Power grid load prediction method based on vehicle network interaction technology

Abstract

The invention relates to the technical field of power load prediction, in particular to a power grid load prediction method based on a vehicle network interaction technology, and the method comprises the steps: periodically obtaining the behavior data of power grid service objects, and dividing the types of the power grid service objects based on the behavior consistency degree of the power grid service objects in a plurality of periods; determining division key parameters based on the power grid service object type; determining a power grid load type based on the division key parameters, and determining a power grid load prediction method based on the power grid load type; obtaining a load peak time period under a corresponding prediction method within a preset time length, and determining whether the power grid load prediction is qualified or not based on the response characteristics of the power grid load peak time period within the preset time length; determining whether to adjust the judgment parameters of the prediction method based on the eligibility of the power grid load prediction; according to the method, the accuracy of power grid service object analysis is improved, so that the accuracy of power grid load prediction method selection is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power load forecasting, and in particular to a power grid load forecasting method based on vehicle-grid interaction technology. Background Art

[0002] Accurate grid load forecasting is crucial for the safe and stable operation, optimal dispatching and resource allocation of the power grid. Traditional grid load forecasting methods are mainly based on analysis of historical load data and meteorological factors. However, with the access of new loads such as electric vehicles, the uncertainty and volatility of grid loads increase, and the prediction accuracy of traditional methods is challenged. Therefore, it is necessary to develop a new grid load forecasting method that can fully consider the characteristics of new loads such as electric vehicles and improve the accuracy and reliability of forecasting.

[0003] Chinese patent application publication number: CN112101663A discloses a study on a power grid load verification and prediction strategy based on big data and partially observable Markov decision process. It is based on the load history data collected by the power grid data center, and takes into account the classification of load types: such as valley load and transferable load. It takes the partially observable Markov decision process as the objective function, combines the orthogonal matching tracking algorithm, establishes a mathematical model and performs current load analysis, verification and short-term load prediction, and generates a day-ahead load analysis report. This study enables power grid load verification and short-term power grid load prediction to be realized. The dispatching center can verify the collected load data and effectively dispatch some controllable loads on the day-ahead according to the results, thereby achieving optimal economic operation of users and power grids.

[0004] It can be seen that the existing technology lacks an analysis process for power grid service objects, resulting in an inability to effectively and accurately divide power grid load types according to the types of power grid service objects, causing inaccurate power grid load prediction. Summary of the invention

[0005] To this end, the present invention provides a grid load prediction method based on vehicle-grid interaction technology to overcome the problem in the prior art of lacking an analysis process for grid service objects, resulting in an inability to effectively and accurately classify grid load types according to grid service object types, causing inaccurate grid load prediction.

[0006] To achieve the above object, the present invention provides a power grid load prediction method based on vehicle-grid interaction technology, comprising: Periodically acquiring behavior data of power grid service objects, and classifying power grid service object types based on the degree of consistency of the behaviors of the power grid service objects in a number of periods, wherein the power grid service object types include fixed types and mobile types; Determine the key parameters for division based on the type of the grid service object, wherein the key parameters for division include the degree of fluctuation of the charging completion time of the grid service object and the charging and discharging switching frequency of the grid service object; Determine the type of power grid load based on the key parameters of the division, and determine the prediction method of the power grid load based on the power grid load type, the power grid load type includes a fluctuating load type and a stable load type, and the prediction method includes a time series prediction method and a similar day prediction method; Obtaining a load peak time period under a corresponding prediction method within a preset time period, and determining whether the power grid load prediction is qualified based on the response characteristics of the power grid load peak time period within the preset time period, wherein the response characteristics include rapid response and delayed response; Based on the eligibility of the power grid load forecast, it is determined whether to adjust the judgment parameters of the forecast method, and the judgment parameters of the forecast method include a preset consistency degree and a preset duration.

