Resident charging pile adjustable resource prediction method and device, equipment and storage medium
By collecting and preprocessing residents' charging pile data, combining factors such as temperature, weather, date type and electricity price, using Spearman correlation coefficient and gray correlation coefficient quantitative influences, and combining SARIMA model for prediction, the problem of insufficient prediction accuracy in the existing technology is solved, and higher prediction accuracy is achieved.
Patent Information
- Application Number
- CN202510021561.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-07
- Publication Date
- 2025-06-03
AI Technical Summary
The existing method of adjustable resource prediction for residents' charging piles has limitations in comprehensively considering a variety of influencing factors, which makes it difficult to guarantee the accuracy of prediction.
A method of adjustable resource prediction for residents' charging piles is adopted to collect and preprocess charging pile data, and to obtain adjustable resource influencing factors including temperature, weather, date type and electricity price. The Spearman correlation coefficient and gray correlation coefficient were used to quantify the effects of each factor and predicted in combination with the SARIMA model.
The accuracy of the charging pile's adjustable resource prediction is improved, and the role of various factors can be considered more accurately, thereby improving the reliability of the prediction results.
Smart Images
Figure CN120088001A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of power systems, and in particular to a method, device, equipment and storage medium for predicting adjustable resources of residential charging piles. Background Art
[0002] With the rapid growth of the number of private electric vehicles, residents' demand for the installation and electricity consumption of private charging piles has continued to rise. With their excellent mobility and potential for interaction with the power system, electric vehicles are expected to participate in temporary auxiliary services of the power grid and achieve orderly power consumption, thereby enhancing the reliability and stability of the power system and bringing significant benefits to society and the economy.
[0003] However, the load characteristics of residential charging piles are quite complex. They are regarded as a special time series data, showing nonlinear characteristics. The load level of charging piles is susceptible to a variety of external factors, including meteorological conditions (such as weather changes, temperature fluctuations, etc.) and electricity price policies. In addition, due to the large number and wide distribution of residential charging piles, they are scattered and disordered in time and space. Coupled with the uncertainty of user charging behavior, it is particularly difficult to evaluate the capacity of residential charging piles as an orderly adjustable resource for electricity consumption.
[0004] At present, the prediction of adjustable resources of charging piles mainly relies on the prediction method based on statistical learning, which estimates the future charging load by modeling and analyzing the historical data of charging piles. Although this method is easy to understand and implement, it has limitations in fully considering multiple influencing factors. For example, key factors such as meteorological conditions and electricity price policies are often difficult to be fully incorporated into the model. At the same time, the accuracy of the prediction results is also restricted by the data quality and sample quantity, making it difficult to effectively guarantee the accuracy of the prediction of adjustable resources of charging piles.
[0005] Therefore, it is urgent to explore and propose an improved method for predicting the adjustable resources of residents' charging piles to improve the accuracy of the prediction of the adjustable resources of charging piles. Summary of the invention
[0006] In order to overcome the deficiencies of the prior art, the purpose of the present application is to provide a method for predicting adjustable resources of residential charging piles, which can improve the accuracy of prediction of adjustable resources of charging piles.
[0007] The purpose of this application is achieved by the following technical solutions: In a first aspect, the present application provides a method for predicting adjustable resources of residential charging piles, comprising: Collect residential charging pile data and preprocess the residential charging pile data, where the residential charging pile data includes at least one of the time when the maximum charging load appears daily, the value of the maximum charging load, the distribution of charging start times, the distribution of charging end times, the distribution of charging durations, the distribution of charging stay durations, the highest and lowest temperatures of the day. Obtain adjustable resource influencing factors, where the adjustable resource influencing factors include at least one of temperature, weather, date type, and electricity price. Predict the adjustable resources of residential charging piles based on the residential charging pile data and the adjustable resource influencing factors, where the adjustable resources of residential charging piles include adjustable load and adjustable duration.
[0008] In one possible implementation, the obtaining of the adjustable resource influencing factors includes: Use the method of calculating the Spearman correlation coefficient to obtain the factors influencing the adjustable resources. Use the method of calculating the grey relational grade coefficient to obtain the weights of the factors influencing the adjustable resources. Determine the adjustable resource influencing factors based on the factors influencing the adjustable resources and the weights of the factors influencing the adjustable resources.
