Resource configuration method and system for new energy sending type hybrid distribution network

Abstract

The present application relates to the technical field of power system resource configuration, and discloses a resource configuration method and system for a new energy sending type hybrid distribution network.The resource configuration method for the new energy sending type hybrid distribution network comprises the following steps: a virtual power plant is established based on a plurality of intelligent controllable terminals; historical power consumption data, real-time running state information data, environmental data and user behavior data of the intelligent controllable terminals are collected; the data is preprocessed and feature extraction is performed; based on the extracted features, a trained prediction model is used to perform phased prediction of the total power consumption of the intelligent controllable terminals; and according to the power consumption prediction result and the power grid load condition, the distribution network is optimally configured with the minimum sum of total configuration and running simulation cost within the planning period as the target.The present application takes the intelligent controllable terminal as the demand end, constructs a virtual power plant, and reduces the power consumption prediction deviation caused by data collection and transmission abnormalities.

Description

Technical Field

[0001] This invention relates to the field of power system resource allocation technology, and specifically to a resource allocation method and system for a hybrid distribution network for transmitting new energy to other regions. Background Technology

[0002] With the rapid development of distributed energy resources and smart grids, traditional centralized power systems face numerous challenges in managing distributed energy resources and smart controllable terminals. The widespread integration of distributed energy resources (such as photovoltaics and wind power) and smart controllable terminals (such as electric vehicles and smart home devices) has fundamentally changed the structure and operation of power systems. Traditional centralized power systems struggle to efficiently coordinate these dispersed resources, leading to problems such as uneven resource allocation, mismatch between power supply and demand, and decreased grid stability. Furthermore, the intermittency of distributed energy resources and the dynamic nature of smart controllable terminals further increase the complexity of power system operation, necessitating an innovative management model to achieve efficient integration and optimized allocation of resources.

[0003] Virtual power plants (VPPs), as an emerging energy management model, require dynamic adjustments to power supply strategies based on electricity consumption forecasts for the controlled area to achieve power supply-demand balance and optimal resource allocation. However, operational failures of electricity-side equipment (such as overload, short circuits, or shutdowns) and anomalies during data acquisition and transmission (such as sensor failures, communication delays, or data loss) can lead to significant deviations between predicted and actual electricity consumption. These deviations not only affect the accuracy of power supply strategies but may also cause grid supply-demand imbalances, resource waste, or decreased system stability, thereby reducing the operational efficiency and reliability of the virtual power plant. Therefore, a comprehensive solution capable of real-time monitoring, fault diagnosis, and predictive optimization is needed to improve the accuracy of electricity consumption forecasts and the adaptability of power supply strategies. Summary of the Invention

[0004] The purpose of this invention is to provide a resource allocation method and system for hybrid distribution networks that transmit new energy to other regions, thereby solving the above-mentioned technical problems:

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] Resource allocation methods for hybrid distribution networks that transmit new energy sources to other regions include:

[0007] A virtual power plant is established based on multiple intelligent and controllable terminals; historical electricity consumption data, real-time operating status information data, environmental data, and user behavior data of the intelligent and controllable terminals are collected; and the data is preprocessed and features are extracted.

[0008] Based on the extracted features, the trained prediction model is used to make phased predictions of the total power consumption of the intelligent controllable terminal; according to the power consumption prediction results and the grid load, the power supply is dynamically adjusted, including the output power of distributed energy, the charging and discharging strategy of the energy storage system, and the power dispatch of the external grid.

[0009] At preset time points, the actual electricity consumption is compared and analyzed with the predicted electricity consumption to evaluate the accuracy of the electricity consumption prediction. Based on the evaluation results, the fault analysis strategy of the intelligent controllable terminal and the optimization strategy of the prediction model are established.

