Virtual power plant dynamic peak shaving method and system based on neural network
Through the dynamic peak shaving method based on neural network, the controllable power generation resources of virtual power plants are analyzed, and a multi-level optimization scheduling model is constructed, which solves the problems of insufficient and uncontrollable peak shaving capabilities in the dynamic peak shaving of virtual power plants, and achieves a more stable The grid operation and load curve are stable.
Patent Information
- Application Number
- CN202410806813.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-21
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-06-21
AI Technical Summary
During the dynamic peak shaving process, virtual power plants have insufficient trough peak shaving capabilities, and the scheduling operation is uncontrollable due to uncertain power supply and load prediction errors.
Using a dynamic peak shaking method based on neural networks, we obtain system scheduling instructions in real time, analyze controllable resource data of power generation, establish adjustable power model of power generation resources, gather the characteristics of power resources, build multi-level optimization scheduling data, and introduce dynamic constraints. Build a output scheduling model to obtain real-time power dynamic peak shaving data.
The peak shaving capability of the power grid is improved, the stability and controllability of the power grid operation are ensured, and the load variance is reduced through energy storage output scheduling, the prediction error is corrected, and a smoother load curve is achieved.
Smart Images

Figure CN118898352A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of dynamic peak regulation of virtual power plants, and in particular to a method and system for dynamic peak regulation of virtual power plants based on neural networks. Background Art
[0002] As a power coordination and management system, virtual power plants can combine various types of clean energy, controllable loads and energy storage systems installed in the distribution network as a special power plant to participate in power grid operation and power market transactions, thereby effectively reducing the difficulty of power grid dispatching, improving the absorption of renewable energy, and ensuring the safe and stable operation of the power system. However, in the actual operation process, virtual power plants currently focus on economic benefits and the absorption of renewable energy, and fail to give full play to their potential for low-carbon emission reduction. Carbon emission rights are traded in the market nationwide as commodities. This market-based carbon trading mechanism requires carbon emissions to consider carbon costs, and the carbon reduction benefits of renewable energy are quantified, which can effectively stimulate the enthusiasm and potential of each carbon emission unit to reduce emissions. This will inevitably have an important impact on the arrangement of virtual power plant operation and dispatch plans. The existing methods for dynamic peak regulation of virtual power plants still have the following problems: (1) With the increasing proportion of low-dispatch resources such as wind power and photovoltaic power and nuclear power units connected to the grid, the peak-to-valley difference of load has increased accordingly, and the peak-to-valley situation of the power grid has become increasingly severe. The demand for nuclear power units to participate in the peak-to-peak operation of the system has become more urgent, resulting in a large gap in the peak-to-peak capacity of the power grid. The annual load tracking and weekly regulation operation of nuclear power units alone are not enough to meet the peak-to-peak demand of the system; (2) Due to the randomness of the output of uncertain power sources such as wind power and photovoltaic power, as well as the existence of random variables such as load forecast errors, the scheduling and operation of dynamic peak regulation of virtual power plants is uncontrollable. How to properly handle uncertain variables becomes the key to ensuring the reliability of power grid operation. Summary of the invention
[0003] The purpose of the present invention is to provide a virtual power plant dynamic peak-shaving method and system based on neural network, so as to solve the technical problems that the existing technology has a large gap in the peak-shaving capacity of the power grid in the off-peak period, the annual load tracking and weekly regulation operation of nuclear power units are insufficient to meet the peak-shaving needs of the system, and the scheduling operation of the dynamic peak-shaving of the virtual power plant is uncontrollable due to uncertain variables.
[0004] In order to solve the above technical problems, the present invention specifically provides the following technical solutions: A first aspect of the present invention provides a virtual power plant dynamic peak regulation method based on a neural network, comprising the following steps: Acquire system dispatch instructions on the power supply side and demand side of the virtual power plant in real time, construct controllable power generation resource data at different time scales for the system dispatch instructions, analyze the power characteristics of the controllable power generation resource data, and establish an adjustable power model of power generation resources at different time scales; Analyze the adjustable power characteristics of the power supply and demand side through the adjustable power model of the power generation resources, aggregate the characteristics of different power resources, establish a virtual power plant resource aggregation model, and obtain the multi-level optimization scheduling data of the virtual power plant; A dynamic peak-shaving strategy is adopted to construct an objective function for the multi-level optimization scheduling data, and dynamic constraints are introduced to construct an output scheduling model for the multi-level optimization scheduling data to obtain real-time power dynamic peak-shaving data.