[0009] In one possible implementation, the predicting of the adjustable resources of residential charging piles based on the residential charging pile data and the adjustable resource influencing factors includes: Normalize the residential charging pile data to obtain normalized data. Predict the adjustable resources of residential charging piles based on the normalized data and the adjustable resource influencing factors.
[0010] In one possible implementation, the predicting of the adjustable resources of residential charging piles based on the normalized data and the adjustable resource influencing factors includes: Determine whether the normalized data is abnormal. If not, input the normalized data and the adjustable resource influencing factors into a preset prediction model to obtain the predicted adjustable resources of residential charging piles.
[0011] In one possible implementation, the predicting of the adjustable resources of residential charging piles based on the normalized data and the adjustable resource influencing factors further includes: If so, correct the normalized data using similar day data, and input the corrected normalized data and the adjustable resource influencing factors into a preset prediction model to obtain the predicted adjustable resources of residential charging piles.
[0012] In one possible implementation, the prediction model is constructed based on the SARIMA model.
[0013] In a second aspect, the present application further provides a device for predicting adjustable resources of residential charging piles, including: A collection module, configured to collect residential charging pile data and preprocess the residential charging pile data. The residential charging pile data includes at least one of the time when the maximum charging load appears daily, the value of the maximum charging load, the distribution of charging start times, the distribution of charging end times, the distribution of charging durations, the distribution of charging stay durations, the highest and lowest temperatures of the day. An acquisition module, configured to acquire factors affecting adjustable resources, where the factors affecting adjustable resources include at least one of temperature, weather, date type, and electricity price. A prediction module, configured to predict the adjustable resources of residential charging piles according to the residential charging pile data and the factors affecting adjustable resources. The adjustable resources of residential charging piles include adjustable load and adjustable duration.
[0014] In a third aspect, the present application further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method described in any one of the first aspects above is implemented.
[0015] In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method described in any one of the first aspects above is implemented.
[0016] Compared with the prior art, the beneficial effects of the present application at least include: In the method for predicting adjustable resources of residential charging piles of the present application, by using the residential charging pile data and the factors affecting adjustable resources including at least one of temperature, weather, date type, and electricity price to predict the adjustable resources of residential charging piles, the accuracy of predicting the adjustable resources of charging piles can be improved.
[0017] In the method for predicting adjustable resources of residential charging piles of the present application, by using the method of calculating the Spearman correlation coefficient to obtain the factors affecting adjustable resources, the influence degree of each factor on the adjustable resources of charging piles can be quantified. By using the method of calculating the grey correlation degree coefficient to obtain the weights of the factors affecting adjustable resources, in this way, not only the correlation between each influencing factor and the adjustable resources of charging piles is considered, but also the weights of each influencing factor are further quantified, and the role of each factor can be considered more precisely, thereby improving the prediction accuracy.
[0018] In the method for predicting adjustable resources of residential charging piles in this application, by normalizing the residential charging pile data, data with different dimensions and value ranges can be converted to a unified scale, which helps to eliminate the dimensional differences between the data, making each data feature comparable and additive in the prediction model, thereby improving the accuracy of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] The drawings related to the embodiments of this application will be briefly introduced below.
[0020] Figure 1 is a schematic flowchart of a method for predicting adjustable resources of residential charging piles provided by an embodiment of this application; Figure 2 is a schematic flowchart of step S200 of a method for predicting adjustable resources of residential charging piles provided by an embodiment of this application; Figure 3 is a schematic flowchart of step S300 of a method for predicting adjustable resources of residential charging piles provided by an embodiment of this application; Figure 4 is a schematic flowchart of a method for predicting adjustable resources of residential charging piles provided by another embodiment of this application; Figure 5 is a structural framework diagram of an apparatus for predicting adjustable resources of residential charging piles provided by an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0021] To facilitate the understanding of this application, this application will be described more comprehensively below with reference to the relevant drawings. The preferred embodiments of this application are shown in the drawings. However, this application can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the understanding of the disclosure content of this application more thorough and comprehensive.