[0010] As a further technical solution, historical electricity consumption data includes: historical electricity consumption records of the intelligent controllable terminal; real-time operating status information includes: current, voltage, power, and temperature during device operation; environmental data includes: temperature, humidity, and weather conditions of the operating environment of the intelligent controllable terminal; and user behavior data includes: frequency, time, and preferences of user device use.

[0011] As a further technical solution, the process of data preprocessing and feature extraction includes:

[0012] The historical electricity consumption data, real-time operating status information data, environmental data, and user behavior data of the intelligent controllable terminal are cleaned and normalized to extract features, including: time features, statistical features, environmental features, and user behavior features.

[0013] As a further technical solution, the process of comparing and analyzing actual electricity consumption with predicted electricity consumption includes:

[0014] statistics Total power consumption forecast of each intelligent controllable terminal in each historical forecasting period This corresponds to the actual total electricity consumption of each smart controllable terminal during the historical prediction period, collected by smart meters. ;

[0015] Through formula Calculate the systematic deviation value of electricity consumption forecast ;

[0016] Through formula Calculate the randomness bias of electricity consumption forecast ,in ;

[0017] Through formula Calculate the mean square error of electricity consumption forecast ,in The error is unavoidable.

[0018] As a further technical solution, the mean square error of electricity consumption prediction will be used. With preset threshold Compare:

[0019] like Then retrieve the corresponding The system collects operational status information data of each intelligent and controllable terminal in each historical prediction stage, and performs fault analysis on the intelligent and controllable terminals based on the operational status information data.

[0020] like If not, fault analysis will not be performed on the intelligent controllable terminal.

[0021] As a further technical solution, the process of fault analysis for intelligent controllable terminals includes:

[0022]

[0023] The failure risk coefficient of the intelligent controllable terminal is calculated using the above formula. ;in, This represents the total number of historical prediction periods. These are the weighting coefficients corresponding to each prediction stage; The curves showing the changes in operating parameters of the intelligent controllable terminal include the current, voltage, output power, and temperature of key components of the intelligent controllable terminal. The standard variation curve of the operating parameters of the intelligent and controllable terminal; The weights corresponding to the running parameters; , These represent the start and end times of the i-th prediction stage;

[0024] The failure risk coefficient of intelligent and controllable terminals With preset risk threshold Compare:

[0025] like If so, it is determined that the corresponding intelligent controllable terminal is at risk of failure;

[0026] like If the corresponding intelligent controllable terminal is deemed to have no fault risk, the prediction model optimization strategy will be executed based on the comparison analysis results of the actual power consumption and the predicted power consumption.

[0027] As a further technical solution, the prediction model optimization strategy includes:

[0028] pass Obtain the ratio of systematic bias to random bias. ;

[0029] ratio With preset coefficient Compare:

[0030] like If the number of features is increased, the regularization coefficient will be decreased.

[0031] like This reduces the number of features and increases the regularization coefficient.

[0032] Finally, based on the obtained predicted power consumption Considering the constraints of the power distribution network and taking power supply as the basic requirement, an objective function is constructed that minimizes the sum of the total construction cost and the system simulation operation cost. By rationally configuring the virtual power plant, the failure risk is reduced, and the reliability of the planning results is effectively improved.

[0033] A resource allocation system for hybrid distribution networks that transmit new energy sources to other regions includes:

[0034] The data acquisition module is used to collect historical power consumption data, real-time operating status information data, environmental data, and user behavior data from the intelligent controllable terminal, and to preprocess the data.

[0035] The power consumption prediction module is used to extract features from the data and, based on these features, use a trained prediction model to make phased predictions of the total power consumption of the smart controllable terminal.

[0036] The power supply control module is used to dynamically adjust the power supply based on the forecast results and the grid load;

[0037] The fault analysis module is used to evaluate the accuracy of electricity consumption forecasts and establish fault analysis strategies and prediction model optimization strategies for intelligent controllable terminals based on the evaluation results.

[0038] The optimization module is used to execute the optimization strategy of the prediction model and obtain relevant information such as the optimal equipment construction location and capacity.