[0007] Furthermore, the classification of grid service object types based on the degree of consistency of behaviors of grid service objects within a number of cycles includes: The degree of consistency of the behavior of the power grid service object within the plurality of cycles is greater than a preset degree of consistency, and the type of the power grid service object is determined to be a fixed type; If the degree of consistency of the behavior of the power grid service object within the plurality of cycles is less than or equal to a preset degree of consistency, it is determined that the type of the power grid service object is a flow type.

[0008] Furthermore, the degree of consistency of the behaviors of the power grid service objects within the several cycles is determined according to the number of power grid service objects whose number of repeated uses of the regional power grid during charging is greater than a preset number of repeated uses.

[0009] Further, the determining of the key parameters for division based on the type of the power grid service object includes: If the type of the grid service object is a fixed type, the key parameter for classification is determined to be the degree of fluctuation of the charging completion time of the grid service object; If the grid service object type is a flow type, the key parameter for classification is determined to be the charging and discharging switching frequency of the grid service object.

[0010] Further, determining the type of power grid load based on the division key parameters includes: If the fluctuation degree of the charging completion time of several same type of grid service objects within a preset period is greater than the preset fluctuation degree or the charging and discharging switching frequency of several same type of grid service objects within a preset period is greater than the preset switching frequency, it is determined that the grid load type is a fluctuating load; If the fluctuation degree of the charging completion time of several same type of grid service objects within a preset period is less than or equal to the preset fluctuation degree or the charging and discharging switching frequency of several same type of grid service objects within a preset period is less than or equal to the preset switching frequency, the grid load type is determined to be a stable load.

[0011] Furthermore, the prediction method for determining the grid load based on the grid load type includes: If the grid load type is fluctuating load, determine to use time series forecasting method to forecast the grid load; If the grid load type is stable load, determine to use similar day forecasting method to forecast the grid load.

[0012] Further, the determining whether the grid load forecast is qualified based on the response characteristics of the grid load peak time period within the preset time period includes: If the response characteristic of the peak load period of the power grid within the preset time period is rapid response, it is determined that the power grid load forecast is qualified; If the response characteristic of the power grid load peak period within the preset time period is a delayed response, it is determined that the power grid load forecast is unqualified.

[0013] Furthermore, determining the response characteristics of the peak load period of the power grid within the preset time period includes: If the response time offset of the peak load period of the power grid within the preset time length is less than the preset offset and the dispatching measure is effective, the response characteristic of the peak load period of the power grid within the preset time length is determined to be a rapid response; If the response time offset of the peak load period of the power grid within the preset time length is greater than or equal to the preset offset or the dispatching measure is invalid, determining that the response characteristic of the peak load period of the power grid within the preset time length is a delayed response; The dispatching measures are effective in controlling the fluctuation range of voltage load within ±5% of the rated voltage during the peak load period of the power grid.

[0014] Further, the determining whether to adjust the judgment parameters of the prediction method based on the eligibility of the power grid load prediction includes: If the grid load forecast is unqualified, it is determined that the judgment parameters of the forecast method need to be adjusted.

[0015] Furthermore, the adjustment amount of the preset consistency degree is positively correlated with the response time offset of the peak power grid load period within the preset time length, and the adjustment amount of the preset time length is negatively correlated with the response time offset of the peak power grid load period within the preset time length.

[0016] Compared with the prior art, the beneficial effect of the present invention lies in that the present invention determines the degree of behavioral consistency by calculating the ratio of the number of grid service objects whose number of repeated uses of the regional power grid during charging is greater than the preset number of repeated uses to the total number of grid service objects in several cycles. This method is based on actual usage data and has high accuracy and objectivity. Compared with subjective judgment or evaluation of a single indicator, this method can more comprehensively reflect the behavior patterns of grid service objects in multiple cycles. Accurate classification of grid service object types is helpful to formulate more targeted grid load forecasting methods and management strategies. For fixed types of service objects, relatively stable forecasting models and scheduling schemes can be adopted to improve the accuracy and reliability of forecasts and reduce grid operation costs. For mobile types of service objects, more flexible forecasting methods and emergency measures can be adopted to cope with their uncertainties. For example, for some temporary power consumption sites or mobile charging facilities, real-time environmental forecasting methods can be used to combine real-time information such as weather and traffic to predict loads, and timely adjust the operating status of the grid to ensure safe and stable operation of the grid.