[0005] As a preferred solution of the present invention, the system dispatching instructions on the power supply side and the demand side of the virtual power plant are obtained in real time, and the controllable resource data of power generation at different time scales are constructed for the system dispatching instructions, including: Counting characteristic indicators of power generation resources corresponding to the system scheduling instructions, and dividing the characteristic indicators into power generation resource adjustment rate, adjustable power and adjustable power; According to the characteristic index, an adjustable power diagram of power generation resources is established at different time scales, with the vertical axis being the power axis and the horizontal axis being the time axis; The adjustable power diagram of the power generation resource is based on the initial operation point of the power generation As the starting point, at any time , Calculate the increase in power separately , reduce power consumption , whose expression is: ; in, , Respectively , The power value at the moment, , They represent the maximum and minimum adjustable power values during the regulation process. Indicates the upper limit of power regulation. Indicates the lower limit of power regulation. Indicates the regulation efficiency value of power generation resources; According to the adjustable power diagram of the power generation resources, to Within the time scale, the power is regulated according to the maximum power operation to obtain the controllable power generation resource data at different time scales.
[0006] As a preferred solution of the present invention, the power characteristics of the controllable power generation resource data are analyzed to establish an adjustable power model of power generation resources at different time scales, including: The controllable power generation resource data is divided according to the virtual power plant power generation scenario Type ,Will Type of single generation data as data center ; statistics Type The data center is selected, and the control probability is set for each data center to obtain the control probability distribution corresponding to the power generation controllable resource data, which is expressed as: ; in, It indicates the power fluctuation amplitude caused by power regulation in the power market; According to the control probability distribution, the total power conservation on the power supply side and the demand side is maintained, and the interruptible load on the demand side is dealt with by the load characteristics during the power time shift process. The expression is: ; in, express , The time difference between the moments Indicates the maximum charging power contained in the controllable power generation resource data. Indicates the power generated at the current moment, Current energy storage capacity, express , The maximum adjustable power within the time period; According to the interruptible load, the dispatchable power resources on the demand side are analyzed in real time, an adjustable power model of power generation resources under different time scales is established, and the adjustable energy storage characteristics of the power supply and demand side are obtained.
[0007] As a preferred solution of the present invention, the adjustable power model of power generation resources is used to analyze the adjustable power characteristics of power supply and demand side, and to gather the characteristics of different power resources, including: Analyze the power generation scenario of the power plant on the power supply side and the demand side according to the adjustable energy storage characteristics, and obtain the equivalent load prediction error between different power sources through the regulation probability distribution; Establish that the equivalent load forecast error obeys the Normal distribution, its expression is: ; in, represents the expected value of the equivalent load forecast error in different time periods, represents the variance of the expected value of the equivalent load forecast error in different time periods, Indicates the power difference of equivalent load in different time periods; The equivalent load prediction error replaces other scenarios with similar characteristics by a typical power plant power generation scenario, and solves the power resource feature set under different power plant power generation scenarios.
[0008] As a preferred solution of the present invention, a virtual power plant resource aggregation model is established according to the characteristics of the different power resources, and multi-level optimization scheduling data of the virtual power plant is obtained, including: According to the characteristics of different power resources, the virtual power plant resources are aggregated and dispatched with the goal of maximizing the overall economic benefits of the virtual power plant. The expression is: ; in, Represents the risk preference coefficient of different power resource characteristics in the aggregation process, represents the length of time periods at different time scales, Representation scene The probability of the following scenario, Indicates the energy efficiency value, Indicates that the virtual power plant benefit is less than the energy efficiency value The proportion of Represents the photovoltaic output income, Indicates the wind turbine output benefit, represents the revenue of energy storage equipment, represents the flexible load benefit, represents the transaction revenue between the virtual power plant and the grid; The aggregated and scheduled virtual power plant resource data is hierarchically executed to obtain multi-level optimized scheduling data.
