Deviation electric quantity processing method, system and equipment for distributed transaction and medium

Abstract

The invention is suitable for the technical field of deviation electric quantity processing, and provides a deviation electric quantity processing method, system and device for distributed transaction and a medium, and the method comprises the steps: collecting the historical power generation data and meteorological data of a distributed power supply within a preset time, and calculating the predicted power generation amount of the distributed power supply through a support vector machine algorithm; acquiring historical electricity consumption data of the user in a preset time, and constructing a seasonal autoregression integral moving average model to calculate predicted electricity consumption of the user; respectively estimating the deviation generating capacity of the distributed power supply and the deviation electricity consumption of the user; judging whether the deviation generating capacity of the distributed power supply and / or the deviation electricity consumption of the user meet a threshold condition or not; if yes, introducing a demand response mechanism, and processing the deviation generating capacity and / or the deviation electricity consumption through the demand response mechanism; and if not, continuing to collect and update the data related to the power generation of the distributed power supply and the power utilization of the user, thereby solving the problem of loss caused by deviation of the electric quantity.

Description

Technical Field

[0001] The present invention relates to the technical field of deviation power processing, and in particular to a deviation power processing method, system, device and medium for distributed transactions. Background Art

[0002] Deviation electricity in distributed transactions refers to the difference between the actual power generation or power consumption and the agreed power in the transaction contract during the distributed energy transaction. There are many reasons for the generation of deviation electricity. From the perspective of power generation, distributed power sources are easily affected by natural environmental factors. For example, photovoltaic power generation will cause fluctuations in power generation due to light intensity and weather changes; wind power generation will also cause changes in power generation due to unstable wind speed. From the perspective of electricity consumption, changes in users' electricity consumption habits, additions or reductions in electrical equipment, etc. will cause inconsistencies between actual electricity consumption and contracted electricity. When the deviation electricity generated is large, it may cause the voltage of the local power grid to increase or decrease, affecting the normal operation of user equipment, damaging equipment, and even causing some equipment to fail to operate normally, thereby causing large-scale power outages.

[0003] In view of this, a method, system, device and medium for processing deviation electricity in distributed transactions are needed. Summary of the invention

[0004] The embodiments of the present application provide a method, system, device and medium for processing deviation electricity in distributed transactions, which are used to solve the problem of losses caused by deviation electricity.

[0005] A first aspect of an embodiment of the present application provides a method for processing a deviation amount of electricity in a distributed transaction, including:

[0006] Collect historical power generation data and meteorological data of distributed power sources within a preset time, and use support vector machine algorithm to calculate the predicted power generation of distributed power sources;

[0007] Collect the user's historical electricity consumption data within the preset time, and build a seasonal autoregressive integral moving average model to calculate the user's predicted electricity consumption;

[0008] estimating the deviation power generation of the distributed power source and the deviation power consumption of the user according to the predicted power generation of the distributed power source and the predicted power consumption of the user respectively;

[0009] Determine whether the deviation power generation of the distributed power source and / or the deviation power consumption of the user meets a threshold condition;

[0010] If yes, a demand response mechanism is introduced to handle the deviation in power generation and / or the deviation in power consumption;

[0011] If not, continue to collect and update data related to the distributed power generation and user electricity consumption.

[0012] Furthermore, the collecting of historical power generation data and meteorological data of the distributed power sources within a preset time and calculating the predicted power generation of the distributed power sources using a support vector machine algorithm include:

[0013] Clean and standardize the collected historical power generation data and meteorological data;

[0014] Determine feature parameters, kernel functions, and model target parameters based on the processed data;

[0015] The preprocessed data is divided into a training set and a test set, and an SVM model is trained based on the data of the training set, the kernel function and the model target parameters to solve the SVM target function;

[0016] The power generation characteristic data of the distributed power source acquired in real time is input into the trained SVM model to obtain the predicted power generation of the distributed power source.