[0017] Furthermore, the present invention determines different key parameters for division according to the types of power grid service objects. For fixed types of service objects, the fluctuation degree of charging completion time is focused on, and for mobile types of service objects, the charging and discharging switching frequency is focused on. It is highly targeted. Different types of service objects have different influencing factors on the power grid load. This distinction can more accurately evaluate the characteristics of the power grid load. Accurately determining the type of power grid load is helpful to formulate a more reasonable power grid load forecasting method. For fluctuating types of power grid loads, more flexible forecasting models and emergency measures can be adopted to cope with their uncertainties; for stable load types of power grid loads, relatively stable forecasting methods and scheduling schemes can be adopted to improve the accuracy and reliability of the forecast. The above method improves the accuracy of the analysis process of power grid service objects and thus improves the accuracy of power grid load forecasting.

[0018] Furthermore, the present invention adopts a time series prediction method for fluctuating power grid loads, which can better capture the dynamic change trend of the load. The time series prediction method is good at processing data with time series characteristics, and can analyze the change pattern of the load over time and the effects of various influencing factors. When the power grid load fluctuates greatly, this method can find out the pattern and trend of the load change through in-depth analysis of historical data, thereby improving the accuracy of the prediction. For the power grid load of stable load type, the similar day prediction method is adopted, which can make full use of the law of relatively stable load periods for prediction. The similar day prediction method is based on the principle that the load has similar performance under similar conditions. When the power grid load is relatively stable, by finding historical dates that are similar to the current conditions, the load conditions of these dates are used to predict future loads. The stable characteristics in the historical data can be effectively utilized to improve the accuracy of the prediction. The above method improves the accuracy of the analysis process of the power grid service object and thus improves the accuracy of the power grid load prediction.

[0019] Furthermore, the present invention uses the response characteristics of the peak power grid load period within a preset time period to judge whether the power grid load forecast is qualified, and establishes a connection between the forecast result and the actual power grid operation response. If the response is quick, it means that the forecast can provide accurate reference for power grid dispatch in advance, so that the dispatch measures can be effectively implemented, and the power grid can be stably passed through the peak load period, which means that the forecast method is reliable and practical; if the response is delayed, it indicates that the forecast is insufficient and needs to be improved, so that the evaluation of the forecast method has a clear basis that fits the actual operating conditions. The above method improves the accuracy of the analysis process of the power grid service object and thus improves the accuracy of the power grid load forecast.

[0020] Furthermore, the present invention determines whether to adjust the judgment parameters of the prediction method according to the eligibility of the power grid load forecast, so that the entire prediction system has the ability of dynamic optimization. When the prediction is unqualified, it means that the currently adopted prediction method may not be consistent with the actual power grid operation in some aspects. At this time, timely adjustment of key judgment parameters such as the preset consistency degree and the preset duration can enable the prediction method to better adapt to the changing characteristics of the power grid load, continuously improve the accuracy and reliability of the prediction, so that it can continue to maintain good performance with the development of the power grid and changes in the external environment. The adjustment amount of the preset consistency degree is set to be positively correlated with the response time offset, and the adjustment amount of the preset duration is set to be negatively correlated with the response time offset. This association is highly targeted. For example, if the response time offset is large, then appropriately increasing the preset consistency degree is helpful to classify the power grid service object type from a stricter perspective and obtain more accurate feature information; at the same time, reducing the preset duration can allow subsequent predictions to focus more on shorter and more timely time periods, thereby reducing problems such as prediction lags caused by improper time scale selection, and achieving accurate optimization of the prediction method. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a flowchart of a method for predicting power grid load based on vehicle-grid interaction technology according to an embodiment of the present invention; Figure 2 A flowchart of dividing the types of power grid service objects in a power grid load forecasting method based on vehicle-grid interaction technology according to an embodiment of the present invention; Figure 3 The present invention is a flowchart of a process for determining whether a power grid load forecast is qualified in a power grid load forecasting method based on vehicle-grid interaction technology according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0023] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0024] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0025] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0026] See also Figure 1 to Figure 3 As shown, Figure 1 This is a flowchart of a method for predicting power grid load based on vehicle-grid interaction technology according to an embodiment of the present invention; Figure 2 A flowchart of dividing the types of power grid service objects in a power grid load forecasting method based on vehicle-grid interaction technology according to an embodiment of the present invention; Figure 3 The present invention is a flowchart of a process for determining whether a power grid load forecast is qualified in a power grid load forecasting method based on vehicle-grid interaction technology according to an embodiment of the present invention.

