Charging Pile Scheduling Method and System Considering Grid Carbon Emission Factors

Abstract

The present invention relates to the technical field of charging pile power scheduling, and discloses a charging pile scheduling method and system considering the carbon emission factor of the power grid, including interacting to obtain a historical data sequence, and setting a power scheduling strategy for the charging pile on the scheduling day according to the historical data sequence; wherein, the historical data sequence includes a historical carbon emission factor sequence, a historical power grid load sequence, a historical charging load sequence and a historical weather data sequence; monitoring the carbon emission factor and the power grid load of the power grid on the scheduling day, and inputting them into a pre-constructed carbon-electricity coupling model to obtain the real-time power allocation value of the charging pile; correcting the power scheduling strategy based on the real-time power allocation value, and executing the charging pile scheduling according to the corrected power scheduling strategy. The method and system of the present invention introduce the carbon emission factor, monitor the power grid condition in real time and dynamically adjust the scheduling strategy, so as to achieve low-carbon charging, optimize the power grid load distribution, and improve the flexibility of scheduling.

Description

Technical Field

[0001] The present invention relates to the technical field of charging pile power scheduling, and in particular to a charging pile scheduling method and system considering the carbon emission factor of the power grid. Background Art

[0002] With the rapid popularization of electric vehicles, the large-scale access of charging piles has posed new challenges to the operation of the power grid. Traditional charging pile scheduling methods mainly perform power distribution based on time-of-use electricity prices or power grid load balancing. For example, centralized charging is guided during low-price periods of electricity to reduce energy consumption costs, or the charging power is dynamically adjusted according to load forecasts to avoid local overload. The charging pile scheduling strategy does not incorporate the carbon emission factor of the power grid into the decision-making system, and does not establish a dynamic mapping relationship between charging power and carbon emissions, making it difficult to quantify the carbon emission reduction contribution of the scheduling strategy. As a result, charging loads are concentrated during high-carbon emission periods (such as when the proportion of coal-fired power generation is high), exacerbating the carbon footprint of the power grid.

[0003] In addition, existing methods mostly adopt fixed power distribution modes or static scheduling based on day-ahead forecasts, and cannot respond in real time to the dynamic changes of the carbon emission factor and load of the power grid. For example, when the output of wind power and photovoltaic power suddenly increases, resulting in a sharp drop in the carbon emission factor, the charging pile power cannot be increased in time to utilize clean energy, causing waste of resources. The day-ahead scheduling is separated from real-time control, resulting in a significant deviation between the scheduling plan and the actual operation.

[0004] In response to the above problems, several improvement schemes have been proposed in the academic community in recent years. The patent with the publication number CN114726024A realizes dynamic scheduling by constructing a load-electricity price linkage model, but still takes economy as the core goal; the literature "Coordinated Scheduling of Electric Vehicles in a Low-Carbon Power Grid" introduces carbon emission constraints, but its static weight distribution mechanism is difficult to adapt to the real-time changes of the power grid energy structure. Summary of the Invention

[0005] To this end, the technical problem to be solved by the present invention is to overcome the problems of the lack of carbon emission factors and insufficient dynamic response ability in the existing charging pile scheduling strategy, and propose a charging pile scheduling method and system considering the carbon emission factor of the power grid. By introducing the carbon emission factor, the power grid condition is monitored in real time and the scheduling strategy is dynamically adjusted to achieve low-carbon charging, optimize the power grid load distribution, and improve the flexibility of scheduling.

[0006] In a first aspect, to solve the above technical problem, the present invention provides a charging pile scheduling method considering the carbon emission factor of the power grid, including,

[0007] Interactively obtain a historical data sequence, and set the power scheduling strategy of the charging pile on the scheduling day according to the historical data sequence; wherein, the historical data sequence includes a historical carbon emission factor sequence, a historical power grid load sequence, a historical charging load sequence, and a historical weather data sequence;

[0008] Monitor the carbon emission factor and grid load of the power grid on the scheduling day, and input them into a pre-constructed carbon-electricity coupling model to obtain the real-time power allocation value of the charging pile;

[0009] Modify the power scheduling strategy based on the real-time power allocation value, and execute the charging pile scheduling according to the modified power scheduling strategy.

[0010] In an embodiment of the present invention, set the power scheduling strategy of the charging pile on the scheduling day according to the historical data sequence, including,

[0011] Construct and train a prediction model;

[0012] The prediction model obtains the prediction result of the scheduling day according to the historical data sequence; wherein, the prediction result includes the predicted value of the carbon emission factor, the predicted value of the grid load, and the predicted value of the charging load;

[0013] Set the power scheduling strategy according to the prediction result.