[0009] As a preferred solution of the present invention, a dynamic peak-shaving strategy is adopted to construct an objective function for the multi-level optimization scheduling data, and a dynamic constraint condition is introduced to construct an output scheduling model for the multi-level optimization scheduling data to obtain real-time power dynamic peak-shaving data, including: The multi-level optimization scheduling data is targeted at minimizing the curve variance of the equivalent load forecast error at different time scales, where the objective function is: ; in, Indicates The power load value at the moment, Indicates Energy storage charging and discharging power at all times, Indicates the average load of the day; Using the energy storage charging and discharging power and remaining power of the virtual power plant as constraints, initial forecast data is obtained; Iteratively process the initial prediction data using a neural network, plan the energy storage output dispatch data, and obtain an equivalent load prediction error curve after peak shaving and valley filling and a predicted value of the remaining energy of the energy storage; Compare the relationship between the predicted value of energy storage remaining energy and the actual value, and adjust the peak shaving and valley filling limit value through the limit adjustment method to correct the energy storage error; Compare the actual load with the predicted load, use the load ratio method to correct the load prediction error, complete the entire peak shaving and valley filling control strategy, and obtain real-time dynamic power peak regulation data.
[0010] As a preferred solution of the present invention, the energy storage charging and discharging power constraint condition is: during the dynamic peak regulation process of the virtual power plant, the energy storage charging and discharging power Always less than the rated power of energy storage , whose expression is: ; The remaining power constraint condition is: during the dynamic peak regulation of the virtual power plant, the remaining power of the energy storage device shall not be less than the minimum allowable charge and discharge capacity of the energy storage, and shall not be greater than the maximum allowable charge and discharge capacity of the energy storage.
[0011] A second aspect of the present invention provides a system for a virtual power plant dynamic peak regulation method based on a neural network, comprising: A power data processing module is used to process multi-source power data input by multi-energy generators, and obtain power load values at different time scales after processing the multi-source power data; The load data prediction module predicts the power load status in real time and monitors the load prediction error after dynamic peak regulation in real time through the equivalent load prediction error curve; The optimization scheduling module is used to generate and execute dynamic peak load regulation strategies, perceive the power load operation status in real time, input the short-term prediction points and scenario prediction points of the power plant into the optimization scheduling module, timely correct the power load operation conditions, and complete the real-time power scheduling plan adjustment; The virtual power plant control module controls the charging and discharging power on the power supply side in real time, detects interruptible loads, and interacts with energy storage devices for power.
[0012] Compared with the prior art, the present invention has the following beneficial effects: The present invention adopts a dynamic peak-shaving strategy. On the basis of short-term power load forecasting, the present invention arranges the day-ahead scheduling plan for the energy storage output with the purpose of using energy storage to help load peak shaving and valley filling. The present invention takes the minimum load variance as the objective function, and the energy storage charging and discharging power and the remaining power as constraints. An energy storage output scheduling model is established to correct the error caused by load forecasting and obtain an energy storage output scheduling plan, which can make the load curve smoother and improve the peak-shaving effect.
[0013] Taking into account the randomness of uncertain power output such as wind power and photovoltaic power, as well as the existence of random variables such as load forecast errors, the flexible adjustability and overall randomness of the aggregation elements are analyzed and modeled, and the dynamic characteristics of key elements are analyzed, which provides theoretical support for the aggregator to select the aggregation elements of the virtual power plant. At the level of aggregation regulation characteristics, on the one hand, the external adjustable flexibility of the virtual power plant is obtained by integrating the flexible and adjustable resources. On the other hand, the probability density function of distributed resources is calculated by convolution to obtain the randomness model of the virtual power plant, which ensures the stability of the power grid operation. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.
[0015] Figure 1 A flow chart of a dynamic peak-shaving method for a virtual power plant provided in an embodiment of the present invention; Figure 2 A structural block diagram of a virtual power plant dynamic peak-shaving system provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0016] 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.
[0017] First embodiment: Figure 1 As shown, the present invention provides a virtual power plant dynamic peak regulation method based on neural network, comprising the following steps: Obtain system dispatch instructions on the power supply and demand sides of the virtual power plant in real time, construct controllable power resource data at different time scales for the system dispatch instructions, analyze the power characteristics of the controllable power resource data, and establish an adjustable power model for power resources at different time scales; In this embodiment, by acquiring the operating characteristics and aggregation effects of distributed resources in real time, the needs of various energy sources are comprehensively considered, the optimal configuration of various resources is achieved at different time scales, and the controllable resource data for power generation is obtained.