[0027] The power grid load prediction method based on vehicle-grid interaction technology in the embodiment of the present invention includes: Step S1, periodically acquiring behavior data of power grid service objects, and classifying power grid service object types based on the degree of consistency of the behaviors of the power grid service objects in several periods, wherein the power grid service object types include fixed types and mobile types; Step S2, determining the key parameters for division based on the type of the grid service object, wherein the key parameters for division include the degree of fluctuation of the charging completion time of the grid service object and the charging and discharging switching frequency of the grid service object; Step S3, determining the grid load type based on the key parameters of the division, and determining the grid load prediction method based on the grid load type, wherein the grid load type includes a fluctuating load type and a stable load type, and the prediction method includes a time series prediction method and a similar day prediction method; Step S4, obtaining a load peak time period under a corresponding prediction method within a preset time period, and determining whether the power grid load prediction is qualified based on the response characteristics of the power grid load peak time period within the preset time period, wherein the response characteristics include rapid response and delayed response; Step S5, based on the eligibility of the grid load forecast, determining whether to adjust the judgment parameters of the forecast method, the judgment parameters of the forecast method including a preset consistency degree and a preset duration.

[0028] In the embodiment of the present invention, the grid service object is an electric vehicle, and the behavior data includes but is not limited to "charging time data, charging location data and charging frequency data". The peak time period of the grid load under the corresponding prediction method within the preset time period is the time period when the load voltage is greater than the preset load voltage under the corresponding prediction method. The preset load voltage is the average value of the grid load voltage within the preset time period. The preset period is set to 30 days, and the preset time period is set to 10 days, but the above values ​​are not limited to this, and technical personnel in this field can also adjust the values ​​according to actual needs.

[0029] Specifically, in step S1, when dividing the types of power grid service objects, the types of power grid service objects are divided according to the comparison results of the behavior consistency of the power grid service objects in several cycles and the preset consistency degree; When the degree of consistency of the behavior of the power grid service object in a number of cycles is greater than a preset degree of consistency, determining the type of the power grid service object as a fixed type; When the degree of consistency of the behavior of the power grid service object within a number of cycles is less than or equal to a preset degree of consistency, it is determined that the type of the power grid service object is a flow type.

[0030] The degree of consistency of the behavior of the power grid service objects within several cycles in the embodiment of the present invention is the ratio of the number of power grid service objects whose number of repeated uses of the regional power grid during charging is greater than the preset number of repeated uses to the total number of power grid service objects within several cycles, the preset number of repeated uses is the average number of repeated uses of the power grid service objects within several cycles, and the preset degree of consistency is the historical average value of the degree of consistency of the behavior of the power grid service objects, but the above values ​​are not limited to this, and technical personnel in this field can also adjust the value according to actual needs.