[0017] Furthermore, the preprocessed data is divided into a training set and a test set, and the SVM model is trained based on the data of the training set, the kernel function and the model target parameters to solve the SVM target function, including:

[0018] When the data in the training set is updated, the weighted SVM model parameters are obtained by weighting the updated data;

[0019] An updated objective function is determined based on the weighted SVM model parameters.

[0020] Furthermore, the expression of the updated objective function is as follows:

[0021] Objective function:

[0022]

[0023] Where: w, b, ξ and ξ * are weights, bias terms, and slack variables for updating the data in the training set, C is the penalty coefficient, and w i is the initial weight of the i-th sample.

[0024] Furthermore, the collecting of historical electricity consumption data of users within a preset time and the construction of a seasonal autoregressive integral moving average model to calculate the predicted electricity consumption of users include:

[0025] Draw a time series diagram of electricity consumption based on the user's historical electricity consumption data, and use a unit root test to determine the stationarity of the time series;

[0026] When the time series data is not stationary, it is differentiated according to the seasonal trend type;

[0027] Determining a model target order according to the time series data after difference processing, and fitting a SARIMA model based on the model target order;

[0028] After verifying that the SARIMA model is effective, the fitted SARIMA model is used to calculate the user's predicted electricity consumption.

[0029] Furthermore, determining the model target order according to the time series data after difference processing, and fitting the SARIMA model based on the model target order, includes:

[0030] The autocorrelation function and partial autocorrelation function of the time series are calculated according to the time series data after difference processing;

[0031] Determining the order of the sliding average part by the autocorrelation function;

[0032] The order of the autoregressive part is determined by the partial autocorrelation function.

[0033] Furthermore, the demand response mechanism is introduced to handle the deviation of power generation and / or deviation of power consumption, including:

[0034] The demand response mechanism includes incentive-based demand response and price-based demand response. The incentive-based demand response mechanism handles deviation power generation and / or deviation power consumption by establishing a response plan, and the response plan uses smart devices to assist in the response; the price-based demand response mechanism handles deviation power generation and / or deviation power consumption by adjusting the price ratio.

[0035] A second aspect of an embodiment of the present application provides a distributed transaction deviation power processing system, including:

[0036] A distributed power generation prediction unit is used to collect historical power generation data and meteorological data of distributed power generation within a preset time, and calculate the predicted power generation of distributed power generation using a support vector machine algorithm;

[0037] A user power consumption prediction unit is used to collect the user's historical power consumption data within a preset time and construct a seasonal autoregressive integral sliding average model to calculate the user's predicted power consumption;

[0038] A deviation power generation and deviation power consumption estimation unit, used to estimate the deviation power generation of the distributed power source and the deviation power consumption of the user according to the predicted power generation of the distributed power source and the predicted power consumption of the user respectively;

[0039] A deviation power generation and / or deviation power consumption judgment unit, used to judge whether the deviation power generation of the distributed power source and / or the deviation power consumption of the user meets a threshold condition;

[0040] A deviation power generation and / or deviation power consumption processing unit, used to introduce a demand response mechanism when the deviation power generation of the distributed power source and / or the deviation power consumption of the user meets a threshold condition, and process the deviation power generation and / or deviation power consumption through the demand response mechanism;

[0041] The unit for collecting and updating relevant data is used to continue collecting and updating relevant data on the power generation of the distributed power source and the power consumption of the user when the deviation power generation of the distributed power source and / or the deviation power consumption of the user do not meet the threshold conditions.

[0042] A third aspect of the embodiments of the present application provides a computer device, including:

[0043] memories, transceivers, processors, and bus systems;

[0044] Wherein, the memory is used to store programs;

[0045] The processor is used to execute the program in the memory, including executing the deviation power processing method of distributed transactions as described in any one of the above;

[0046] The bus system is used to connect the memory and the processor so that the memory and the processor can communicate with each other.

[0047] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, comprising instructions, which, when executed on a computer, enables the computer to execute a method for processing deviation electricity in a distributed transaction as described in any one of the above.