[0039] The beneficial effects of this invention are:

[0040] This invention constructs a virtual power plant using intelligent controllable terminals as the demand side, reducing electricity consumption forecasting errors caused by data acquisition and transmission anomalies. Furthermore, by evaluating and analyzing electricity consumption forecasts, it establishes fault diagnosis strategies and optimization strategies for prediction models, thereby improving system reliability and stability. This invention is applicable to hybrid distribution networks, efficiently integrating distributed energy resources and intelligent controllable terminals to achieve power supply and demand balance and optimal resource allocation. Attached Figure Description

[0041] The invention will now be further described with reference to the accompanying drawings.

[0042] Figure 1 This is a flowchart of the resource allocation method for a hybrid distribution network for transmitting new energy sources in this invention;

[0043] Figure 2 This is a summary block diagram of the resource allocation system for a hybrid distribution network for transmitting new energy to other regions in this invention; Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] Please see Figure 1 As shown, the resource allocation method for a hybrid distribution network for transmitting new energy sources includes:

[0046] A virtual power plant is established based on multiple intelligent controllable terminals. Historical electricity consumption data, real-time operating status information, environmental data (such as temperature and humidity), and user behavior data are collected from these terminals. Preprocessing operations such as cleaning and normalization are performed on this data to facilitate the extraction of key features.

[0047] Based on the extracted features, a trained prediction model, such as a machine learning model or a deep learning model, is used to make phased predictions of the total electricity consumption of smart and controllable terminals, such as hourly, daily, or weekly predictions. The power supply strategy is then dynamically adjusted based on the electricity consumption prediction results and the grid load. Specifically, this includes adjusting the output power of distributed energy sources (such as photovoltaic and wind power), optimizing the charging and discharging strategies of energy storage systems to balance supply and demand, and coordinating power dispatch from the external grid to ensure stable grid operation.

[0048] This method compares and analyzes actual electricity consumption with predicted electricity consumption at preset time points to evaluate the accuracy of electricity consumption prediction. Based on the evaluation results, it establishes fault analysis strategies for intelligent controllable terminals and optimization strategies for prediction models. Specifically, it compares and analyzes actual electricity consumption with predicted values ​​at preset time points to evaluate the accuracy of prediction models. Based on the evaluation results, it formulates fault analysis strategies for intelligent controllable terminals and optimization strategies for prediction models, such as adjusting model parameters or retraining. This method can also be combined with demand response mechanisms to guide users to adjust their electricity consumption behavior through price signals or incentives, further optimizing resource allocation.

[0049] Through the above technical solution, this embodiment constructs a virtual power plant using intelligent controllable terminals as the demand side, reducing the deviation in electricity consumption forecasting caused by data acquisition and transmission anomalies. Furthermore, by evaluating and analyzing electricity consumption forecasts, establishing fault diagnosis strategies and optimizing prediction models, the reliability and stability of the system can be improved. This invention is applicable to hybrid distribution networks, efficiently integrating distributed energy resources and intelligent controllable terminals to achieve power supply and demand balance and optimal resource allocation.

[0050] Historical electricity consumption data includes: historical electricity consumption records of the intelligent controllable terminal; real-time operating status information includes: current, voltage, power, and temperature during device operation; environmental data includes: temperature, humidity, and weather conditions of the operating environment of the intelligent controllable terminal; user behavior data includes: frequency, time, and preferences of user device use.

[0051] Historical electricity consumption data, real-time operational status information, environmental data, and user behavior data from intelligent controllable terminals are cleaned and normalized to extract features, including time features, statistical features, environmental features, and user behavior features. Specifically, historical electricity consumption data, real-time operational status information, environmental data, and user behavior data are fused to construct a multi-dimensional feature matrix, improving the accuracy of the prediction model. Convolutional neural networks are used to extract local features from time-series data, long short-term memory networks are used to extract long-term dependency features of electricity consumption, and clustering algorithms (such as K-means) are used to classify user behavior, such as high-energy-consuming users and low-energy-consuming users.