[0031] The present invention determines the degree of behavioral consistency by calculating the ratio of the number of grid service objects that reuse the regional power grid more than a preset number of times during charging to the total number of grid service objects in several cycles. This method is based on actual usage data and has high accuracy and objectivity. Compared with subjective judgment or evaluation of a single indicator, this method can more comprehensively reflect the behavior patterns of grid service objects in multiple cycles. Accurate classification of grid service object types is helpful to formulate more targeted grid load forecasting methods and management strategies. For fixed types of service objects, relatively stable forecasting models and scheduling schemes can be used to improve the accuracy and reliability of forecasts and reduce grid operation costs. For mobile types of service objects, more flexible forecasting methods and emergency measures can be adopted to cope with their uncertainties. For example, for some temporary power usage sites or mobile charging facilities, real-time environmental forecasting methods can be used to combine real-time information such as weather and traffic to predict loads, and timely adjust the operating status of the grid to ensure safe and stable operation of the grid.

[0032] Specifically, in step S2, when determining the key parameters for division, the key parameters for division are determined according to the type of the power grid service object; When the type of the grid service object is a fixed type, the key parameter for classification is determined to be the degree of fluctuation of the charging completion time of the grid service object; When the grid service object type is a flow type, the key parameter for division is determined to be the charging and discharging switching frequency of the grid service object.

[0033] Specifically, in step S2, when determining the type of power grid load, the type of power grid load is determined according to the division key parameters; When the fluctuation degree of the charging completion time of several same type of grid service objects within a preset period is greater than the preset fluctuation degree or the charging and discharging switching frequency of several same type of grid service objects within a preset period is greater than the preset switching frequency, the grid load type is determined to be a fluctuating load; When the fluctuation degree of the charging completion time of several same type of grid service objects within a preset period is less than or equal to the preset fluctuation degree or the charging and discharging switching frequency of several same type of grid service objects within a preset period is less than or equal to the preset switching frequency, the grid load type is determined to be a stable load.

[0034] In the embodiment of the present invention, the degree of fluctuation of the charging completion time of several identical grid service objects within the preset period is the average variance of the charging completion time of several identical grid service objects within the preset period, the charge and discharge switching frequency of several identical grid service objects within the preset period is the ratio of the number of charge and discharge switching times of several identical grid service objects within the preset period to the preset period duration, the identical grid service objects are electric vehicles of the same model, the preset degree of fluctuation is the average variance of the charging completion time of all identical grid service objects, and the preset switching frequency is the average value of the charge and discharge switching frequencies of all identical grid service objects, but the above values ​​are not limited thereto, and those skilled in the art may also adjust the values ​​according to actual needs.

[0035] The present invention determines different key parameters for division according to the types of power grid service objects. For fixed types of service objects, the fluctuation degree of charging completion time is focused on; for mobile types of service objects, the charging and discharging switching frequency is focused on. The present invention is highly targeted. Different types of service objects have different influencing factors on the power grid load. This distinction can more accurately evaluate the characteristics of the power grid load. Accurately determining the type of power grid load is helpful to formulate a more reasonable power grid load forecasting method. For fluctuating types of power grid loads, more flexible forecasting models and emergency measures can be adopted to cope with their uncertainties. For stable load types of power grid loads, relatively stable forecasting methods and scheduling schemes can be adopted to improve the accuracy and reliability of the forecast. The above method improves the accuracy of the analysis process of the power grid service objects and thus improves the accuracy of the power grid load forecast.

[0036] Specifically, in step S3, when determining the prediction method of the power grid load, the prediction method of the power grid load is determined according to the power grid load type; When the grid load type is fluctuating load, it is determined to use the time series prediction method to predict the grid load; When the grid load type is stable load, it is determined to predict the grid load using the similar day prediction method.

[0037] The time series prediction method for predicting the power grid load described in the embodiment of the present invention includes collecting historical time series data of the power grid load, cleaning the collected data, removing outliers and erroneous data, and dividing the preprocessed time series data into a training set, a validation set, and a test set. Generally speaking, 70% to 80% of the data is used for training, 10% to 15% of the data is used for validation, and the remaining data is used for testing. For example, for power grid load data with 1,000 time points, 700 to 800 points can be selected as a training set, 100 to 150 points can be selected as a validation set, and 100 to 150 points can be selected as a test set. The trained model is used to predict the power grid load for a future time (such as the next hour, day, etc.) based on the known historical time series data. For example, in a prediction model based on LSTM, the power grid load data of the past several hours is input into the model, and the model outputs the predicted value of the power grid load for the next hour.