[0014] In an embodiment of the present invention, setting the power scheduling strategy further includes,

[0015] Divide the scheduling day into multiple time periods;

[0016] According to the prediction result of the prediction model, set the basic power allocation value for each time period respectively, and the basic power allocation values of each time period constitute the power scheduling strategy;

[0017] Among them, set the basic power allocation value of each time period according to the following method:

[0018] ;

[0019] represents the basic power allocation value of charging pile i in the t-th time period; i represents the charging pile number; t represents the time period number; represents the maximum power of charging pile i; represents the maximum grid load capacity; represents the predicted value of the grid load in the t-th time period; N represents the number of charging piles; represents the predicted value of the charging load of charging pile i in the t-th time period; represents the time period length.

[0020] In an embodiment of the present invention, after setting the basic power allocation value, it further includes,

[0021] Compare the predicted value of the carbon emission factor with the carbon emission factor threshold:

[0022] If the predicted value of the carbon emission factor is less than the first carbon emission factor threshold, the basic power allocation value is increased according to the first factor;

[0023] If the predicted value of the carbon emission factor is greater than the second carbon emission factor threshold, the basic power allocation value is decreased according to the second factor; wherein, the first carbon emission factor threshold is less than the second carbon emission factor threshold;

[0024] The power scheduling strategy is generated from the increased or decreased basic power allocation value;

[0025] Wherein, the first factor A1 and the second factor A2 are respectively:

[0026] ;

[0027] ;

[0028] Wherein, represents the predicted value of the carbon emission factor at time t; and respectively represent the maximum value and the minimum value of the carbon emission factor.

[0029] In an embodiment of the present invention, after setting the basic power allocation value, it further includes adjusting the basic power allocation value according to the user priority of the charging pile, and generating the power scheduling strategy from the adjusted basic power allocation value.

[0030] In an embodiment of the present invention, the prediction model includes a first prediction channel and a second prediction channel,

[0031] The first prediction channel predicts the predicted value of the carbon emission factor on the scheduling day according to the historical carbon emission factor sequence and the historical weather data sequence;

[0032] The second prediction channel predicts the predicted value of the grid load on the scheduling day according to the historical grid load sequence and the historical weather data sequence;

[0033] Wherein, constructing the prediction model includes extracting features from the historical weather data sequence to obtain a shared weather feature vector; wherein, the shared weather feature vector has dual-task adaptability; the dual tasks are the task of predicting the carbon emission factor and the task of predicting the grid load.

[0034] In an embodiment of the present invention, constructing the prediction model further includes caching the shared weather feature vector into a buffer for the first prediction channel and the second prediction channel to call when predicting in their respective channels;

[0035] When a call request is initiated by Prediction Channel 1 or Prediction Channel 2, the shared weather feature vector is preferentially read from the buffer; if the cache is not hit or the data has expired, real-time feature calculation is triggered and the cache is updated.

[0036] In an embodiment of the present invention, constructing the carbon-electricity coupling model includes,

[0037] Taking the carbon emission factor, grid load, and charging pile status information as inputs and the real-time power distribution values of each charging pile as outputs, a carbon-electricity coupling model is constructed;

[0038] Converting the carbon-electricity coupling model into a state space model, and updating the state variables of the state space model based on the Kalman filtering algorithm;

[0039] Taking minimizing the carbon emission factor and maximizing user satisfaction as optimization objectives, and solving the state space model based on the distributed optimization algorithm to obtain the real-time power distribution values of each charging pile in different time periods.

[0040] In an embodiment of the present invention, modifying the power scheduling strategy based on the real-time power distribution value includes,

[0041] Calculating the difference between the predicted value of the carbon emission factor and the monitored value of the carbon emission factor;

[0042] Introducing a carbon emission factor adjustment coefficient ;

[0043] According to the difference and the carbon emission factor adjustment coefficient Adjusting the basic power distribution value;

[0044] Among them, the adjustment formula is:

[0045] ;

[0046] represents the adjusted basic power distribution value of charging pile i at time t according to the real-time power distribution value; represents the difference between the predicted value of the carbon emission factor and the monitored value of the carbon emission factor; represents the predicted value of the carbon emission factor at time t.