[0022] Embodiment 1 Figure 1 is a schematic flowchart of a method for predicting adjustable resources of residential charging piles provided by an embodiment of this application. Refer to Figure 1 As shown, the method may include the following steps: S100, collect residential charging pile data and preprocess the residential charging pile data, where the residential charging pile data includes at least one of the moment when the maximum charging load appears daily, the value of the maximum charging load, the distribution of the charging start time, the distribution of the charging end time, the distribution of the charging duration, the distribution of the charging stay duration, the highest and lowest temperatures on the same day.
[0023] Among them, the moment when the maximum charging load appears daily, that is, the specific time point when the charging demand reaches the peak every day, reflects the charging habits of users and the peak period. The value of the maximum charging load, that is, at the above-mentioned maximum load moment, the charging power or electricity carried by the charging pile, reflects the maximum power support that the power grid needs to provide at this moment. The distribution of the charging start time, that is, to count the time distribution of when users start charging, reflects the charging behavior pattern of users, such as whether they concentrate on charging after returning home at night, etc. The distribution of the charging end time, which corresponds to the start time, can reflect the idle period of the charging pile. The distribution of the charging duration, that is, to record the duration of each charging, can reflect the charging demands of different vehicle models and battery capacities, as well as the charging preferences of users. The distribution of the charging stay duration, that is, the total duration that the vehicle is connected to the charging pile, including the actual charging time and possible waiting time, can reflect the usage efficiency of the charging pile. The highest and lowest temperatures of the day, that is, considering the impact of environmental factors on charging behavior, such as high temperature or low temperature may affect battery performance and thus affect the charging demand.
[0024] Among them, the preprocessing of the residential charging pile data specifically may include: Using the average interpolation method to repair and fill in the abnormal data, where the abnormal data includes null values or 0 values. As a data processing technology, the core idea of the average interpolation method is to use the average value of adjacent data to reasonably estimate and fill in these abnormal values. The specific implementation steps are as follows: First, traverse the entire data set to accurately identify all null values or 0 values, which will be regarded as abnormal data that needs to be repaired.
[0025] For each abnormal data point, it is necessary to find the normal data points before and after it, and these normal data points will form the numerical range for interpolation.
[0026] Using the normal data points within the determined interpolation range, the average value can be calculated. This average value will be used as the estimated value to fill in the abnormal data point.
[0027] Finally, fill the calculated average value into the corresponding abnormal data position, thus completing the data repair and filling work.
[0028] By using the average interpolation method, it is possible to effectively process abnormal data such as null values or 0 values in the data set, ensure the integrity and accuracy of the data set, and lay a solid foundation for subsequent data analysis and processing work. This method is particularly effective when dealing with time series data or continuous numerical data because it can make reasonable estimates and fillings by fully utilizing the correlation and continuity between data.
[0029] Specifically, data collection can be carried out on residential charging piles based on residential communities. The collected data includes one or more of the time when the maximum charging load appears daily, the value of the maximum charging load, the distribution of charging start times, the distribution of charging end times, the distribution of charging durations, the distribution of charging stay durations, and the highest and lowest temperatures of the day. Due to the influence of various random factors, the collected load and load influencing factor data may have missing or abnormal values, such as 0 or null values. If the data with missing or abnormal values is used, the prediction results will be inaccurate to a certain extent, with low accuracy. Therefore, before performing load prediction, the data can be preprocessed and corrected to eliminate data with too short charging times. The average interpolation method can be used to repair and fill in null values, 0 values, or other abnormal data. The calculation formula of the average interpolation method is as follows: In the formula: x i is the data after repair or filling; x i-1 and x i+1 are the data before and after the null value or abnormal data respectively.
[0030] It should be noted that to determine whether the charging time is too short, the charging stay duration and the charging duration need to be obtained.
[0031] Among them, the calculation formula of the charging stay duration is as follows: T s = T end - T start In the formula: T s is the charging stay duration, T end is the moment when the charging plug is unplugged, and T start is the moment when the charging plug is plugged in.
[0032] The calculation formula of the charging duration is as follows: T c = T stop - T charge In the formula: T c is the charging duration, T stop is the charging end moment, and T charge is the charging start moment.