[0048] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0049] This application predicts the power generation of distributed power sources and the power consumption of users, and then estimates the deviation power generation and deviation power consumption based on the predicted power generation and predicted power consumption, and finally determines whether the triggering conditions for introducing a demand response mechanism are met based on the estimated deviation power generation and deviation power consumption, and then processes the deviation power according to the demand response mechanism. This application improves the prediction accuracy of the power generation of distributed power sources by using a support vector machine algorithm, and improves the prediction accuracy of user power consumption by using a seasonal autoregressive integral sliding average model. Through the demand response mechanism, the power supply and demand of the power generation system can be balanced, the economic losses caused by the deviation power can be reduced, and the losses caused to user power consumption can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic diagram of an embodiment of a method for processing deviation power of distributed transactions in the present invention;

[0051] Figure 2 It is a structural block diagram of an embodiment of a distributed transaction deviation power processing system in the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0053] The method for processing the deviation power of distributed transactions in this embodiment is used to reduce the economic losses caused by the deviation power and the losses caused to the user's electricity consumption. The implementation method in this embodiment can be implemented in the system, can be implemented in the server, and can also be implemented in the terminal, without specific limitation.

[0054] Embodiment 1

[0055] See also Figure 1 An embodiment of a method for processing a deviation amount of electricity in a distributed transaction of the present invention comprises the following steps:

[0056] S11. Collect historical power generation data and meteorological data of distributed power sources within a preset time, and use support vector machine algorithm to calculate the predicted power generation of distributed power sources;

[0057] In this embodiment, collecting historical power generation data and meteorological data of distributed power sources within a preset time and using a support vector machine algorithm to calculate the predicted power generation of distributed power sources includes the following steps:

[0058] 1. Clean and standardize the collected historical power generation data and meteorological data;

[0059] The preset time here is the time interval of a certain time, which is determined according to the actual situation. The historical power generation data is obtained by widely collecting information such as output power, operating time, equipment status, etc. of different types of distributed power sources, such as solar photovoltaic power stations and wind power plants. Meteorological data includes light intensity, wind speed, temperature, humidity, etc., which are factors affecting renewable energy power generation. Clean the data, remove outliers and missing values, and standardize them to make the data of different variables comparable.

[0060] 2. Determine the feature parameters, kernel function and model target parameters based on the processed data;

[0061] Analyze the characteristics of the processed data, calculate the correlation between the characteristics of each data and the power generation, and select the characteristic data with high correlation as the input of the model based on the correlation coefficient analysis. Here, the Gaussian kernel function is selected to map the data to a high-dimensional space. Based on the selected kernel function, determine the relevant parameters and penalty parameters.

[0062] 3. Divide the preprocessed data into training set and test set, perform SVM model training based on the data of the training set, kernel function and model target parameters, and solve the objective function of SVM;

[0063] The preprocessed data is divided into a training set and a test set. Generally, 70%-80% of the data is used as the training set, and the remaining data is used as the test set. Using the training set data, according to the selected kernel function and parameters, the objective function of the SVM is solved through the optimization algorithm to find the optimal hyperplane.

[0064] The objective function expression is as follows:

[0065]

[0066] The constraint expressions are as follows:

[0067] y i ―w T x i ―b≤∈+ξ i ;

[0068] w T x i +b―y i ≤∈+ξ i* ;

[0069] ξ i ,ξ i * ≥0;

[0070] Where: w is the weight vector, b is the bias term, ξ i ,ξ i * is the slack variable, ∈ is the parameter of the insensitive loss function, y i is the actual power generation, x i is the input feature vector.

[0071] During the model training process, w, b, ξ are adjusted through the optimization algorithm. i ,ξ i * The objective function is minimized by using the same parameters. Finally, the mean square error and mean absolute error are used to evaluate the model.

[0072] 4. Input the distributed power generation characteristic data obtained in real time into the trained SVM model to obtain the predicted power generation of the distributed power generation.