[0018] Analyze the adjustable power characteristics of power supply and demand side through the adjustable power model of power generation resources, aggregate the characteristics of different power resources, establish a virtual power plant resource aggregation model, and obtain multi-level optimization scheduling data of virtual power plants; In this embodiment, on the power supply side, according to the controllability of the unit, it can be divided into controllable resources and semi-controlled resources; on the load side, based on the two response principles of price type and incentive type, it can be divided into time-shifted demand side resources and interruptible demand side resources. According to the flexibility of the adjustable resources, the dispatching instructions of the controllable power supply response system and the power fluctuations of the tracking system are aggregated to obtain multi-level optimized dispatching data of the virtual power plant.
[0019] A dynamic peak-shaving strategy is used to construct the objective function of the multi-level optimization scheduling data, and dynamic constraints are introduced to build an output scheduling model for the multi-level optimization scheduling data to obtain real-time dynamic power peak-shaving data.
[0020] In this embodiment, firstly, the adjustable flexibility and randomness of different distributed power sources, traditional units, controllable loads and energy storage are analyzed, and a virtual power plant resource aggregation model of each aggregation element is established at the level of flexible adjustable resources; secondly, at the level of aggregation regulation characteristics, on the one hand, the external adjustable flexibility of the virtual power plant is obtained by integrating the flexible adjustable resources, and on the other hand, the probability density function of the distributed resources is calculated by convolution to obtain the randomness model of the virtual power plant. Finally, based on the investment portfolio theory, an optimization configuration model considering the randomness of the virtual power plant is established, and by introducing dynamic constraints to analyze the internal unit capacities that need to be dynamically matched, a configuration solution is provided for the multi-level optimization scheduling of internal resources of the virtual power plant.
[0021] Obtain system dispatch instructions on the power supply side and demand side of the virtual power plant in real time, and construct controllable power generation resource data at different time scales for the system dispatch instructions, including: Statistical system dispatch instructions corresponding to the characteristic indicators of power generation resources, the characteristic indicators are divided into power generation resource adjustment rate, adjustable power and adjustable power; In this embodiment, characteristic indicators of power generation resources are obtained based on the reliability and stability of controllable resources and the dispatching instructions of the response system, and the characteristic indicators are established as an adjustable power diagram of power generation resources with the vertical axis being the power axis and the horizontal axis being the time axis.
[0022] According to the characteristic indicators, an adjustable power diagram of power generation resources is established at different time scales, with the vertical axis being the power axis and the horizontal axis being the time axis; In this embodiment, the area enclosed by the power change curve and the coordinate axis at different times in the adjustable power diagram of power generation resources represents the flexibility range of the adjustable resources, the curve area enclosed by the maximum power value represents the maximum upward flexibility of the adjustable resources, and the curve area enclosed by the minimum power value represents the minimum downward flexibility of the adjustable resources.
[0023] Adjustable power diagram of power generation resources based on initial operation point of power generation As the starting point, at any time , Calculate the increase in power separately , reduce power consumption , whose expression is: ; in, , Respectively , The power value at the moment, , They represent the maximum and minimum adjustable power values during the regulation process. Indicates the upper limit of power regulation. Indicates the lower limit of power regulation. Indicates the regulation efficiency value of power generation resources; In this embodiment, the adjustable power of the power generation resources at any time can be obtained by subtracting the maximum and minimum output powers of the known adjustable resources from the current operating power.
[0024] According to the adjustable power diagram of power generation resources to Within the time scale, the power is regulated according to the maximum power operation to obtain the controllable power generation resource data at different time scales.
[0025] In this embodiment, when the adjustable power of the power generation resource is in the initial stage of regulation, that is, when the output is low, the adjustable power of the unit is a constant value, and its flexibility is limited by the regulation rate of the power generation resource; when the unit output gradually increases, when the upward adjustment space is less than the climbing capacity, its flexibility is limited by its own regulation ability; this flexibility adjustment index reflects the maximum adjustable capacity of the unit upward and downward, and at the same time characterizes the regulation rate of the power generation resource and the rising rate of the power generation resource power.
[0026] Analyze the power characteristics of controllable power generation resource data and establish adjustable power generation resource models at different time scales, including: Divide the controllable power generation resource data according to the virtual power plant power generation scenario Type ,Will Type of single generation data as data center ; statistics Type The data center is selected, and the control probability is set for each data center to obtain the control probability distribution of the corresponding power generation controllable resource data, which is expressed as: ; in, It indicates the power fluctuation amplitude caused by power regulation in the power market; In this embodiment, according to the different power generation scenarios of the virtual power plant, a dynamic peak-shaving scenario data center is established to obtain hidden information in the operation of the virtual power plant, and a probability distribution area belonging to the virtual power plant management and control is established to obtain higher peak-shaving profits.
