model integration and interaction: Integrate the constructed new energy output model, load model and energy storage model to form a unified virtual power plant interaction model. Set up an interaction mechanism in the model to enable real-time interaction and response between new energy output, load demand and energy storage status. When the new energy output increases, the model should be able to automatically adjust the charging and discharging strategy of the energy storage system to meet the load demand while achieving the optimal energy distribution.

[0082] Step 4: Model verification and optimization: Use a portion of data that is not involved in model construction to verify the virtual power plant interaction model, evaluate the model's prediction accuracy and performance, and optimize and adjust the model based on the verification results to improve the model's prediction accuracy and adaptability;

[0083] Step 5: Deployment and real-time monitoring: deploy the constructed virtual power plant interaction model to the actual system, realize real-time interaction with the actual power grid, monitor and maintain the model in real time, and ensure the stable operation of the model and real-time updating of data.

[0084] Further improved, the historical output data in step 1 includes solar energy, water source heat pumps and ground source heat pumps, including hourly or minute output, light intensity, water source temperature and soil temperature; the historical load data includes daily, weekly, monthly and annual load changes, as well as the time and regularity of load peaks and valleys; the relevant data of the energy storage system includes the battery charging and discharging rates, and the battery SOC; the data is preprocessed to process missing values, fill in missing values ​​using interpolation or statistical-based methods, perform data smoothing, eliminate noise and outliers, and use sliding average or median filtering to normalize the data, converting all data to the same dimension and range for model training and calculation.

[0085] Further improved, the new energy output model in step 2 extracts features of the new energy output data, including time features and meteorological features, selects one of the random forest, gradient boosting regression tree or deep learning model for training, and obtains a new energy output prediction model. The load model extracts features of the load data, including time features, meteorological features and economic features, and constructs a load prediction model according to the importance and correlation of the features. The energy storage model analyzes the characteristics of the energy storage system, including battery capacity, charging and discharging efficiency, and charging and discharging rate. According to historical data and the characteristics of the energy storage system, an energy storage prediction model is constructed to predict the charging and discharging state and energy storage of the energy storage system.

[0086] Further improvement is made in that the model integration and interaction in step 3 designs interaction rules and strategies among new energy output, load demand and energy storage status, and automatically adjusts the charging and discharging strategy of the energy storage system according to the predicted new energy output, load demand and energy storage status to achieve optimal energy distribution.

[0087] Further improved, the model verification and optimization in step 4 uses an independent data set that does not participate in the model construction to verify the model, calculate the error index of the prediction result, analyze the gap and cause between the prediction result and the actual data, evaluate the prediction accuracy and performance of the model, and adjust the model parameters or optimize the algorithm according to the verification results to improve the prediction accuracy and stability of the model.

[0088] Further improved, the deployment and real-time monitoring in step 5 deploy the optimized virtual power plant interaction model into the actual system, integrate it with the power grid dispatching system and control system, configure data transmission and communication interfaces, and ensure that the model can receive and process power grid operation data in real time; monitor the operating status and prediction results of the virtual power plant interaction model in real time to ensure the stability and accuracy of the model, and regularly collect and analyze actual operation data for model updating and improvement.

[0089] Further improved, the new energy output model includes a photovoltaic module output model, a water source heat pump output model and a ground source heat pump output model;

[0090] The algorithm formula of the photovoltaic module output model is as follows:

[0091] Calculation of photovoltaic module output power:

[0092] P=S×η cell ×η sys ×G

[0093] P is the output power of the photovoltaic module (W),

[0094] S is the area of ​​the photovoltaic module (m 2 ),

[0095] η cell is the conversion efficiency of the photovoltaic cell (dimensionless),

[0096] η sys is the efficiency of the photovoltaic system, taking into account the overall efficiency of the system including factors such as inverter efficiency and line losses (dimensionless),

[0097] G is the solar radiation intensity (W / m 2 ), in actual applications, it will vary according to factors such as geographical location, season, and weather;

[0098] Calculation of daily power generation of photovoltaic modules:

[0099] E=Ppeak×H

[0100] E is the daily electricity generation (kWh),

[0101] P peak is the peak power of the PV module (kWp),

[0102] H is the peak sunshine hours (h / day), which represents the equivalent number of hours in a day that the standard test conditions of the photovoltaic modules are met;

[0103] Calculation of annual power generation of photovoltaic modules:

[0104] Eyear=E×365

[0105] E year is the annual electricity generation (kWh / year), assuming the same sunshine conditions every day;

[0106] Calculation of photovoltaic system power generation efficiency:

[0107]

[0108] η s is the photovoltaic system power generation efficiency,

[0109] P is the electricity generated by the photovoltaic system (kWh),

[0110] Ap is the surface area of ​​the PV module (m2)

[0111] The algorithm formula of the water source heat pump output model is as follows:

[0112]

[0113] Q represents the cooling or heating capacity of the heat pump, usually in kilowatts (kW) or kilocalories per hour (kcal / h), C op is the energy efficiency ratio of the heat pump, that is, the ratio of heating or cooling capacity to input electrical energy, dimensionless,

[0114] is the mass flow rate of water in kilograms per second (kg / s) or kilograms per hour (kg / h),

[0115] ΔT is the temperature difference between the inlet and outlet of water, in degrees Celsius (℃);

[0116] The algorithm formula of the ground source heat pump output model is as follows:

[0117] Assuming that the cooling load is large, taking the cooling mode as an example, the heat exchange capacity Q of the buried pipe is calculated:

[0118]

[0119] Among them, COP cool It is the coefficient of performance under refrigeration conditions, indicating the cooling capacity (kW) that can be provided without consuming 1 kilowatt of electrical energy;

[0120] If the heat load is large, take the heating mode as an example to calculate the heat exchange capacity Q of the buried pipe:

[0121]

[0122] Among them, COP heat It is the performance coefficient under heating conditions, indicating the amount of heat (kW) that can be provided without consuming 1 kilowatt of electrical energy;

[0123] Further improved, the load model includes an overall load forecasting model, a renewable energy generation model and a load elasticity model;

[0124] The algorithm formula of the overall load forecasting model is as follows:

[0125]

[0126] L total (t) is the predicted total load of the virtual power plant (MW) at time t,

[0127] L i (t) is the predicted load of the ith distributed energy source or load at time t,

[0128] L flex (t) is the flexibility load provided by the adjustable load at time t;

[0129] The algorithm formula of the renewable energy power generation model is as follows:

[0130] P rencwable (t) = C capacoty ×R availability (t)

[0131] P rencwable(t) is the power generated by renewable energy at time t (MW),

[0132] C capacoty is the installed capacity (MW),

[0133] R availability (t) is the resource availability factor, which takes into account weather conditions, geographical location, etc., and is usually a value between 0 and 1;

[0134] The algorithm formula of the load elasticity model is as follows:

[0135] L flex (t) = B × (L base (t)-L setpoint (t))

[0136] L flex (t) is the load flexibility achieved through DR,

[0137] B is the baseline load.

[0138] L base (t) is the expected load without DR intervention,

[0139] L setpoint (t) is the load target set by the DR program;

[0140] Further improved, the algorithm formula of the energy storage model is as follows:

[0141]

[0142] dSOC(t) is the state of the energy storage system at time t, usually expressed as a percentage,

[0143] P charge (t) the power charged to the energy storage system at time t (positive value),

[0144] P discharge (t) the power discharged from the energy storage system at time t (negative value),

[0145] E rated The rated energy capacity of the energy storage system, used to normalize the rate of change of SOC.

[0146] Compared with the prior art, the present invention has the following beneficial effects:

[0147] The present invention can improve the accuracy of prediction of future energy output, power demand and energy storage status by constructing new energy output model, load model and energy storage model, and using historical data for training and verification. Through model integration and interaction, the virtual power plant can realize real-time interaction and response between new energy, load and energy storage system. This enables the virtual power plant to adjust energy allocation strategy more flexibly, optimize energy use, improve energy utilization and stability of power grid operation, and adopt appropriate data encryption, access control and privacy protection technology to ensure the data security of the virtual power plant and prevent unauthorized access and data leakage.