[0033] S200, obtain the influencing factors of adjustable resources, and the influencing factors of adjustable resources include at least one of temperature, weather, date type, and electricity price.
[0034] Among them, temperature is one of the key factors affecting the adjustable resources of residential charging piles. High or low temperature environments may both affect the charging efficiency and performance of electric vehicle batteries, thus indirectly influencing the load demand of charging piles. For example, under extreme weather conditions, the battery may require a longer charging time or a higher charging power to meet the normal operation needs of electric vehicles. Temperature data can usually be obtained from meteorological departments or professional weather forecasting services. These data are usually updated hourly, daily, or weekly, providing real-time or near-real-time temperature information for the prediction of adjustable resources.
[0035] Weather, such as sunny days, rainy days, snowy days, etc., also affects the load demand of residential charging piles. For example, under adverse weather conditions, residents may be more inclined to use electric vehicles instead of traditional fuel vehicles for travel, thus increasing the load demand of charging piles. In addition, weather may also affect the cruising range and charging efficiency of electric vehicles. Weather data can also be obtained from meteorological departments or professional weather forecasting services. These data include multiple indicators such as weather type, precipitation, wind speed, humidity, etc., providing comprehensive weather information for the prediction of adjustable resources.
[0036] Date types, such as weekdays, weekends, holidays, etc., have a significant impact on the load demand of residential charging piles. On weekdays, residents may reduce their charging demand due to going to work or school; while on weekends and holidays, residents may be more inclined to use electric vehicles for leisure activities or long-distance trips, thus increasing the charging demand. Date type data can usually be obtained from calendars or date calculation tools. These data provide a clear time frame and background information for the prediction of adjustable resources.
[0037] Electricity price is an important economic factor affecting the adjustable resources of residential charging piles. By implementing time-of-use electricity price or peak-valley electricity price policies, residents can be guided to charge during periods with lower electricity prices, thus reducing the charging cost and alleviating the load pressure on the power grid. At the same time, electricity price policies may also affect the purchase intention and usage frequency of electric vehicles. Electricity price data can usually be obtained from the official websites of power companies or energy management departments. These data include information such as electricity price levels at different times and electricity price adjustment policies, providing important economic references for the prediction of adjustable resources.
[0038] It should be noted that when predicting the adjustable resources of charging piles, since the adjustable resources of charging piles are significantly affected by various external factors, adding appropriate influencing factors can significantly improve the accuracy of predicting the potential of adjustable resources. Specifically, external factors such as temperature, weather conditions, and date types need to be comprehensively considered. To analyze these impacts more precisely, temperature data, rainfall information, holiday arrangements, and historical load records can be combined, and the Spearman correlation coefficient can be used to extract temperature and weather factors that have a significant impact on the adjustable resources of charging piles. By calculating and analyzing these correlation coefficient values, the correlation degree between each characteristic factor and the adjustable resources of charging piles can be accurately judged.
[0039] S300. Predict the adjustable resources of residential charging piles according to the residential charging pile data and the influencing factors of adjustable resources, where the adjustable resources of residential charging piles include adjustable load and adjustable duration.
[0040] Based on the collected residential charging pile data and the identified influencing factors of adjustable resources, the adjustable resources of residential charging piles can be predicted. Here, the "adjustable resources of residential charging piles" specifically refers to the load amount and charging duration that can be flexibly adjusted to meet the grid demand or optimize the charging strategy, that is, adjustable load and adjustable duration. Specifically, first, according to the characteristics of the data and the prediction target, a suitable prediction model can be selected, which can include time series models (such as SARIMA), machine learning models (such as random forest, neural network), or combined models, and the model is trained using historical data, and the parameters are adjusted to optimize the prediction performance. Then, using the trained model, combined with the current residential charging pile data and real-time information on the influencing factors of adjustable resources, the adjustable load and adjustable duration for a period of time in the future are predicted. Among them, the adjustable load prediction focuses on the variation of the maximum or average charging power that the charging pile can provide in different time periods; the adjustable duration prediction focuses on the ability of the charging pile to flexibly adjust the charging duration within a day or a specific time period.