[0027] According to the control probability distribution, the total power conservation on the power supply side and the demand side is maintained, and the interruptible load on the demand side is dealt with by the load characteristics during the power time shift process. The expression is: ; in, express , The time difference between the moments Indicates the maximum charging power contained in the controllable power generation resource data. Indicates the power generated at the current moment, Current energy storage capacity, express , The maximum adjustable power within the time period; According to the interruptible load, the dispatchable power resources on the demand side are analyzed in real time, and the adjustable power model of power generation resources under different time scales is established to obtain the adjustable energy storage characteristics of the power supply and demand side.
[0028] In this embodiment, the virtual power plant selects the optimal value of the control probability distribution corresponding to its hidden information to maximize its own utility, thereby exposing the real hidden information and solving the problem of information asymmetry between the power supply side and the demand side during the peak-shaving process. In the complete information scenario where the real information of the virtual power plant can be obtained by the data center, the data center knows the type of each virtual power plant in advance, including the real peak-shaving revenue of the virtual power plant. In this scenario, the data center participates in the peak-shaving transaction process of the virtual power plant and can retain the utility of the virtual power plant by minimizing its utility, thereby obtaining higher utility than in the incomplete information scenario.
[0029] The adjustable power model of power generation resources is used to analyze the adjustable power characteristics of power supply and demand side, and to gather the characteristics of different power resources, including: Analyze the power generation scenarios of power plants on the power supply and demand side based on the adjustable energy storage characteristics, and obtain the equivalent load prediction error between different power sources by adjusting the probability distribution; Establish the equivalent load forecast error under different time scales Normal distribution, its expression is: ; in, represents the expected value of the equivalent load forecast error in different time periods, represents the variance of the expected value of the equivalent load forecast error in different time periods, Indicates the power difference of equivalent load in different time periods; The equivalent load forecast error replaces other scenarios with similar characteristics with typical power plant power generation scenarios to solve the set of power resource characteristics under different power plant power generation scenarios.
[0030] In this embodiment, in this embodiment, the equivalent load prediction error is calculated based on the corresponding control probability distribution area.
[0031] According to the characteristics of different power resources, a virtual power plant resource aggregation model is established to obtain multi-level optimization scheduling data of the virtual power plant, including: According to the characteristics of different power resources, the virtual power plant resources are aggregated and dispatched with the goal of maximizing the overall economic benefits of the virtual power plant. The expression is: ; in, Represents the risk preference coefficient of different power resource characteristics in the aggregation process, represents the length of time periods at different time scales, Representation scene The probability of the following scenario, Indicates the energy efficiency value, Indicates that the virtual power plant benefit is less than the energy efficiency value The proportion of Represents the photovoltaic output income, Indicates the wind turbine output benefit, represents the revenue of energy storage equipment, represents the flexible load benefit, represents the transaction revenue between the virtual power plant and the grid; The aggregated and scheduled virtual power plant resource data is hierarchically executed to obtain multi-level optimized scheduling data.
[0032] The dynamic peak-shaving strategy is used to construct the objective function for the multi-level optimization dispatching data, and the dynamic constraint conditions are introduced to construct the output dispatching model for the multi-level optimization dispatching data to obtain the real-time power dynamic peak-shaving data, including: The objective of multi-level optimization dispatching data is to minimize the curve variance of equivalent load forecast error at different time scales, where the objective function is: ; in, Indicates The power load value at the moment, Indicates Energy storage charging and discharging power at all times, Indicates the average load of the day; In this embodiment, the objective function is used to ensure that the remaining power of the energy storage is within a reasonable range, which is beneficial to battery maintenance and convenient for scheduling under abnormal conditions. In addition, the charging and discharging power of the energy storage should also be limited to its rated power range, otherwise the output scheduling will not be fully executed.
[0033] Using the energy storage charging and discharging power and remaining power of the virtual power plant as constraints, initial forecast data is obtained; The neural network is used to iteratively process the initial prediction data, plan the energy storage output dispatch data, and obtain the equivalent load prediction error curve after peak shaving and valley filling and the energy storage remaining energy prediction value; In this embodiment, based on the initial forecast data, the output of the energy storage power station is dispatched taking into account the maximum output limit of the wind and photovoltaic power plant, the remaining power of the energy storage power station, and the output constraints, so as to achieve the best overall economic benefits, maximize the absorption of wind power and photovoltaic power, and achieve the purpose of smoothing the equivalent load forecast error curve.