[0047] Second, based on the same inventive concept, to solve the above technical problems, the present invention also provides a charging pile scheduling system considering the grid carbon emission factor, including,

[0048] A historical data acquisition module, which is used to interactively obtain a historical data sequence; the historical data sequence includes a historical carbon emission factor sequence, a historical grid load sequence, a historical charging load sequence, and a historical weather data sequence;

[0049] A scheduling strategy generation module, configured to generate a power scheduling strategy for a charging pile on a scheduling day according to the historical data sequence;

[0050] A data monitoring and analysis module, configured to monitor the carbon emission factor data and load data of the power grid on a scheduling day, and input them into a pre-constructed carbon-electricity coupling model to obtain the real-time power distribution value of the charging pile;

[0051] A scheduling execution module, configured to correct the power scheduling strategy according to the real-time power distribution value, and execute the charging pile scheduling according to the corrected power scheduling strategy.

[0052] The above technical solution of the present invention has the following beneficial effects compared with the prior art:

[0053] The charging pile scheduling method and system considering the carbon emission factor of the power grid according to the present invention introduce the carbon emission factor, monitor the power grid condition in real time and dynamically adjust the scheduling strategy, realizing low-carbon charging and optimizing the power grid load distribution; in addition, by monitoring the carbon emission factor and power grid load of the power grid in real time on the scheduling day, it can quickly respond to the changes in the power grid condition, calculate the power distribution value of the charging pile in real time through the carbon-electricity coupling model, and dynamically adjust the power scheduling strategy, thereby improving the flexibility and adaptability of the scheduling, and ensuring that the power scheduling of the charging pile can be adapted to the power grid load fluctuation and carbon emission level in real time.

[0054] By combining the carbon emission factor with the power grid load and adjusting the power distribution in real time, it overcomes the problems of the lack of consideration of carbon emissions and insufficient dynamic response ability in the traditional charging pile scheduling strategy, significantly improves the scheduling efficiency, reduces the carbon emissions, optimizes the power grid load, and at the same time makes a positive contribution to the consumption of green energy and the sustainable development of the power grid. Description of the Drawings

[0055] In order to make the content of the present invention easier to be clearly understood, the following further details the present invention according to the specific embodiments of the present invention and in combination with the drawings, where

[0056] Figure 1 is a flowchart of the charging pile scheduling method considering the carbon emission factor of the power grid in the preferred embodiment of the present invention;

[0057] Figure 2 is a flowchart of setting the power scheduling strategy of the charging pile on the scheduling day according to the historical data sequence in one embodiment of the present invention;

[0058] Figure 3 is a flowchart of setting the power scheduling strategy of the charging pile on the scheduling day according to the historical data sequence in another embodiment of the present invention;

[0059] Figure 4Flow chart for constructing the carbon - electricity coupling model in the preferred embodiment of the present invention;

[0060] Figure 5 Structural block diagram of the charging pile scheduling system considering the carbon emission factor of the power grid in the preferred embodiment of the present invention. Detailed implementation manners

[0061] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand the present invention and be able to implement it. However, the specific embodiments given are not intended to limit the present invention.

[0062] Embodiment 1

[0063] Referring to Figure 1 As shown, the embodiment of the present invention discloses a charging pile scheduling method considering the carbon emission factor of the power grid, including:

[0064] S100. Interactively obtain the historical data sequence, and set the power scheduling strategy of the charging pile on the scheduling day according to the historical data sequence; wherein, the historical data sequence includes the historical carbon emission factor sequence, the historical power grid load sequence, the historical charging load sequence, and the historical weather data sequence;

[0065] S200. Monitor the carbon emission factor and the power grid load of the power grid on the scheduling day, and input them into the pre - constructed carbon - electricity coupling model to obtain the real - time power distribution value of the charging pile;

[0066] S300. Modify the power scheduling strategy based on the real - time power distribution value, and perform the charging pile scheduling according to the modified power scheduling strategy.

[0067] For the charging pile scheduling method and system considering the carbon emission factor of the power grid of the present invention, by introducing the carbon emission factor, monitoring the power grid condition in real time and dynamically adjusting the scheduling strategy, low - carbon charging and optimized power grid load distribution are realized; in addition, by monitoring the carbon emission factor and the power grid load of the power grid in real time on the scheduling day, the change of the power grid condition can be quickly responded to. The real - time power distribution value of the charging pile is calculated through the carbon - electricity coupling model, and the power scheduling strategy is dynamically adjusted, so as to improve the flexibility and adaptability of the scheduling, and ensure that the power scheduling of the charging pile can be adapted to the power grid load fluctuation and the carbon emission level in real time.