[0038] In the embodiment of the present invention, the grid load prediction method using the similar day prediction method includes selecting standards to construct similar days, and the standards include date type (working day or rest day), weather conditions and special events (holidays or large-scale activities). For the selected similar days, the complete grid load data is collected, including load data at hourly, half-hourly or even finer time intervals. These data can be obtained from the monitoring system of the power grid and the power data center. The collected historical load data is cleaned to remove abnormal values ​​and erroneous data in the data. For each selected similar day, the pre-processed grid load data is used as the preliminary grid load prediction value of the prediction day. On the prediction day, as time goes by, real-time grid load data and related factor data (such as weather changes, electric vehicle charging conditions, etc.) are obtained, and the prediction results are corrected using these real-time data to make the prediction more accurate. For example, if the actual grid load on the morning of the prediction day is higher than the preliminary prediction value, and the weather conditions are hotter than expected, then the predicted load value in the afternoon can be appropriately adjusted upward.

[0039] The present invention adopts the time series prediction method for the fluctuating type of power grid load, which can better capture the dynamic change trend of the load. The time series prediction method is good at processing data with time series characteristics, and can analyze the change law of load over time and the role of various influencing factors. When the power grid load fluctuates greatly, this method can find out the pattern and trend of load change through in-depth analysis of historical data, thereby improving the accuracy of the prediction. For the power grid load of the stable load type, the similar day prediction method is adopted, which can make full use of the law of the relatively stable period of load for prediction. The similar day prediction method is based on the principle that the load has similar performance under similar conditions. When the power grid load is relatively stable, by finding historical dates similar to the current conditions, the load conditions of these dates are used to predict future loads. The stable characteristics in the historical data can be effectively utilized to improve the accuracy of the prediction. The above method improves the accuracy of the analysis process of the power grid service object and thus improves the accuracy of the power grid load prediction.

[0040] Specifically, in step S4, when determining whether the grid load forecast is qualified, it is determined whether the grid load forecast is qualified according to the response characteristics of the grid load peak time period within the preset time period; When the response characteristic of the peak load period of the power grid within the preset time period is rapid response, it is determined that the power grid load forecast is qualified; When the response characteristic of the power grid load peak time period within the preset time period is a delayed response, it is determined that the power grid load forecast is unqualified.

[0041] Specifically, in step S4, when determining the response characteristics of the peak load period of the power grid within the preset duration, the response characteristics of the peak load period of the power grid within the preset duration are determined according to the response time offset of the peak load period of the power grid within the preset duration and whether the dispatching measures are effective; When the response time offset of the peak load period of the power grid within the preset time length is less than the preset offset and the dispatching measure is effective, determining that the response characteristic of the peak load period of the power grid within the preset time length is a rapid response; When the response time offset of the peak load period of the power grid within the preset time length is greater than or equal to the preset offset or the dispatching measure is invalid, determining that the response characteristic of the peak load period of the power grid within the preset time length is a response lag; Among them, the dispatching measure is effective when the fluctuation range of voltage load during the peak load period of the power grid is controlled within ±5% of the rated voltage, and the dispatching measure is invalid when the fluctuation range of voltage load during the peak load period of the power grid is outside ±5% of the rated voltage.

[0042] In the embodiment of the present invention, the response time offset of the power grid load peak time period within the preset time length is the average of the differences between the start timestamps of several power grid load forecast peak time periods within the preset time length and the actual start timestamps of the power grid load peak time period. The preset offset is the average response time offset of the power grid load peak time periods within several preset time lengths, but the above values ​​are not limited to this, and technical personnel in this field can also adjust the value according to actual needs.