[0041] The method for predicting the adjustable resources of residential charging piles in this application uses residential charging pile data and influencing factors of adjustable resources including at least one of temperature, weather, date type, and electricity price to predict the adjustable resources of residential charging piles, which can improve the accuracy of predicting the adjustable resources of charging piles.
[0042] In a specific embodiment, the influencing factors of adjustable resources can be obtained by using the method of calculating the Spearman correlation coefficient and the method of calculating the grey correlation degree coefficient.
[0043] Figure 2 It is a schematic flowchart of step S200 of the method for predicting the adjustable resources of residential charging piles provided by an embodiment of this application. See Figure 2As shown, the step S200 may specifically include the following steps: S210, using the method of calculating the Spearman correlation coefficient, obtain the factors affecting the adjustable resources.
[0044] Specifically, calculating the Spearman (i.e., Spearman) correlation coefficient can be achieved through the following formula: In the formula: ρ represents the Spearman correlation coefficient between variables x and y, reflecting the dependence of two adjustable resources on the influencing characteristics.
[0045] x, y are two variables for which the correlation coefficient is to be calculated.
[0046] Taking the example of obtaining temperature as a factor affecting adjustable resources by using the method of calculating the Spearman correlation coefficient, first, collect the charging pile load data and temperature data over a period of time, remove outliers and missing values; then sort the load data and temperature data respectively and assign ranks; then for each pair of load and temperature observations, calculate the difference in their ranks. Use the above formula for calculating the Spearman correlation coefficient to calculate the Spearman correlation coefficient between load and temperature. Suppose the calculated correlation coefficient is 0.8, indicating a strong positive correlation between load and temperature, that is, as the temperature rises, the load of the charging pile also tends to increase.
[0047] S220, using the method of calculating the grey relational grade coefficient, obtain the weights of the factors affecting the adjustable resources.
[0048] When calculating the grey relational grade coefficient, taking the calculation of the correlation coefficient between y(k) and x i (k) as an example, its calculation formula is as follows: In the formula: ξ i (k) is the correlation coefficient; min i min k |x 0 (k)-x i (k)| is the minimum value in the matrix, max i max k |x 0 (k)-x i (k)| is the maximum value in the matrix, and ρ is the resolution coefficient, which can be taken as 0.5.
[0049] Specifically, taking the determination of the influence weights of temperature, weather, date type, and electricity price on the residential charging pile load as an example, first, collect the load data of residential charging piles and the corresponding temperature, weather, date type, and electricity price data over a period of time, and perform standardization processing on the collected data to make them have the same dimension and range. Then, for each influencing factor (comparison sequence), calculate its grey correlation coefficient with the load data (reference sequence) at each moment. Next, calculate the average value of the correlation coefficients of each influencing factor to obtain the grey correlation degree. Finally, based on the grey correlation degrees of each influencing factor, determine its influence weight in the change of the residential charging pile load. For example, assuming that the correlation degree of temperature is the highest, its influence weight is also the largest, indicating that temperature has the most significant influence on the charging pile load.
[0050] S230. Determine the adjustable resource influencing factors according to the factors influencing the adjustable resources and the weights of the factors influencing the adjustable resources.
[0051] Exemplarily, according to the weights obtained in step S220, it is known that electricity price and date type are the key factors affecting the charging station load change, while the influence of temperature and weather is relatively small. Then, electricity price and date type can be finally determined as the adjustable resource influencing factors.
[0052] Figure 3 It is a schematic flowchart of step S300 of the method for predicting adjustable resources of residential charging piles provided by an embodiment of the present application. Refer to Figure 3 As shown, step S300 may include the following steps: S310. Perform normalization processing on the residential charging pile data to obtain normalized data.
[0053] Before predicting the adjustable resources of the charging pile, in view of the dimension difference existing in the load data, the load historical data can be normalized. The normalized data can serve the model training and prediction process more effectively. Specifically, the Min-Max normalization method can be used to perform normalization processing on the residential charging pile data, and the formula is: Exemplarily, the Min-Max normalization method can be selected to perform normalization processing on data such as charging duration, charging start and end times, etc., to ensure that the data is compared under a unified scale.