[0073] When power generation needs to be predicted, the power generation characteristic data of the distributed power source, such as meteorological data or equipment parameters, is obtained in real time, the data is input into the trained model, and the predicted value of power generation is calculated according to the prediction function of the model.

[0074] Specifically, the preprocessed data is divided into a training set and a test set, and the SVM model is trained based on the data of the training set, the kernel function and the model target parameters to solve the SVM target function, which also includes the following steps:

[0075] When the data of the training set is updated, the weighted SVM model parameters are obtained by weighting the updated data; and the updated objective function is determined based on the weighted SVM model parameters.

[0076] Since the generation of distributed power sources of various types is affected by many factors, the collected data may change over time. Therefore, this embodiment introduces an incremental learning algorithm to update the SVM model instead of retraining the entire model. In this way, the parameters of the model can be adaptively adjusted according to the changes in the data, which can maintain good prediction performance and effectively improve the prediction accuracy of the model.

[0077] For the existing training data set {(x1,y1),(x2,y2),…,(x n ,y n )}, define the initial weight of each sample, such as w i =1(i=1,2,…,m), when a new sample (x j ,y j ), giving the new sample a larger initial weight w j =2. Add the weight term to the objective function of SVM, and the expression is updated as follows:

[0078]

[0079] Where w i is the initial weight of the i-th sample.

[0080] After the corresponding constraints are updated, the power generation forecast results are re-solved based on the weighted objective function and constraints.

[0081] S12. Collect the user's historical electricity consumption data within a preset time, and construct a seasonal autoregressive integral moving average model to calculate the user's predicted electricity consumption;

[0082] In this embodiment, collecting the historical electricity consumption data of the user within a preset time and constructing a seasonal autoregressive integral moving average model to calculate the user's predicted electricity consumption includes the following steps:

[0083] 1. Draw a time series diagram of electricity consumption based on the user's historical electricity consumption data, and use a unit root test to determine the stationarity of the time series;

[0084] After collecting the user's historical electricity consumption data, data cleaning is performed, and a time series diagram of electricity consumption is drawn based on the processed data to observe whether the data shows obvious trends (such as year-on-year growth or decline), seasonality (such as high electricity consumption in summer and low electricity consumption in winter) or periodicity. Then a unit root test (such as ADF test-augmented Dickey-Fuller test) is used to determine the stationarity of the time series.

[0085] 2. When the time series data is not stationary, differential processing is performed according to the seasonal trend type;

[0086] When the data is not stationary, a difference operation is needed to make it stationary. Seasonal trends are divided into non-seasonal trends and seasonal trends. For non-seasonal trends, first-order or multi-order differences may be required. The calculation method of the first-order difference is Δy t =y t ―y t―1 If the data also has a seasonal trend and the seasonal period is s, seasonal differences need to be performed, such as the first-order seasonal difference Δsy t =y t ―y t―s The order of difference (including non-seasonal difference order and seasonal difference order D) needs to be determined according to the characteristics of the data and the test results.

[0087] 3. Determine the model target order based on the time series data after difference processing, and fit the SARIMA model based on the model target order;

[0088] 4. After verifying that the SARIMA model is effective, use the fitted SARIMA model to calculate the user's predicted electricity consumption.

[0089] After determining the target order of the model, p, q, P, Q, and s, the maximum likelihood estimation method is used to estimate the model parameters. After verifying that the model is effective, the fitted SARIMA model is used to predict future user electricity consumption.

[0090] Specifically, the target order of the model is determined according to the time series data after difference processing, and the SARIMA model is fitted based on the target order of the model, including:

[0091] The autocorrelation function and partial autocorrelation function of the time series are calculated respectively according to the time series data after difference processing; the order of the sliding average part is determined by the autocorrelation function; and the order of the autoregressive part is determined by the partial autocorrelation function.