[0068] In specific application scenarios, historical data sequences for interaction can be obtained through API interfaces. By docking with system interfaces of power system operators, weather forecast services, carbon emission data providers, etc., historical data can be automatically acquired. For example, historical carbon emission factor sequences (sampled every 15 minutes), historical grid load sequences (sampled hourly), and historical charging load sequences (recorded by charging piles and sampled hourly) can be obtained from the grid data center; historical weather data sequences (temperature, humidity, wind speed, light intensity, etc., sampled hourly) can be obtained from the meteorological department.

[0069] Historical data sequences help identify the variation patterns of grid loads, the fluctuation patterns of carbon emission factors, the spatio-temporal distribution of charging demands, and the impact of climate factors on charging behaviors. Through the accumulation of historical data, optimized strategies for charging pile scheduling are set based on historical rules. By analyzing historical weather data, the impact of weather on grid and charging pile demands can be predicted more accurately. The introduction of weather data enhances the accuracy of scheduling, especially for power demand changes in high-temperature or cold weather. Combining historical charging load data, the charging demands during specific periods can be analyzed, and scheduling strategies can be optimized to make the charging load more balanced, thereby avoiding grid load fluctuations.

[0070] On the scheduling day, the carbon emission factors and grid loads of the power grid are monitored in real time. The monitoring results of grid loads and carbon emission factors are used as inputs and transmitted to a pre-constructed carbon-electricity coupling model. Through the analysis of real-time data, the carbon-electricity coupling model calculates the real-time power distribution values of the charging piles. By correcting the scheduling strategy based on the real-time power distribution values, the charging piles can adapt to the changes in grid loads and carbon emission factors in real time, rather than following fixed preset schemes, achieving the combination of day-ahead scheduling and real-time control, reducing the deviation between the scheduling plan and actual operation, and enhancing the flexibility of the charging pile scheduling strategy.

[0071] In a further implementation plan, as shown in Figure 2 The power scheduling strategy for charging piles on the scheduling day set according to the historical data sequence includes: constructing and training a prediction model; the prediction model obtaining the prediction results for the scheduling day based on the historical data sequence; where the prediction results include predicted values of carbon emission factors, grid loads, and charging loads; setting the power scheduling strategy according to the prediction results.

[0072] In specific implementation, the prediction model adopts a multi-task prediction model architecture. The multi-tasks include carbon emission factor prediction, power grid load prediction, and charging load prediction. The multi-task prediction model architecture includes a carbon emission factor prediction branch, a power grid load prediction branch, and a charging load prediction branch. Among them, the carbon emission factor prediction branch adopts an autoregressive integrated moving average (ARIMA) model, the power grid load prediction branch adopts a long short-term memory network (LSTM) model, and the charging load prediction branch adopts an extreme gradient boosting (XGBoost) model. The prediction model should be understood as obtaining the variation law of the power grid load, the fluctuation pattern of the carbon emission factor, and the spatio-temporal distribution of the charging demand through deep learning of the historical data sequence, and predicting the carbon emission factor, the power grid load, and the charging load on the scheduling day according to the learning results.

[0073] Referring to Figure 2 As shown, in an embodiment of the present invention, setting the power scheduling strategy further includes: dividing the scheduling day into multiple time periods; according to the prediction results of the prediction model, setting basic power allocation values for each of the time periods respectively, and the basic power allocation values of each of the time periods constitute the power scheduling strategy.

[0074] That is to say, by subdividing the scheduling day and setting the basic power allocation for each time period based on the prediction model, a complete scheduling strategy is finally formed. In specific application scenarios:

[0075] Step 1: Divide the scheduling day into multiple time periods: for example, divide the scheduling day into 96 time periods, with each 15 minutes as a time period; of course, the rules for dividing the time periods can also be set according to the operating characteristics of the power grid, user charging behaviors, and historical data analysis results, etc. For example, more specific divisions can be made according to peak and valley demand periods: morning peak, evening peak, night valley, etc. Subdividing the scheduling day into multiple time periods can adjust the charging strategy more refinedly according to the real-time or predicted state of the power grid, and achieve more refined management of the charging strategy.

[0076] Step 2: Set the basic power allocation value for each time period according to the prediction results of the prediction model; set the basic power for each time period through accurate prediction data to ensure the effective use of power resources when they are most needed. For example, increase charging activities during periods with lower carbon emissions, and vice versa, reduce charging activities when carbon emissions are higher.