[0052] Through the above technical solutions, this embodiment cleanses, normalizes, and extracts features from historical electricity consumption data, real-time operating status information, environmental data, and user behavior data to construct a high-quality feature set, providing a reliable data foundation for electricity consumption prediction, equipment fault analysis, and user behavior optimization. These features not only improve the accuracy of the prediction model but also provide strong support for the refined management and optimal allocation of resources in the power grid.

[0053] The process of comparing and analyzing actual electricity consumption with predicted electricity consumption includes:

[0054] statistics Total power consumption forecast of each intelligent controllable terminal in each historical forecasting period This corresponds to the actual total electricity consumption of each smart controllable terminal during the historical prediction period, collected by smart meters. ;

[0055] Through formula Calculate the systematic deviation value of electricity consumption forecast ;

[0056] Through formula Calculate the randomness bias of electricity consumption forecast ,in ;

[0057] Through formula Calculate the mean square error of electricity consumption forecast ,in The error is unavoidable.

[0058] Through the above technical solution, this embodiment provides a comparative analysis process, specifically, through... , , Calculate and obtain the systematic deviation value of electricity consumption forecast separately. random bias and mean square error .

[0059] Mean square error of electricity consumption prediction With preset threshold Compare:

[0060] like Then retrieve the corresponding The system collects operational status information data of each intelligent and controllable terminal in each historical prediction stage, and performs fault analysis on the intelligent and controllable terminals based on the operational status information data.

[0061] like If not, fault analysis will not be performed on the intelligent controllable terminal.

[0062] Through the above technical solution, this embodiment provides the triggering condition for fault analysis, which is only triggered when the mean square error of the electricity consumption prediction is... Exceeding the preset threshold At the same time, fault analysis of intelligent and controllable terminals can be used to eliminate large deviations in power consumption prediction caused by equipment failure.

[0063] The process of fault analysis for intelligent and controllable terminals includes:

[0064]

[0065] The failure risk coefficient of the intelligent controllable terminal is calculated using the above formula. ;in, This represents the total number of historical prediction periods. These are the weighting coefficients for each prediction stage, with the weighting coefficients being larger for prediction stages closer to the current time point. The curves showing the changes in operating parameters of the intelligent controllable terminal include the current, voltage, output power, and temperature of key components of the intelligent controllable terminal. The standard variation curve of the operating parameters of the intelligent and controllable terminal; The weights corresponding to the running parameters; , These represent the start and end times of the i-th prediction stage;

[0066] The failure risk coefficient of intelligent and controllable terminals With preset risk threshold Compare:

[0067] like If so, it is determined that the corresponding intelligent controllable terminal is at risk of failure;

[0068] like If the corresponding intelligent controllable terminal is deemed to have no fault risk, the prediction model optimization strategy will be executed based on the comparison analysis results of the actual power consumption and the predicted power consumption.

[0069] Through the above technical solution, this embodiment provides a process for fault analysis of intelligent controllable terminals. Specifically, it first calculates the fault risk coefficient of the intelligent controllable terminal using a formula. It should be noted that in the formula The weighting coefficients for each prediction stage are defined, with larger coefficients for prediction stages closer to the current time point. Through this process, when the predicted and actual electricity consumption values ​​deviate significantly, historical data from the intelligent controllable terminal can be analyzed offline, reducing the burden on the system caused by the high computational load during real-time monitoring and analysis. Fault analysis can eliminate the impact of equipment malfunctions on electricity consumption prediction, allowing for the selection of appropriate prediction algorithm optimization strategies.

[0070] Predictive model optimization strategies include:

[0071] pass Obtain the ratio of systematic bias to random bias. ;

[0072] ratio With preset coefficient Compare:

[0073] like Increasing the number of features and decreasing the regularization coefficient are both beneficial when a model is underfitting. Increasing the number of features introduces more information, helping the model capture complex patterns in the data; decreasing the regularization coefficient reduces the limitations on model complexity and improves fitting ability.