[0092] Calculate the autocorrelation function and partial autocorrelation function of the stationary time series after difference. The autocorrelation function can help determine the orders q and Q of the moving average (MA) part, and the partial autocorrelation function can help determine the orders p and P of the autoregressive (AR) part. Generally speaking, if the autocorrelation function suddenly truncates (close to 0) after a certain lag order, it may indicate the order of the MA model; if the partial autocorrelation function truncates after a certain lag order, it may indicate the order of the AR model. For the seasonal part, it is necessary to observe the characteristics of the autocorrelation function and partial autocorrelation function at multiples of the seasonal cycle. For example, if the seasonal cycle is (monthly), observe the values ​​of the autocorrelation function and the partial autocorrelation function at lags of 12, 24, 36, etc.

[0093] S13. Estimating the deviation power generation of the distributed power source and the deviation power consumption of the user according to the predicted power generation of the distributed power source and the predicted power consumption of the user;

[0094] In this embodiment, the deviation variable can be calculated by the following formula:

[0095] ΔE=E actual ―E predicted ;

[0096] Where: ΔE is the deviation charge, E actual is the actual power generation or power consumption, E predicted To predict power generation or power consumption.

[0097] S14. Determine whether the deviation power generation of the distributed power source and / or the deviation power consumption of the user meets the threshold condition;

[0098] In this embodiment, the threshold condition here is the trigger condition for the system to execute the demand response mechanism, and the threshold of the trigger condition can be set according to the actual situation. For example, when the calculated deviation power generation value is relatively large, it can reach more than 30%, then the mechanism needs to be triggered at this time. For example, if the deviation of the user's deviation power consumption exceeds 20%, the demand response mechanism is triggered to reduce power consumption. It should be noted that the trigger condition set here is that only by using the demand response mechanism can the optimal effect of handling deviation power be achieved, not that the mechanism must be used as long as a deviation occurs. For example, when the calculated deviation power generation is less than 30%, the system can adopt another processing method, and can internally regulate the power generation equipment to supplement or store energy.

[0099] And / or refers to that when one of the deviation power generation of the distributed power source and the deviation power generation of the distributed power source meets the threshold condition, the demand response mechanism is triggered, and when both meet the threshold condition, the demand response mechanism is triggered at the same time.

[0100] S15. When the deviation power generation of the distributed power source and / or the deviation power consumption of the user meets the threshold condition, a demand response mechanism is introduced to process the deviation power generation and / or deviation power consumption through the demand response mechanism;

[0101] Specifically, the demand response mechanism includes incentive-based demand response and price-based demand response:

[0102] The incentive-based demand response mechanism handles deviation in power generation and / or deviation in power consumption by establishing a response plan, wherein the response plan is to use smart devices to assist in the response;

[0103] The use of smart devices to assist is to monitor the user's electricity consumption in real time through smart meters and smart home systems, and push electricity usage suggestions to users. When power generation is insufficient, the smart system can remind users to turn off unnecessary electrical devices, and smart sockets can automatically cut off the power supply of non-critical electrical appliances according to preset modes. However, the control operation of smart devices must be authorized by the user in advance and must not affect the user's normal life and the operation of key equipment.

[0104] The price-based demand response mechanism handles deviations in power generation and / or power consumption by adjusting the price ratio. When power generation is insufficient, the electricity price is appropriately increased to encourage users to reduce electricity consumption. The electricity price can be increased by a certain percentage in a short period of time, and the increase can be greater for high-energy-consuming enterprises. At the same time, the tiered electricity price policy is strengthened, and users who consume more than a certain amount of electricity face a higher electricity price gradient. However, the increase in electricity prices must not exceed a certain upper limit, and the price increase plan and time range must be announced to users in advance. When there is excess power generation, lowering electricity prices stimulates users to increase electricity consumption. The electricity price can be lowered to a certain percentage of the normal level, and special electricity price packages can be introduced to attract users to concentrate on electricity consumption during periods of excess power generation, such as "low-price electricity packages at night". However, the reduction in electricity prices must not cause the economic losses of power suppliers to exceed a certain limit, and the validity period of special electricity price packages must be clearly specified.