[0043] The present invention judges whether the power grid load forecast is qualified based on the response characteristics of the power grid load peak time period within a preset time length, and establishes a connection between the forecast result and the actual power grid operation response. If the response is quick, it means that the forecast can provide accurate reference for power grid dispatch in advance, so that the dispatch measures can be effectively implemented, and the power grid can be guaranteed to pass the peak load period stably, which means that the forecast method is reliable and practical; if the response is delayed, it means that the forecast is insufficient and needs to be improved, so that the evaluation of the forecast method has a clear basis that fits the actual operation situation. The above method improves the accuracy of the analysis process of the power grid service object and thus improves the accuracy of the power grid load forecast.

[0044] Specifically, in step S5, when determining whether to adjust the judgment parameters of the prediction method, it is determined whether to adjust the judgment parameters of the prediction method according to the eligibility of the power grid load prediction; When the grid load forecast is qualified, it is determined that there is no need to adjust the judgment parameters of the forecast method; When the grid load forecast is unqualified, it is determined that the judgment parameters of the forecast method need to be adjusted.

[0045] Specifically, when it is determined that the judgment parameters of the prediction method need to be adjusted, it is determined that the preset consistency degree is adjusted with a first adjustment coefficient, and the preset duration is adjusted with a second adjustment coefficient.

[0046] In the embodiment of the present invention, the value range of the first adjustment coefficient is set to 1.05~1.21, the value of the first adjustment coefficient is preferably 1.12, the value range of the second adjustment coefficient is set to 0.82~0.96, the value range of the second adjustment coefficient is set to 0.88, the adjustment amount of the preset consistency degree is positively correlated with the response time offset of the peak time period of the power grid load within the preset time length, and the adjustment amount of the preset time length is negatively correlated with the response time offset of the peak time period of the power grid load within the preset time length, but the above values ​​are not limited to this, and technical personnel in this field can also adjust the value according to actual needs.

[0047] The present invention determines whether to adjust the judgment parameters of the prediction method according to the eligibility of the power grid load prediction, so that the entire prediction system has the ability of dynamic optimization. When the prediction is unqualified, it means that the currently adopted prediction method may not be consistent with the actual power grid operation in some aspects. At this time, timely adjustment of key judgment parameters such as the preset consistency degree and the preset duration can enable the prediction method to better adapt to the changing characteristics of the power grid load, continuously improve the accuracy and reliability of the prediction, so that it can continue to maintain good performance with the development of the power grid and the change of the external environment. The adjustment amount of the preset consistency degree is set to be positively correlated with the response time offset, and the adjustment amount of the preset duration is set to be negatively correlated with the response time offset. This association is highly targeted. For example, if the response time offset is large, then appropriately increasing the preset consistency degree is helpful to divide the power grid service object type from a stricter perspective and obtain more accurate feature information; at the same time, reducing the preset duration can allow subsequent predictions to focus more on shorter and more timely time periods, thereby reducing problems such as prediction lag caused by improper time scale selection, and realizing accurate optimization of the prediction method.

[0048] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0049] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for predicting power grid load based on vehicle-grid interaction technology, characterized in that: include: Periodically acquiring behavior data of power grid service objects, and classifying power grid service object types based on the degree of consistency of the behaviors of the power grid service objects in a number of periods, wherein the power grid service object types include fixed types and mobile types; Determine the key parameters for division based on the type of the grid service object, wherein the key parameters for division include the degree of fluctuation of the charging completion time of the grid service object and the charging and discharging switching frequency of the grid service object; Determine the type of power grid load based on the key parameters of the division, and determine the prediction method of the power grid load based on the power grid load type, the power grid load type includes fluctuating load and stable load, and the prediction method includes a time series prediction method and a similar day prediction method; Obtaining a peak load time period under a corresponding prediction method within a preset time period, and determining whether the power grid load prediction is qualified based on the response characteristics of the peak load time period of the power grid within the preset time period, wherein the response characteristics include rapid response and delayed response; Based on the eligibility of the power grid load forecast, it is determined whether to adjust the judgment parameters of the forecast method, and the judgment parameters of the forecast method include a preset consistency degree and a preset duration.