[0077] Step 3: The basic power allocation values of each of the time periods constitute the power scheduling strategy. The power allocation strategies of each time period are summarized into a complete scheduling strategy for a day, ensuring the orderliness and coherence of the charging strategy from morning to night.

[0078] Specifically, the basic power allocation value for each time period is set according to the following method:

[0079] ;

[0080] represents the basic power allocation value of charging pile i in period t; i represents the charging pile serial number; t represents the period serial number; represents the maximum power of charging pile i; represents the maximum load capacity of the power grid; represents the power grid load prediction value in period t; N represents the number of charging piles; represents the charging load prediction value of charging pile i in period t; represents the period length.

[0081] Referring to Figure 3 as shown, in another embodiment of the present invention, setting the power scheduling strategy further includes, after setting the basic power allocation value, comparing the predicted carbon emission factor with the carbon emission factor threshold:

[0082] If the predicted carbon emission factor is less than the first carbon emission factor threshold, increase the basic power allocation value according to the first factor; if the predicted carbon emission factor is greater than or equal to the first carbon emission factor threshold and less than or equal to the second carbon emission factor threshold, set the basic power allocation value according to the prediction result of the prediction model; if the predicted carbon emission factor is greater than the second carbon emission factor threshold, reduce the basic power allocation value according to the second factor; wherein, the first carbon emission factor threshold is less than the second carbon emission factor threshold;

[0083] Generate the power scheduling strategy from the increased or decreased basic power allocation value;

[0084] Wherein, the first factor A1 and the second factor A2 are respectively:

[0085] ;

[0086] ;

[0087] Wherein, represents the predicted carbon emission factor in period t; and respectively represent the maximum value and the minimum value of the carbon emission factor.

[0088] In specific application scenarios, the first carbon emission factor threshold and the second carbon emission factor threshold are key parameters for determining whether to adjust the basic power allocation value, and can be obtained through various methods, such as historical data analysis method, standards provided by the power grid operator, or empirical rules and industry practices, etc. Among them, the first carbon emission factor threshold represents a lower emission level (i.e., "green" operation state), while the second carbon emission factor threshold represents a medium emission level.

[0089] When the predicted value of the carbon emission factor on the scheduling day is less than the first carbon emission factor threshold, it indicates that the predicted carbon emission factor on the scheduling day is relatively low, meaning that the power grid is at a relatively low carbon emission level. Usually, at this time, the supply of renewable energy is relatively large and the grid load is relatively light. At this time, increasing the basic power allocation value according to the first factor A1 can maximize the utilization of green energy and promote environmental protection without increasing carbon emissions; when the predicted value of the carbon emission factor on the scheduling day is greater than the second carbon emission factor threshold, it indicates that the predicted carbon emission factor on the scheduling day is relatively high, and it may be in a state of high load or a relatively high proportion of fossil energy. At this time, reducing the basic power allocation value according to the second factor A2 and reducing charging during high carbon emission periods can reduce the environmental burden; when the predicted value of the carbon emission factor on the scheduling day is between the first carbon emission factor threshold and the second carbon emission factor threshold, it indicates that the predicted carbon emission factor on the scheduling day is in a suitable state. At this time, directly adopt the basic power allocation value set according to the prediction result. By setting the carbon emission factor threshold and comparing it with the carbon emission factor prediction day, the basic power allocation value can be made more accurate. Optimizing the charging pile scheduling in this way can not only reduce carbon emissions but also optimize the grid load management.

[0090] It should be noted that after setting the basic power allocation value according to the prediction result, adjust the power according to the first factor or the second factor on the basis of this power allocation value, and use the adjusted power as the basic power allocation value for subsequent scheduling. If the predicted value of the carbon emission factor is greater than or equal to the first carbon emission factor threshold and less than or equal to the second carbon emission factor threshold, then the basic power allocation value set according to the prediction result of the prediction model is directly used as the basic power allocation value for subsequent scheduling.

[0091] Furthermore, after setting the basic power allocation value, it also includes adjusting the basic power allocation value according to the user priority of the charging pile, and generating the power scheduling strategy from the adjusted basic power allocation value. The user priorities include that emergency service vehicles (such as ambulances and fire trucks) have the highest priority, public transportation (such as electric buses and taxis) have a relatively high priority, personal electric vehicles, electric delivery vehicles, etc. have a medium priority, and rental and shared vehicles have a low priority, etc. Of course, the user priorities support customization. Adjusting the basic power allocation value according to the user priority makes the scheduling strategy more flexible.