[0105] S16. When the deviation power generation of the distributed power source and / or the deviation power consumption of the user do not meet the threshold condition, continue to collect and update the data related to the power generation of the distributed power source and the power consumption of the user.

[0106] When the deviation power generation of the distributed power source and / or the deviation power consumption of the user do not meet the threshold conditions, the data related to the user's power consumption and the distributed power generation will continue to be collected and updated. This includes more frequent acquisition of smart meter data, meteorological data (for the case where the distributed power generation is affected by the weather) and any other factors that may affect the power balance, such as changes in the operating status of the user's equipment. In addition, warning conditions can be set, and different warning levels can be set according to the size and trend of the forecast deviation. For example, the forecast deviation within a smaller range is defined as a low-level warning, and a slightly larger range is defined as a medium-level warning. Or provide information feedback to the user, through mobile phone applications, text messages or smart meter display screens, etc., to inform the user that the current forecast deviation is within the normal range, but still needs attention. At the same time, some general energy-saving suggestions can be provided to encourage users to develop good electricity usage habits in their daily lives. These suggestions can include setting the air-conditioning temperature reasonably, avoiding long-term standby appliances, etc.

[0107] The above embodiment uses the support vector machine algorithm to improve the prediction accuracy of distributed power generation, and uses the seasonal autoregressive integrated moving average model to improve the prediction accuracy of user power consumption, thereby improving the evaluation accuracy of deviation power. The introduction of the demand response mechanism under the threshold condition can balance the power supply and demand in terms of power grid operation; it can reduce the economic losses caused by power deviation in terms of power generation enterprises; and it can reduce user losses and improve power efficiency in terms of users.

[0108] Embodiment 2

[0109] See also Figure 2 , an embodiment of a distributed transaction deviation power processing system of the present invention comprises the following steps:

[0110] A distributed power generation prediction unit is used to collect historical power generation data and meteorological data of distributed power generation within a preset time, and calculate the predicted power generation of distributed power generation using a support vector machine algorithm;

[0111] A user power consumption prediction unit is used to collect the user's historical power consumption data within a preset time and construct a seasonal autoregressive integral sliding average model to calculate the user's predicted power consumption;

[0112] A deviation power generation and deviation power consumption estimation unit, used to estimate the deviation power generation of the distributed power source and the deviation power consumption of the user according to the predicted power generation of the distributed power source and the predicted power consumption of the user respectively;

[0113] A deviation power generation and / or deviation power consumption judgment unit, used to judge whether the deviation power generation of the distributed power source and / or the deviation power consumption of the user meets the threshold condition;

[0114] A deviation power generation and / or deviation power consumption processing unit, used to introduce a demand response mechanism when the deviation power generation of the distributed power source and / or the deviation power consumption of the user meets the threshold condition, and process the deviation power generation and / or deviation power consumption through the demand response mechanism;

[0115] The unit for collecting and updating relevant data is used to continue collecting and updating relevant data on the distributed power generation and the user's electricity consumption when the deviation power generation of the distributed power source and / or the deviation power consumption of the user do not meet the threshold conditions.

[0116] Embodiment 3

[0117] The present invention provides a computer device, comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the steps of the above method are implemented.

[0118] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0119] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored. In addition, each functional unit in each embodiment of the present invention can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units.

[0120] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0121] It can be understood that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for processing deviation electricity in distributed transactions, characterized in that: include: Collect historical power generation data and meteorological data of distributed power sources within a preset time, and use support vector machine algorithm to calculate the predicted power generation of distributed power sources; Collect the user's historical electricity consumption data within the preset time, and build a seasonal autoregressive integral moving average model to calculate the user's predicted electricity consumption; estimating the deviation power generation of the distributed power source and the deviation power consumption of the user according to the predicted power generation of the distributed power source and the predicted power consumption of the user respectively; Determine whether the deviation power generation of the distributed power source and / or the deviation power consumption of the user meets a threshold condition; If yes, a demand response mechanism is introduced to handle the deviation in power generation and / or the deviation in power consumption; If not, continue to collect and update data related to the distributed power generation and user electricity consumption.