2. The method for predicting power grid load based on vehicle-grid interaction technology according to claim 1 is characterized in that: The classification of grid service object types based on the degree of consistency of the behaviors of grid service objects in a number of cycles includes: The degree of consistency of the behavior of the power grid service object within the plurality of cycles is greater than a preset degree of consistency, and the type of the power grid service object is determined to be a fixed type; If the degree of consistency of the behavior of the power grid service object within the plurality of cycles is less than or equal to a preset degree of consistency, it is determined that the type of the power grid service object is a flow type.

3. The method for predicting power grid load based on vehicle-grid interaction technology according to claim 2 is characterized in that: The degree of consistency of the behaviors of the power grid service objects within the plurality of cycles is determined according to the number of power grid service objects whose number of repeated uses of the regional power grid during charging is greater than a preset number of repeated uses.

4. The method for predicting power grid load based on vehicle-grid interaction technology according to claim 3 is characterized in that: The determining of the key parameters for division based on the type of the power grid service object comprises: If the type of the grid service object is a fixed type, the key parameter for classification is determined to be the degree of fluctuation of the charging completion time of the grid service object; If the grid service object type is a flow type, the key parameter for classification is determined to be the charging and discharging switching frequency of the grid service object.

5. The method for predicting power grid load based on vehicle-grid interaction technology according to claim 4 is characterized in that: Determining the type of power grid load based on the key parameters of the division includes: If the fluctuation degree of the charging completion time of several same type of grid service objects within a preset period is greater than the preset fluctuation degree or the charging and discharging switching frequency of several same type of grid service objects within a preset period is greater than the preset switching frequency, it is determined that the grid load type is a fluctuating load; If the fluctuation degree of the charging completion time of several same type of grid service objects within a preset period is less than or equal to the preset fluctuation degree or the charging and discharging switching frequency of several same type of grid service objects within a preset period is less than or equal to the preset switching frequency, the grid load type is determined to be a stable load.

6. The method for predicting power grid load based on vehicle-grid interaction technology according to claim 5 is characterized in that: The prediction method for determining the power grid load based on the power grid load type includes: If the grid load type is fluctuating load, determine to use time series forecasting method to forecast the grid load; If the grid load type is stable load, determine to use similar day forecasting method to forecast the grid load.

7. The method for predicting power grid load based on vehicle-grid interaction technology according to claim 6, characterized in that: The determining whether the grid load forecast is qualified based on the response characteristics of the grid load peak time period within the preset time period includes: If the response characteristic of the peak load period of the power grid within the preset time period is rapid response, it is determined that the power grid load forecast is qualified; If the response characteristic of the power grid load peak period within the preset time period is a delayed response, it is determined that the power grid load forecast is unqualified.

8. The method for predicting power grid load based on vehicle-grid interaction technology according to claim 7 is characterized in that: The response characteristics of the peak load period of the power grid within the preset time period include: If the response time offset of the peak load period of the power grid within the preset time length is less than the preset offset and the dispatching measure is effective, the response characteristic of the peak load period of the power grid within the preset time length is determined to be a rapid response; If the response time offset of the peak load period of the power grid within the preset time length is greater than or equal to the preset offset or the dispatching measure is invalid, determining that the response characteristic of the peak load period of the power grid within the preset time length is a delayed response; The dispatching measures are effective in controlling the fluctuation range of voltage load within ±5% of the rated voltage during the peak load period of the power grid.

9. The method for predicting power grid load based on vehicle-grid interaction technology according to claim 8, characterized in that: The determining whether to adjust the judgment parameters of the prediction method based on the eligibility of the power grid load prediction includes: If the grid load forecast is unqualified, it is determined that the judgment parameters of the forecast method need to be adjusted.

10. The method for predicting power grid load based on vehicle-grid interaction technology according to claim 9, characterized in that: The adjustment amount of the preset consistency degree is positively correlated with the response time offset of the peak power grid load period within the preset time length, and the adjustment amount of the preset time length is negatively correlated with the response time offset of the peak power grid load period within the preset time length.

Images