[0148] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0149] Finally, a few points should be explained: First, in the description of this application, it should be noted that, unless otherwise specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, and can be mechanical or electrical connections, or internal connectivity between two components, or direct connections. "Up", "down", "left", "right", etc. are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may change.

[0150] 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 method for constructing a virtual power plant interactive model, characterized in that: The method for constructing a virtual power plant interactive model includes the following steps: Step 1: Data collection and preprocessing: Collect historical output data of new energy sources, historical load data, and relevant data of energy storage systems, clean and preprocess the collected data, remove outliers, process missing data, and ensure data quality and consistency; Step 2: Construction of interactive model, which includes new energy output model, load model and energy storage model; Step 3: Model integration and interaction: Integrate the constructed new energy output model, load model and energy storage model to form a unified virtual power plant interaction model. Set up an interaction mechanism in the model to enable real-time interaction and response between new energy output, load demand and energy storage status. When the new energy output increases, the model should be able to automatically adjust the charging and discharging strategy of the energy storage system to meet the load demand while achieving the optimal energy distribution. Step 4: Model verification and optimization: Use a portion of data that is not involved in model construction to verify the virtual power plant interaction model, evaluate the model's prediction accuracy and performance, and optimize and adjust the model based on the verification results to improve the model's prediction accuracy and adaptability; Step 5: Deployment and real-time monitoring: deploy the constructed virtual power plant interaction model to the actual system, realize real-time interaction with the actual power grid, monitor and maintain the model in real time, and ensure the stable operation of the model and real-time updating of data.

2. A method for constructing a virtual power plant interactive model according to claim 1, characterized in that: The historical output data in step 1 include solar energy, water source heat pumps and ground source heat pumps, including hourly or minute output, light intensity, water source temperature and soil temperature. The historical load data include daily, weekly, monthly and annual load changes, as well as the time and regularity of load peaks and valleys. The relevant data of the energy storage system include battery charging and discharging rates, and battery SOC. Preprocessing is performed on the data to process missing values, fill in missing values ​​using interpolation or statistical methods, perform data smoothing, eliminate noise and outliers, and use sliding average or median filtering methods to normalize the data, converting all data to the same dimension and range for model training and calculation.

3. The method for constructing a virtual power plant interactive model according to claim 1, characterized in that: The new energy output model in step 2 extracts features from the new energy output data, including time features and meteorological features, and selects one of the random forest, gradient boosting regression tree or deep learning model for training to obtain a new energy output prediction model. The load model extracts features from the load data, including time features, meteorological features and economic features, and constructs a load prediction model according to the importance and relevance of the features. The energy storage model analyzes the characteristics of the energy storage system, including battery capacity, charging and discharging efficiency, and charging and discharging rate. According to historical data and the characteristics of the energy storage system, an energy storage prediction model is constructed to predict the charging and discharging state and energy storage of the energy storage system.

4. The method for constructing a virtual power plant interactive model according to claim 1, characterized in that: The model integration and interaction in step 3 designs the interaction rules and strategies between the new energy output, load demand and energy storage status, and automatically adjusts the charging and discharging strategy of the energy storage system according to the predicted new energy output, load demand and energy storage status to achieve the optimal distribution of energy.

5. The method for constructing a virtual power plant interactive model according to claim 1, characterized in that: The model verification and optimization in step 4 uses an independent data set that is not involved in the model construction to verify the model, calculate the error index of the prediction result, analyze the gap and cause between the prediction result and the actual data, evaluate the prediction accuracy and performance of the model, and adjust the model parameters or optimize the algorithm according to the verification results to improve the prediction accuracy and stability of the model.

6. A method for constructing a virtual power plant interactive model according to claim 1, characterized in that: The deployment and real-time monitoring in step 5 deploys the optimized virtual power plant interaction model into the actual system, integrates it with the power grid dispatching system and control system, configures data transmission and communication interfaces, and ensures that the model can receive and process power grid operation data in real time; monitors the operating status and prediction results of the virtual power plant interaction model in real time to ensure the stability and accuracy of the model, and regularly collects and analyzes actual operation data for model updating and improvement.