[0054] S320. Predict the adjustable resources of the residential charging pile according to the normalized data and the adjustable resource influencing factors. Specifically, step S320 may include: S321. Determine whether the normalized data is abnormal. Specifically, reasonable upper and lower threshold values can be set for the normalized data according to historical data or experience. If the data exceeds these thresholds, it is determined as abnormal data. Statistical principles can also be used, such as calculating statistical quantities such as the mean and standard deviation of the data, and then determining whether the data deviates from the normal range. Abnormal detection algorithms (such as Isolation Forest, One-Class SVM, etc.) can also be used to identify abnormal points in the data.
[0055] S322. If not, input the normalized data and the adjustable resource influencing factors into a preset prediction model to obtain the predicted adjustable resources of residential charging piles. Specifically, a suitable prediction model can be selected according to data characteristics and prediction requirements, such as linear regression, time series models (such as ARIMA), neural networks (such as LSTM), etc. Then, the normalized data and adjustable resource influencing factors are used as input features and input into the preset prediction model. Finally, run the prediction model to obtain the prediction result of the adjustable resources of residential charging piles.
[0056] S323. If so, correct the normalized data using similar-day data, and input the corrected normalized data and the adjustable resource influencing factors into a preset prediction model to obtain the predicted adjustable resources of residential charging piles.
[0057] Specifically, similar dates to the current day (such as similar weather conditions, holiday situations, seasons, etc.) can be found from historical data, and the normalized data of these dates can be obtained. Then, replace the abnormal data with the corresponding values of the similar-day data. Here, the similar-day data is historical data that is similar or close to the day to be predicted under specific conditions (such as weather, holidays, etc.). Finally, input the corrected normalized data and adjustable resource influencing factors into the preset prediction model for prediction.
[0058] In summary, in order to improve the prediction accuracy, rich historical data such as weather type, temperature, and date type can be incorporated into the analysis process to conduct a comprehensive abnormal detection of the input data. Once abnormal data is found, use the data of similar days for detailed correction. These corrected data will be used as the key input items of the model to help obtain a more accurate final prediction result. The correction using similar-day data can be achieved through the following formula: In the formula: is the average power of the residential charging pile on the correction day, is the average power of the residential charging pile on the similar day, is the ratio of the total capacity of the residential charging pile on the correction day to that on the similar day, T isThe charging residence duration on the correction date, T js The charging residence duration on the similar date, T ic The charging duration on the correction date, T jc The charging duration on the similar date.
[0059] Among them, the prediction model can be constructed based on the SARIMA model.
[0060] The SARIMA (Seasonal Autoregressive Integrated Moving Average) model consists of four parts: the seasonal (S) part, the autoregressive (AR) part, the differencing (I) part, and the moving average (MA) part. Among them, the seasonality can be any regular fluctuation with a certain period (hourly changes within a day, daily changes within a month, etc.), which can effectively capture and model the periodic changes in the load characteristics of residential charging piles, and achieve accurate prediction of the orderly charging potential of residential charging piles. The expression of the model is as follows: In the formula: is the regression coefficient, φ 1 , φ 2 , …, φ P is the seasonal autoregressive coefficient, θ 1 , θ 2 ,... θ q is the moving average coefficient, Θ 1 , Θ 2 ,... Θ Q is the seasonal moving average coefficient, B is the differencing operator, d is the number of differencing times, D is the number of seasonal differencing times, s is the seasonal cycle length, X t is the observed value at time point t, ε t is the error term.
[0061] The specific parameters can be determined by observing the autocorrelation function (ACF) and partial autocorrelation function (PACF) graphs, or using an automatic parameter optimization tool, which will not be elaborated here.
[0062] Among them, predicting the adjustable resources of residential charging piles, that is, the instantaneous adjustable potential of orderly charging of residential charging piles, can be expressed by the formula: In the formula: P i represents the instantaneous adjustable power of the residential charging pile, P a represents the rated power of the residential charging pile, P b represents the instantaneous active power of the residential charging pile, Wi Represents the total adjustable resource power in a period of time.