[0074] like To reduce model overfitting, reduce the number of features and increase the regularization coefficient. Reducing the number of features can decrease model complexity when the model is overfitting; increasing the regularization coefficient can increase the penalty on the model weights and prevent the model from overfitting the training data.

[0075] After fitting, based on the obtained predicted power... Considering the constraints of the power distribution network and taking power supply as the basic requirement, an objective function is constructed that minimizes the sum of the total construction cost and the system simulation operation cost. By rationally configuring the virtual power plant, the failure risk is reduced, and the reliability of the planning results is effectively improved.

[0076] Through the above technical solution, this embodiment provides a prediction model optimization strategy. Systematic bias refers to the average difference between the model's predicted values ​​and the actual values, reflecting the model's fitting ability. High systematic bias usually means the model is too simple and cannot capture the complex patterns in the data, i.e., underfitting. Random bias refers to the degree of fluctuation in the model's predicted values, reflecting the model's sensitivity to the training data. High random bias usually means the model is too complex and overfits the training data. Finally, based on the predicted data obtained from training, the distribution network is optimized with the goal of minimizing the total economic cost.

[0077] Please see Figure 2 As shown, a resource allocation system for a hybrid distribution network for transmitting new energy sources includes:

[0078] The data acquisition module is used to collect historical power consumption data, real-time operating status information data, environmental data, and user behavior data from the intelligent controllable terminal, and to preprocess the data.

[0079] The electricity consumption prediction module is used to extract features from the data and, based on the features, use a trained prediction model to make a phased prediction of the total electricity consumption of the smart controllable terminal. Specifically, it uses regression models (such as linear regression and support vector regression) or time series models (such as LSTM and ARIMA) to make predictions and output the phased electricity consumption prediction results.

[0080] The power supply control module dynamically adjusts the power supply based on forecast results and grid load. Specifically, it adjusts the output power of distributed energy sources such as photovoltaics and wind power to match electricity demand. If the forecasted electricity demand is lower than the available output power of distributed energy sources, priority is given to using distributed energy sources for power supply, and excess electricity is stored or sold to the external grid. If the forecasted electricity demand is higher than the available output power of distributed energy sources, power is supplemented from the external grid or the energy storage system is called upon.

[0081] The fault analysis module is used to evaluate the accuracy of electricity consumption forecasts and establish fault analysis strategies and prediction model optimization strategies for intelligent controllable terminals based on the evaluation results.

[0082] The optimization module executes optimization strategies for the prediction model. Specifically, it increases the number of features and decreases the regularization coefficient, suitable for cases of model underfitting. It decreases the number of features and increases the regularization coefficient, suitable for cases of model overfitting. Based on the prediction data obtained from training, it optimizes the distribution network configuration with the goal of minimizing the total economic cost.

[0083] Through the above technical solution, this embodiment provides a resource allocation system for a hybrid distribution network for renewable energy transmission. The resource allocation system, through the coordinated operation of modules such as data acquisition, electricity consumption forecasting, power supply control, fault analysis, and model optimization, achieves refined management and dynamic optimization of hybrid distribution network resources.

[0084] This invention can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the invention.