[0092] In a further improvement of the embodiment of the present invention, the prediction model includes a first prediction channel, a second prediction channel, and a third prediction channel. The first prediction channel predicts the predicted value of the carbon emission factor on the scheduling day according to the historical carbon emission factor sequence and the historical weather data sequence; the second prediction channel predicts the predicted value of the grid load on the scheduling day according to the historical grid load sequence and the historical weather data sequence; the third prediction channel predicts the predicted value of the charging load on the scheduling day according to the historical charging load sequence.

[0093] Both the carbon emission factor prediction and the power grid load prediction are related to the historical weather data series. In order to reduce computational redundancy and improve the operating efficiency of the prediction model, constructing the prediction model includes extracting features from the historical weather data series to obtain a shared weather feature vector; wherein the shared weather feature vector has dual-task adaptability; the dual tasks are the task of predicting the carbon emission factor and the task of predicting the power grid load.

[0094] In the specific application scenario, the prediction model is configured with a unified feature extraction module. First, the historical weather data sequence is reduced in dimension through the principal component analysis (PCA) algorithm to obtain a low-dimensional feature vector. The unified feature extraction module has a shared encoder, which takes the low-dimensional feature vector as input. During training, the loss functions of carbon emission prediction and power grid load prediction are optimized at the same time, forcing the encoder to extract features that are explanatory for both tasks. At the same time, a multi-head attention layer is added to the shared encoder to automatically identify weather features that are important for both tasks. SHAPShapley (Additive Explanations) or LIME (Local Interpretable Model-agnostic Explanations) is used to quantify the contribution of each weather feature to the two tasks, and the encoder structure is adjusted to enhance the expression of shared important features. A task discriminator is added after the encoder, and adversarial training is used to force the feature vector to be unable to distinguish the task type, thereby ensuring the universality of the features, and finally obtaining a shared weather feature vector. The shared weather feature vector has dual-task adaptability for predicting carbon emission factors and predicting power grid loads. The shared weather feature vector is called by prediction channel one and prediction channel two, thereby reducing computational redundancy and improving the operating efficiency of the prediction model.

[0095] Furthermore, constructing the prediction model also includes executing the caching of the shared weather feature vector to a cache for the prediction channel one and the prediction channel two to call when predicting their respective channels; wherein, when the prediction channel one or the prediction channel two initiates a call request, the shared weather feature vector is preferentially read from the cache; if the cache misses or the data expires, real-time feature calculation is triggered and the cache is updated. By setting up a cache mechanism, millisecond-level response to prediction requests is achieved, supporting high-concurrency scenarios; in addition, the cache mechanism can be horizontally expanded to multi-regional power grid systems. The cache mechanism combined with weather feature sharing can effectively reduce more than 60% of repeated calculations, reduce computing resource consumption, and ensure prediction accuracy and real-time performance.

[0096] In other embodiments of the present invention, referring to Figure 4As shown, constructing the carbon-electricity coupling model includes taking the carbon emission factor, grid load, and charging pile status information as inputs and the real-time power distribution value of each charging pile as the output to construct the carbon-electricity coupling model; converting the carbon-electricity coupling model into a state-space model and updating the state variables of the state-space model based on the Kalman filtering algorithm; taking minimizing the carbon emission factor and maximizing user satisfaction as the optimization objectives, and solving the state-space model based on the distributed optimization algorithm to obtain the real-time power distribution values of each charging pile in different time periods. The solution of the present invention can estimate the system state in real time through the Kalman filter, can handle the uncertainties brought by the fluctuations of the power grid and carbon emission factors, has higher adaptability to load changes and carbon emission prediction errors, thus effectively avoiding the occurrence of power grid overload and high carbon emissions, and improving the stability and security of the power grid. Different from the traditional way that relies on centralized calculation, this solution adopts a distributed optimization algorithm, which can perform parallel calculations on multiple charging pile nodes, greatly reducing the calculation amount required by the centralized optimization method and being able to give real-time power distribution decisions more quickly.