[0034] Compare the relationship between the predicted value of energy storage remaining energy and the actual value, and adjust the peak shaving and valley filling limit value through the limit adjustment method to correct the energy storage error; Compare the actual load with the predicted load, use the load ratio method to correct the load prediction error, complete the entire peak shaving and valley filling control strategy, and obtain real-time dynamic power peak regulation data.
[0035] In this embodiment, the energy storage output is dispatched in real time according to the load forecast data, and the equivalent load forecast error curve after peak shaving and valley filling and the energy storage remaining energy forecast value are obtained. Then, the actual load is compared with the predicted load. In the case of excessive deviation, the energy storage output is re-dispatched. Then, the actual remaining energy situation is compared with the predicted remaining energy situation. In the case of excessive deviation, the energy storage output is re-dispatched by adjusting the limit value of peak shaving and valley filling, so as to realize precise peak shaving from two aspects: the energy storage output and the remaining energy of the energy storage.
[0036] The energy storage charging and discharging power constraint condition is: during the dynamic peak load regulation of the virtual power plant, the energy storage charging and discharging power Always less than the rated power of energy storage , whose expression is: ; The remaining power constraint condition is: during the dynamic peak regulation of the virtual power plant, the remaining power of the energy storage device shall not be less than the minimum allowable charge and discharge capacity of the energy storage, and shall not be greater than the maximum allowable charge and discharge capacity of the energy storage.
[0037] In this embodiment, the energy storage output scheduling plan needs to take into account the peak-shaving effect of the net load after the addition of the wind, solar and storage systems and the economic benefits of users and power plants in order to meet the basic expectations of a peak-shaving strategy. The three objective functions of minimizing the net load variance, minimizing the electricity cost, and minimizing the combined output cost of wind, solar and storage are selected for multi-objective optimization. The smaller the net load variance, the flatter the net load curve, and the better the peak-shaving effect of the wind, solar and storage combined system. Therefore, the minimum variance of the net load after the combined peak shaving and valley filling is taken as the objective function.
[0038] Second embodiment: Figure 2 As shown, a system of a virtual power plant dynamic peak regulation method based on a neural network includes: The power data processing module is used to process the multi-source power data input by the multi-energy generator sets, and obtain the power load values at different time scales after processing the multi-source power data; The load data prediction module predicts the power load status in real time and monitors the load prediction error after dynamic peak regulation in real time through the equivalent load prediction error curve; The optimization scheduling module is used to generate and execute dynamic peak load regulation strategies, perceive the power load operation status in real time, input the short-term prediction points and scenario prediction points of the power plant into the optimization scheduling module, timely correct the power load operation conditions, and complete the real-time power scheduling plan adjustment; The virtual power plant control module controls the charging and discharging power on the power supply side in real time, detects interruptible loads, and interacts with energy storage devices for power.
[0039] In this embodiment, the virtual power plant control module accepts the dispatching instructions from the optimization dispatching module, and then the load data prediction module combines the information of the load prediction system and the requirements of the wind, solar and storage joint dispatching, and issues charging and discharging instructions after calculation. This can not only increase the wind and solar power consumption ratio, but also reduce costs, with good economic benefits.
[0040] The present invention adopts a dynamic peak-shaving strategy. On the basis of short-term power load forecasting, the present invention arranges the day-ahead scheduling plan for the energy storage output with the purpose of using energy storage to help load peak shaving and valley filling. The present invention takes the minimum load variance as the objective function, and the energy storage charging and discharging power and the remaining power as constraints. An energy storage output scheduling model is established to correct the error caused by load forecasting and obtain an energy storage output scheduling plan, which can make the load curve smoother and improve the peak-shaving effect.
[0041] Taking into account the randomness of uncertain power output such as wind power and photovoltaic power, as well as the existence of random variables such as load forecast errors, the flexible adjustability and overall randomness of the aggregation elements are analyzed and modeled, and the dynamic characteristics of key elements are analyzed, which provides theoretical support for the aggregator to select the aggregation elements of the virtual power plant. At the level of aggregation regulation characteristics, on the one hand, the external adjustable flexibility of the virtual power plant is obtained by integrating the flexible and adjustable resources. On the other hand, the probability density function of distributed resources is calculated by convolution to obtain the randomness model of the virtual power plant, which ensures the stability of the power grid operation.