[0063] It should be noted that MAE (Mean Absolute Error) and RMSE (Root Mean Square Error) can be selected as the core evaluation indicators to calculate the prediction result error. Among them, MAE focuses on evaluating the error size from the dimension of the absolute error of the prediction result, while RMSE focuses on analyzing the influence degree of extreme abnormal data in the prediction on the overall error. Finally, by statistically analyzing the probability density distribution of the error, the prediction performance of the model can be comprehensively and deeply analyzed and calculated. Regarding the specific calculation methods of these two evaluation indicators, they can be realized through the following formulas respectively: In the formula: MAE and RMSE are the mean absolute percentage error and root mean square error respectively; y i is the actual value, is the predicted value, and n is the number of samples.
[0064] Embodiment 2 Figure 4 is another process schematic diagram of the adjustable resource prediction method for residential charging piles provided by an embodiment of the present application. Refer to Figure 4 As shown, in the adjustable resource prediction method for residential charging piles in the embodiment of the present application, first, the method of calculating the Spearman correlation coefficient is used to accurately identify the key factors affecting the adjustable resources of residential charging piles, including meteorological conditions and temperature conditions. Subsequently, when constructing the adjustable resource prediction model, multiple factors such as historical charging load, charging stay duration, charging duration, and the corresponding weather conditions, temperature levels, and holidays are comprehensively incorporated. In view of the significant impact of different date types and electricity price policies on the electricity consumption behavior of residential charging piles, the grey correlation coefficient can be further used to accurately calculate the weight influence coefficient of the electricity price policy on the usage behavior of residential charging piles. Finally, this coefficient is integrated with the SARIMA prediction model to implement combined prediction in order to obtain more accurate and reliable prediction results.
[0065] More specifically, refer to Figure 4 As shown, in the adjustable resource prediction method for residential charging piles in the embodiment of the present application, first, the prediction purpose and content are clarified, and then the historical characteristic data of the charging piles is collected and processed. Further, the spearman coefficient can be used to select the key meteorological and temperature factors affecting the load, and the grey correlation degree coefficient can be used to select external factors to complete data preprocessing and selection of load influencing factors. On this basis, an adjustable resource prediction model for orderly charging of residential charging piles is constructed.
[0066] Then, input historical charging loads, charging residence durations, charging durations, and corresponding factors such as weather, temperature, and holidays into the adjustable resource prediction model. Further, it is possible to first determine whether there are abnormal conditions in the input data. If so, similar-day data is used to correct the abnormal data. Otherwise, the SARIMA model is used to predict the adjustable resources of residential charging piles.
[0067] Among them, when using the SARIMA model to predict the adjustable resources of residential charging piles, it involves model parameter estimation. On this basis, the adjustable resource load and adjustable duration of residential charging piles are mined.
[0068] Finally, it is possible to determine whether the prediction result meets the error requirements. If not, model calibration and improvement are carried out. If so, the prediction result of the adjustable resources of residential charging piles is output, that is, the potential of the adjustable resources for orderly charging of residential charging piles.
[0069] The method for predicting the adjustable resources of residential charging piles provided by an embodiment of the present application can accurately predict the adjustable potential of residential charging piles. By comprehensively considering the adjustable resources of residential charging piles and the load demand of the actual operation of the power grid, this method aims to effectively guide residential charging piles to participate in the orderly power consumption plan, thereby improving the utilization efficiency of the adjustable resources of charging piles. In addition, when there is a power supply gap, it can reduce the stable operation pressure of the power grid, provide a solid support for the efficient and reasonable utilization of energy, and effectively reduce the load gap during high-power demand periods.
[0070] Embodiment 3 Figure 5 is the structural framework diagram of the device for predicting the adjustable resources of residential charging piles provided by an embodiment of the present application. Refer to Figure 5 As shown, the device may include: The acquisition module 100 is used to acquire residential charging pile data and preprocess the residential charging pile data. The residential charging pile data includes at least one of the moment when the maximum charging load appears daily, the value of the maximum charging load, the distribution of charging start times, the distribution of charging end times, the distribution of charging durations, the distribution of charging residence durations, the highest and lowest temperatures on the same day.
[0071] The acquisition module 200 is used to acquire adjustable resource influencing factors. The adjustable resource influencing factors include at least one of temperature, weather, date type, and electricity price.