[0097] In other embodiments of the present invention, modifying the power scheduling strategy based on the real-time power distribution value includes calculating the difference between the predicted carbon emission factor value and the monitored carbon emission factor value; introducing a carbon emission factor adjustment coefficient ; adjusting the basic power distribution value according to the difference and the carbon emission factor adjustment coefficient ; where the adjustment formula is:

[0098] ;

[0099] represents the adjusted basic power distribution value of charging pile i at time t according to the real-time power distribution value; represents the difference between the predicted carbon emission factor value and the monitored carbon emission factor value; represents the predicted carbon emission factor value at time t.

[0100] In the solution of the embodiment of the present invention, by calculating the difference between the predicted value and the actual monitored value, the carbon emission factor prediction error is sensed in real time, and the power distribution of the charging pile is adjusted through the difference. This mechanism can quickly correct the deviation caused by the prediction error, ensure that the power scheduling always matches the actual situation of the power grid, and avoid energy waste or excessive carbon emissions caused by inaccurate prediction.

[0101] Embodiment 2

[0102] Based on the same inventive concept, as shown in Figure 5 , the embodiment of the present invention provides a charging pile scheduling system considering the carbon emission factor of the power grid, including

[0103] A historical data acquisition module, which is used to interactively obtain a historical data sequence; the historical data sequence includes a historical carbon emission factor sequence, a historical power grid load sequence, a historical charging load sequence, and a historical weather data sequence;

[0104] A scheduling strategy generation module, which is used to generate a power scheduling strategy for the charging pile on the scheduling day according to the historical data sequence;

[0105] A data monitoring and analysis module, which is used to monitor the carbon emission factor data and load data of the power grid on the scheduling day, and input them into a pre-constructed carbon-electricity coupling model to obtain the real-time power allocation value of the charging pile;

[0106] A scheduling execution module, which is used to correct the power scheduling strategy according to the real-time power allocation value, and execute the charging pile scheduling according to the corrected power scheduling strategy.

[0107] The embodiment of the present invention is used to execute the above-mentioned charging pile scheduling method considering the carbon emission factor of the power grid, and has the same technical effect in the process of solving the technical problem, which will not be elaborated here.

[0108] In summary, for the charging pile scheduling method and system considering the carbon emission factor of the power grid of the present invention, by introducing the carbon emission factor, monitoring the power grid condition in real time and dynamically adjusting the scheduling strategy, low-carbon charging and optimized power grid load distribution are realized; in addition, on the scheduling day, the carbon emission factor and power grid load of the power grid are monitored in real time, which can quickly respond to the changes in the power grid condition, calculate the power allocation value of the charging pile in real time through the carbon-electricity coupling model, and dynamically adjust the power scheduling strategy, so as to improve the flexibility and adaptability of the scheduling, and ensure that the power scheduling of the charging pile can be adapted to the power grid load fluctuation and carbon emission level in real time.

[0109] Combining the carbon emission factor with the power grid load and adjusting the power allocation in real time overcomes the problems of lack of carbon emission consideration and insufficient dynamic response ability in the traditional charging pile scheduling strategy, significantly improves the scheduling efficiency, reduces the carbon emission, optimizes the power grid load, and at the same time makes a positive contribution to the consumption of green energy and the sustainable development of the power grid.

[0110] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0111] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing device produce means for implementing the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 means for implementing the functions specified in one or more blocks.

[0112] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means for implementing the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 means for implementing the functions specified in one or more blocks.

[0113] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operational steps are performed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 means for implementing the functions specified in one or more blocks.

[0114] Obviously, the above embodiments are merely examples for clear illustration and are not limitations on the implementation manners. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to enumerate all implementation manners here. And the obvious changes or modifications derived therefrom are still within the protection scope of the present invention.

Claims

1. A charging pile scheduling method taking into account the carbon emission factor of the power grid, characterized by: include, Interactively obtaining a historical data sequence, wherein the historical data sequence includes a historical carbon emission factor sequence, a historical power grid load sequence, a historical charging load sequence, and a historical weather data sequence; The power dispatching strategy of the charging pile on the dispatching day is set according to the historical data sequence: it includes constructing and training a prediction model, the prediction model obtains the prediction result of the dispatching day according to the historical data sequence; wherein the prediction result includes the carbon emission factor prediction value, the power grid load prediction value and the charging load prediction value, and the power dispatching strategy is set according to the prediction result; On the scheduling day, the carbon emission factor and the load of the power grid are monitored, and a pre-built carbon-electricity coupling model is input to obtain a real-time power allocation value of the charging pile; Modify the power scheduling strategy based on the real-time power allocation value, and execute the charging pile scheduling according to the modified power scheduling strategy; Wherein, setting the power scheduling strategy also includes: dividing the scheduling day into a plurality of time periods; According to the prediction result of the prediction model, a basic power allocation value is set for each of the time periods, and the power scheduling strategy is formed by the basic power allocation value of each of the time periods; The basic power allocation value for each time period is set according to the following method: ； Indicates the basic power allocation value of charging pile i in time period t; i represents the charging pile number; t represents the time period number; Indicates the maximum power of charging pile i; Indicates the maximum load capacity of the power grid; represents the predicted value of the grid load during period t; N represents the number of charging piles; represents the predicted value of charging load of charging pile i in period t; Indicates the length of the time period.