[0042] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.
Claims
1. A virtual power plant dynamic peak regulation method based on neural network, characterized in that: The following steps are involved: Acquire system dispatch instructions on the power supply side and demand side of the virtual power plant in real time, construct controllable power generation resource data at different time scales for the system dispatch instructions, analyze the power characteristics of the controllable power generation resource data, and establish an adjustable power model of power generation resources at different time scales; Analyze the adjustable power characteristics of the power supply and demand side through the adjustable power model of the power generation resources, aggregate the characteristics of different power resources, establish a virtual power plant resource aggregation model, and obtain the multi-level optimization scheduling data of the virtual power plant; A dynamic peak-shaving strategy is adopted to construct an objective function for the multi-level optimization scheduling data, and dynamic constraints are introduced to construct an output scheduling model for the multi-level optimization scheduling data to obtain real-time power dynamic peak-shaving data.
2. According to the neural network-based virtual power plant dynamic peak-shaving method of claim 1, it is characterized in that: Obtain system dispatch instructions on the power supply side and demand side of the virtual power plant in real time, and construct controllable power generation resource data at different time scales for the system dispatch instructions, including: Counting characteristic indicators of power generation resources corresponding to the system scheduling instructions, and dividing the characteristic indicators into power generation resource adjustment rate, adjustable power and adjustable power; According to the characteristic index, an adjustable power diagram of power generation resources is established at different time scales, with the vertical axis being the power axis and the horizontal axis being the time axis; The adjustable power diagram of the power generation resource is based on the initial operation point of the power generation As the starting point, at any time , Calculate the increase in power separately , reduce power consumption , whose expression is: ; in, , Respectively , The power value at the moment, , They represent the maximum and minimum adjustable power values during the regulation process. Indicates the upper limit of power regulation. Indicates the lower limit of power regulation. Indicates the regulation efficiency value of power generation resources; According to the adjustable power diagram of the power generation resources, to Within the time scale, the power is regulated according to the maximum power operation to obtain the controllable power generation resource data at different time scales.
3. The method for dynamic peak load regulation of a virtual power plant based on a neural network according to claim 2, characterized in that: Analyze the power characteristics of the controllable power resource data and establish an adjustable power model of power resources at different time scales, including: The controllable power generation resource data is divided according to the virtual power plant power generation scenario Type ,Will Type of single generation data as data center ; statistics Type The data center is selected, and the control probability is set for each data center to obtain the control probability distribution corresponding to the power generation controllable resource data, which is expressed as: ; in, It indicates the power fluctuation amplitude caused by power regulation in the power market; According to the control probability distribution, the total power conservation on the power supply side and the demand side is maintained, and the interruptible load on the demand side is dealt with by the load characteristics during the power time shift process. The expression is: ; in, express , The time difference between the moments Indicates the maximum charging power contained in the controllable power generation resource data. Indicates the power generated at the current moment, Current energy storage capacity, express , The maximum adjustable power within the time period; According to the interruptible load, the dispatchable power resources on the demand side are analyzed in real time, an adjustable power model of power generation resources under different time scales is established, and the adjustable energy storage characteristics of the power supply and demand side are obtained.
4. The method for dynamic peak load regulation of a virtual power plant based on a neural network according to claim 3 is characterized in that: The adjustable power model of power generation resources is used to analyze the adjustable power characteristics of power supply and demand side, and to gather the characteristics of different power resources, including: Analyze the power generation scenario of the power plant on the power supply side and the demand side according to the adjustable energy storage characteristics, and obtain the equivalent load prediction error between different power sources through the regulation probability distribution; Establish that the equivalent load forecast error obeys the Normal distribution, its expression is: ; in, represents the expected value of the equivalent load forecast error in different time periods, represents the variance of the expected value of the equivalent load forecast error in different time periods, Indicates the power difference of equivalent load in different time periods; The equivalent load prediction error replaces other scenarios with similar characteristics by a typical power plant power generation scenario, and solves the power resource feature set under different power plant power generation scenarios.