[0085] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0086] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0087] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A resource allocation method for a hybrid distribution network for transmitting new energy to other regions, characterized in that, The method includes: A virtual power plant is established based on multiple intelligent and controllable terminals; historical electricity consumption data, real-time operating status information data, environmental data, and user behavior data of the intelligent and controllable terminals are collected; and the data is preprocessed and features are extracted. Based on the extracted features, the trained prediction model is used to make phased predictions of the total power consumption of the intelligent controllable terminal; according to the power consumption prediction results and the grid load, the power supply is dynamically adjusted, including the output power of distributed energy, the charging and discharging strategy of the energy storage system, and the power dispatch of the external grid. At preset time points, the actual electricity consumption is compared and analyzed with the predicted electricity consumption value to evaluate the accuracy of the electricity consumption prediction. Based on the evaluation results, the fault analysis strategy of the intelligent controllable terminal and the optimization strategy of the prediction model are established. The process of comparing and analyzing actual electricity consumption with predicted electricity consumption includes: statistics Total power consumption forecast of each intelligent controllable terminal in each historical forecasting period This corresponds to the actual total electricity consumption of each smart controllable terminal during the historical prediction period, collected by smart meters. ; Through formula Calculate the systematic deviation value of electricity consumption forecast ; Through formula Calculate the randomness bias of electricity consumption forecast ,in ; Through formula Calculate the mean square error of electricity consumption forecast ,in To minimize errors; Mean square error of electricity consumption prediction With preset threshold Compare: like Then retrieve the corresponding The system collects operational status information data of each intelligent and controllable terminal in each historical prediction stage, and performs fault analysis on the intelligent and controllable terminals based on the operational status information data. like In this case, fault analysis will not be performed on the intelligent controllable terminal. The process of fault analysis for intelligent and controllable terminals includes: The failure risk coefficient of the intelligent controllable terminal is calculated using the above formula. ;in, This represents the total number of historical prediction periods. These are the weighting coefficients corresponding to each prediction stage; The curves showing the changes in operating parameters of the intelligent controllable terminal include the current, voltage, output power, and temperature of key components of the intelligent controllable terminal. The standard variation curve of the operating parameters of the intelligent and controllable terminal; The weights corresponding to the running parameters; , These represent the start and end times of the i-th prediction stage; The failure risk coefficient of intelligent and controllable terminals With preset risk threshold Compare: like If so, it is determined that the corresponding intelligent controllable terminal is at risk of failure; like If the corresponding intelligent controllable terminal is deemed to have no fault risk, the prediction model optimization strategy will be executed based on the comparative analysis results of the actual power consumption and the predicted power consumption. Predictive model optimization strategies include: pass Obtain the ratio of systematic bias to random bias. ; ratio With preset coefficient Compare: like If the number of features is increased, the regularization coefficient will be decreased. like This reduces the number of features and increases the regularization coefficient.

2. The resource allocation method for a hybrid distribution network for new energy transmission as described in claim 1, characterized in that, Historical electricity consumption data includes: historical electricity consumption records of the intelligent controllable terminal; real-time operating status information includes: current, voltage, power, and temperature during device operation; environmental data includes: temperature, humidity, and weather conditions of the operating environment of the intelligent controllable terminal; user behavior data includes: frequency, time, and preferences of user device use.

3. The resource allocation method for a hybrid distribution network for new energy transmission as described in claim 2, characterized in that, The process of data preprocessing and feature extraction includes: The historical electricity consumption data, real-time operating status information data, environmental data, and user behavior data of the intelligent controllable terminal are cleaned and normalized to extract features, including: time features, statistical features, environmental features, and user behavior features.

4. A resource allocation system for a hybrid distribution network for transmitting new energy to other regions, characterized in that, The resource allocation system is used to execute the resource allocation method for a hybrid distribution network for new energy transmission as described in any one of claims 1 to 3, wherein the resource allocation system includes: The data acquisition module is used to collect historical power consumption data, real-time operating status information data, environmental data, and user behavior data from the intelligent controllable terminal, and to preprocess the data. The power consumption prediction module is used to extract features from the data and, based on these features, use a trained prediction model to make phased predictions of the total power consumption of the smart controllable terminal. The power supply control module is used to dynamically adjust the power supply based on the forecast results and the grid load; The fault analysis module is used to evaluate the accuracy of electricity consumption forecasts and establish fault analysis strategies and prediction model optimization strategies for intelligent controllable terminals based on the evaluation results. The optimization module is used to execute optimization strategies for the prediction model.

Images