2. The charging pile scheduling method taking into account the carbon emission factor of the power grid according to claim 1 is characterized in that: After setting the basic power allocation value, the method further includes: Compare the predicted carbon emission factor with the carbon emission factor threshold: If the predicted value of the carbon emission factor is less than the carbon emission factor threshold value 1, increasing the basic power allocation value according to the first factor; If the predicted value of the carbon emission factor is greater than the second carbon emission factor threshold, the basic power allocation value is reduced according to the second factor; wherein the first carbon emission factor threshold is less than the second carbon emission factor threshold; Generating the power scheduling strategy based on the increased or decreased basic power allocation value; The first factor A1 and the second factor A2 are respectively: ； ； in, represents the predicted value of carbon emission factor in period t; and Represent the maximum and minimum values ​​of the carbon emission factor respectively.

3. The charging pile scheduling method taking into account the carbon emission factor of the power grid according to claim 1 or 2, characterized in that: After setting the basic power allocation value, the method further includes adjusting the basic power allocation value according to the user priority of the charging pile, and generating the power scheduling strategy based on the adjusted basic power allocation value.

4. The charging pile scheduling method taking into account the carbon emission factor of the power grid according to claim 1 is characterized in that: The prediction model includes prediction channel one and prediction channel two. Prediction channel 1 predicts the predicted value of the carbon emission factor on the scheduling day based on the historical carbon emission factor sequence and the historical weather data sequence; Prediction channel 2 predicts the grid load forecast value of the dispatching day according to the historical grid load sequence and the historical weather data sequence; Among them, constructing the prediction model includes extracting features from the historical weather data sequence to obtain a shared weather feature vector; wherein the shared weather feature vector has dual-task adaptability; the dual tasks are the task of predicting carbon emission factors and the task of predicting power grid loads.

5. The charging pile scheduling method taking into account the carbon emission factor of the power grid according to claim 4 is characterized in that: Constructing the prediction model further includes executing caching the shared weather feature vector into a cache for the prediction channel 1 and the prediction channel 2 to call when predicting their respective channels; Among them, when prediction channel 1 or prediction channel 2 initiates a call request, the shared weather feature vector is read from the cache first; if the cache misses or the data expires, real-time feature calculation is triggered and the cache is updated.

6. The charging pile scheduling method taking into account the carbon emission factor of the power grid according to claim 1 is characterized in that: Constructing the carbon-electric coupling model includes: The carbon-electricity coupling model is constructed with carbon emission factors, grid load and charging pile status information as input and the real-time power allocation value of each charging pile as output; Converting the carbon-electric coupling model into a state-space model, and updating state variables of the state-space model based on a Kalman filter algorithm; The optimization goals are to minimize the carbon emission factor and maximize user satisfaction, and the state space model is solved based on the distributed optimization algorithm to obtain the real-time power allocation value of each charging pile in different time periods.

7. A charging pile scheduling system taking into account the carbon emission factor of the power grid, used to execute the charging pile scheduling method taking into account the carbon emission factor of the power grid as claimed in any one of claims 1 to 6, characterized in that: include, A historical data acquisition module, which is used to interactively obtain a historical data sequence; the historical data sequence includes a historical carbon emission factor sequence, a historical power grid load sequence, a historical charging load sequence and a historical weather data sequence; A scheduling strategy generation module, used to generate a power scheduling strategy for the charging pile on the scheduling day according to the historical data sequence; A data monitoring and analysis module, used to monitor the carbon emission factor data and load data of the power grid on the dispatching day, and input a pre-built carbon-electricity coupling model to obtain the real-time power allocation value of the charging pile; The scheduling execution module is used to modify the power scheduling strategy according to the real-time power allocation value, and execute the charging pile scheduling according to the modified power scheduling strategy.

Images