7. A method for constructing a virtual power plant interactive model according to claim 3, characterized in that: The new energy output model includes a photovoltaic module output model, a water source heat pump output model and a ground source heat pump output model; The algorithm formula of the photovoltaic module output model is as follows: Calculation of photovoltaic module output power: P=S×η cell ×η sys ×G P is the output power of the photovoltaic module (W), S is the area of ​​the photovoltaic module (m 2 ), η ce ll is the conversion efficiency of the photovoltaic cell (dimensionless), η sys is the efficiency of the photovoltaic system, taking into account the overall efficiency of the system including factors such as inverter efficiency and line losses (dimensionless), G is the solar radiation intensity (W / m 2 ), in actual applications, it will vary according to factors such as geographical location, season, and weather; Calculation of daily power generation of photovoltaic modules: E=Ppeak×H E is the daily electricity generation (kWh), P peak is the peak power of the PV module (kWp), H is the peak sunshine hours (h / day), which represents the equivalent number of hours in a day that the standard test conditions of the photovoltaic modules are met; Calculation of annual power generation of photovoltaic modules: Eyear=E×365 E year is the annual electricity generation (kWh / year), assuming the same sunshine conditions every day; Calculation of photovoltaic system power generation efficiency: η s is the photovoltaic system power generation efficiency, P is the electricity generated by the photovoltaic system (kWh), Ap is the surface area of ​​the PV module (m2) The algorithm formula of the water source heat pump output model is as follows: Q represents the cooling or heating capacity of the heat pump, usually in kilowatts (kW) or kilocalories per hour (kcal / h). C op is the energy efficiency ratio of the heat pump, that is, the ratio of heating or cooling capacity to input electrical energy, dimensionless, is the mass flow rate of water in kilograms per second (kg / s) or kilograms per hour (kg / h), ΔT is the temperature difference between the inlet and outlet of water, in degrees Celsius (℃); The algorithm formula of the ground source heat pump output model is as follows: Assuming that the cooling load is large, taking the cooling mode as an example, the heat exchange capacity Q of the buried pipe is calculated: Among them, COP cool It is the coefficient of performance under refrigeration conditions, indicating the cooling capacity (kW) that can be provided without consuming 1 kilowatt of electrical energy; If the heat load is large, take the heating mode as an example to calculate the heat exchange capacity Q of the buried pipe: Among them, COP heat It is the performance coefficient under heating conditions, which indicates the amount of heat (kW) that can be provided without consuming 1 kilowatt of electrical energy.

8. A method for constructing a virtual power plant interactive model according to claim 3, characterized in that: The load model includes an overall load forecasting model, a renewable energy generation model and a load elasticity model; The algorithm formula of the overall load forecasting model is as follows: L total (t) is the predicted total load of the virtual power plant (MW) at time t, L i (t) is the forecast load of the ith distributed energy source or load at time t, L flex (t) is the flexibility load provided by the adjustable load (e.g., demand response resources) at time t; The algorithm formula of the renewable energy power generation model is as follows: P rencwable (t)=C capacoty ×R availability (t) P rencwable (t) is the power generated by renewable energy at time t (MW), C capacoty is the installed capacity (MW), R availability (t) is the resource availability factor, which takes into account weather conditions, geographical location, etc., and is usually a value between 0 and 1; The algorithm formula of the load elasticity model is as follows: L flex (t)=B×(L base (t)-L setpoint (t)) L flex (t) is the load flexibility achieved through DR, B is the baseline load. L base (t) is the expected load without DR intervention, L setpoint (t) is the load target set by the DR program.

9. A method for constructing a virtual power plant interactive model according to claim 3, characterized in that: The algorithm formula of the energy storage model is as follows: dSOC(t) is the state of the energy storage system at time t, usually expressed as a percentage, P charge (t) the power charged to the energy storage system at time t (positive value), P discharge (t) the power discharged from the energy storage system at time t (negative value), E rated The rated energy capacity of the energy storage system, used to normalize the rate of change of SOC.

Images