5. The method for dynamic peak load regulation of a virtual power plant based on a neural network according to claim 3, characterized in that: According to the characteristics of the different power resources, a virtual power plant resource aggregation model is established to obtain multi-level optimization scheduling data of the virtual power plant, including: According to the characteristics of different power resources, the virtual power plant resources are aggregated and dispatched with the goal of maximizing the overall economic benefits of the virtual power plant. The expression is: ; in, Represents the risk preference coefficient of different power resource characteristics in the aggregation process, represents the length of time periods at different time scales, Representation scene The probability of the following scenario, Indicates the energy efficiency value, Indicates that the profit of virtual power plant is less than the energy efficiency value The proportion of Represents the photovoltaic output income, Indicates the wind turbine output benefit, represents the revenue of energy storage equipment, represents the flexible load benefit, represents the transaction revenue between the virtual power plant and the grid; The aggregated and scheduled virtual power plant resource data is hierarchically executed to obtain multi-level optimized scheduling data.
6. A method for dynamic peak load regulation of virtual power plant based on neural network according to claim 5, characterized in that: A dynamic peak-shaving strategy is adopted to construct an objective function for the multi-level optimization dispatching data, and a dynamic constraint condition is introduced to construct an output dispatching model for the multi-level optimization dispatching data to obtain real-time power dynamic peak-shaving data, including: The multi-level optimization scheduling data is targeted at minimizing the curve variance of the equivalent load forecast error at different time scales, where the objective function is: ; in, Indicates The power load value at the moment, Indicates Energy storage charging and discharging power at all times, Indicates the average load of the day; Using the energy storage charging and discharging power and remaining power of the virtual power plant as constraints, initial forecast data is obtained; Iteratively process the initial prediction data using a neural network, plan the energy storage output dispatch data, and obtain an equivalent load prediction error curve after peak shaving and valley filling and a predicted value of the remaining energy of the energy storage; Compare the relationship between the predicted value of energy storage remaining energy and the actual value, and adjust the peak shaving and valley filling limit value through the limit adjustment method to correct the energy storage error; Compare the actual load with the predicted load, use the load ratio method to correct the load prediction error, complete the entire peak shaving and valley filling control strategy, and obtain real-time dynamic power peak regulation data.
7. A method for dynamic peak load regulation of virtual power plant based on neural network according to claim 6, characterized in that: The energy storage charging and discharging power constraint condition is: during the dynamic peak regulation process of the virtual power plant, the energy storage charging and discharging power Always less than the rated power of energy storage , whose expression is: ; The remaining power constraint condition is: during the dynamic peak regulation of the virtual power plant, the remaining power of the energy storage device shall not be less than the minimum allowable charge and discharge capacity of the energy storage, and shall not be greater than the maximum allowable charge and discharge capacity of the energy storage.
8. A system for dynamic peak-shaving method of virtual power plant based on neural network, characterized in that: include: A power data processing module is used to process multi-source power data input by multi-energy generators, and obtain power load values at different time scales after processing the multi-source power data; The load data prediction module predicts the power load status in real time and monitors the load prediction error after dynamic peak regulation in real time through the equivalent load prediction error curve; The optimization scheduling module is used to generate and execute dynamic peak load regulation strategies, perceive the power load operation status in real time, input the short-term prediction points and scenario prediction points of the power plant into the optimization scheduling module, timely correct the power load operation conditions, and complete the real-time power scheduling plan adjustment; The virtual power plant control module controls the charging and discharging power on the power supply side in real time, detects interruptible loads, and interacts with energy storage devices for power.
Citation Information
Patent Citations
Virtual power plant optimization trading strategy model based on two-stage stochastic programming
CN108960510A
Source-load bilateral thermoelectric combined random optimization scheduling method based on virtual power plant
CN110516851A
Scene generation method considering wind and light prediction error
CN113191063A
Multi-subject virtual power plant optimization scheduling method based on risk preference
CN113469402A
Virtual power plant capacity optimization configuration method of multi-agent game
CN113888204A
Cited By
Multi-time-scale virtual flexible load scheduling method and system for coping with power fluctuation
CN119093383A
Virtual power plant scheduling method and system based on layer area dynamic construction, medium and equipment
CN119918904A
Virtual power plant scheduling method, system, medium and device based on dynamic construction of layer areas
CN119918904B
Intelligent scheduling method for comprehensive virtual power plant
CN120046958A
Virtual power plant regulation and control method and system based on dynamic parameter identification and hierarchical collaboration
CN120127652A