[0072] The prediction module 300 is used to predict the adjustable resources of residential charging piles according to the residential charging pile data and the adjustable resource influencing factors. The adjustable resources of residential charging piles include adjustable load and adjustable duration.
[0073] It should be noted here that the examples and application scenarios implemented by the above modules and the corresponding steps are the same, but are not limited to the content disclosed in the methods of the above embodiments. It should be noted that the above modules, as part of the device, can run in an environment such as Figure 1 shown, and can be implemented by software or by hardware.
[0074] Embodiment 4 An embodiment of the present application further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the adjustable resource prediction method for residential charging piles in the above Embodiment 1 is implemented. Specific examples in this embodiment can refer to the examples described in the above Embodiment 1, and will not be repeated here.
[0075] Embodiment 5 An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. The computer program, when executed by a processor, implements the adjustable resource prediction method for residential charging piles in the above Embodiment 1. Similarly, specific examples in this embodiment can refer to the examples described in the above Embodiment 1, and will not be repeated here.
[0076] In addition, those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.
[0077] The above embodiments only represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several deformations and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.
Claims
1. A method for predicting adjustable resources of residential charging piles, characterized in that: include: Collecting residential charging pile data and preprocessing the residential charging pile data, wherein the residential charging pile data includes at least one of the time when the maximum charging load occurs every day, the value of the maximum charging load, the distribution of the charging start time, the distribution of the charging end time, the distribution of the charging duration, the distribution of the charging stay duration, and the maximum and minimum temperatures of the day; Acquire an adjustable resource influencing factor, wherein the adjustable resource influencing factor includes at least one of temperature, weather, date type, and electricity price; The adjustable resources of the resident charging piles are predicted according to the resident charging pile data and the adjustable resource influencing factors. The adjustable resources of the resident charging piles include adjustable load and adjustable duration.
2. The method according to claim 1, characterized in that The factors affecting the acquisition of adjustable resources include: The factors affecting adjustable resources are obtained by calculating the Spearman correlation coefficient. The weights of the factors affecting the adjustable resources are obtained by calculating the grey correlation coefficient; The adjustable resource influencing factor is determined according to the factor influencing the adjustable resource and the weight of the factor influencing the adjustable resource.
3. The method according to claim 2, characterized in that The predicting the adjustable resources of the resident charging piles according to the resident charging pile data and the adjustable resource influencing factors includes: Normalizing the residential charging pile data to obtain normalized data; The adjustable resources of the residents' charging piles are predicted according to the normalized data and the adjustable resource influencing factors.
4. The method according to claim 3, characterized in that The step of predicting the adjustable resources of the resident charging piles according to the normalized data and the adjustable resource influencing factors includes: Determining whether the normalized data is abnormal; If not, the normalized data and the adjustable resource influencing factors are input into a preset prediction model to obtain the predicted adjustable resources of the resident charging piles.
5. The method according to claim 4, characterized in that The step of predicting the adjustable resources of the resident charging piles according to the normalized data and the adjustable resource influencing factors further includes: If so, similar day data is used to correct the normalized data, and the corrected normalized data and the adjustable resource influencing factors are input into a preset prediction model to obtain the predicted adjustable resources of the residents' charging piles.
6. The method according to claim 5, characterized in that The prediction model is constructed based on the SARIMA model.
7. The method according to any one of claims 1 to 6, characterized in that: The preprocessing of the resident charging pile data includes: The average interpolation method is used to repair and fill in the abnormal data, and the abnormal data includes null values or 0 values.
8. A device for predicting adjustable resources of residential charging piles, characterized in that: include: A collection module is used to collect and pre-process the data of the residential charging piles, wherein the data of the residential charging piles includes at least one of the time when the maximum charging load occurs every day, the value of the maximum charging load, the distribution of the charging start time, the distribution of the charging end time, the distribution of the charging duration, the distribution of the charging stay duration, and the maximum and minimum temperatures of the day; An acquisition module, used for acquiring adjustable resource influencing factors, wherein the adjustable resource influencing factors include at least one of temperature, weather, date type and electricity price; The prediction module is used to predict the adjustable resources of the resident charging piles according to the resident charging pile data and the adjustable resource influencing factors, and the adjustable resources of the resident charging piles include adjustable load and adjustable duration.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.