2. The method for processing the deviation amount of distributed transactions according to claim 1 is characterized in that: The collecting of historical power generation data and meteorological data of the distributed power source within a preset time and calculating the predicted power generation of the distributed power source using a support vector machine algorithm include: Clean and standardize the collected historical power generation data and meteorological data; Determine feature parameters, kernel functions, and model target parameters based on the processed data; The preprocessed data is divided into a training set and a test set, and an SVM model is trained based on the data of the training set, the kernel function and the model target parameters to solve the SVM target function; The power generation characteristic data of the distributed power source acquired in real time is input into the trained SVM model to obtain the predicted power generation of the distributed power source.

3. The method for processing the deviation amount of distributed transactions according to claim 2 is characterized in that: The preprocessed data is divided into a training set and a test set, and the SVM model is trained based on the data of the training set, the kernel function and the model target parameters to solve the SVM target function, including: When the data in the training set is updated, the weighted SVM model parameters are obtained by weighting the updated data; An updated objective function is determined based on the weighted SVM model parameters.

4. The method for processing the deviation amount of distributed transactions according to claim 3 is characterized in that: The expression of the updated objective function is as follows: Objective function: Where: w, b, ξ and ξ * are weights, bias terms, and slack variables for updating the data in the training set, C is the penalty coefficient, and w i is the initial weight of the i-th sample.

5. The method for processing the deviation amount of distributed transactions according to claim 1 is characterized in that: The collecting of the user's historical electricity consumption data within a preset time and the construction of a seasonal autoregressive integral sliding average model to calculate the user's predicted electricity consumption include: Draw a time series diagram of electricity consumption based on the user's historical electricity consumption data, and use a unit root test to determine the stationarity of the time series; When the time series data is not stationary, it is differentiated according to the seasonal trend type; Determining a model target order according to the time series data after difference processing, and fitting a SARIMA model based on the model target order; After verifying that the SARIMA model is effective, the fitted SARIMA model is used to calculate the user's predicted electricity consumption.

6. The method for processing the deviation amount of distributed transactions according to claim 5 is characterized in that: Determining the model target order according to the time series data after difference processing, and fitting the SARIMA model based on the model target order, includes: The autocorrelation function and partial autocorrelation function of the time series are calculated according to the time series data after difference processing; Determining the order of the sliding average part by the autocorrelation function; The order of the autoregressive part is determined by the partial autocorrelation function.

7. The method for processing the deviation amount of distributed transactions according to claim 1 is characterized in that: The demand response mechanism is introduced to handle the deviation of power generation and / or deviation of power consumption, including: The demand response mechanism includes incentive-based demand response and price-based demand response. The incentive-based demand response mechanism handles deviation power generation and / or deviation power consumption by establishing a response plan, and the response plan uses smart devices to assist in the response; the price-based demand response mechanism handles deviation power generation and / or deviation power consumption by adjusting the price ratio.

8. A distributed transaction deviation power processing system, characterized in that: include: A distributed power generation prediction unit is used to collect historical power generation data and meteorological data of distributed power generation within a preset time, and calculate the predicted power generation of distributed power generation using a support vector machine algorithm; A user power consumption prediction unit is used to collect the user's historical power consumption data within a preset time and construct a seasonal autoregressive integral sliding average model to calculate the user's predicted power consumption; A deviation power generation and deviation power consumption estimation unit, used to estimate the deviation power generation of the distributed power source and the deviation power consumption of the user according to the predicted power generation of the distributed power source and the predicted power consumption of the user respectively; A deviation power generation and / or deviation power consumption judgment unit, used to judge whether the deviation power generation of the distributed power source and / or the deviation power consumption of the user meets a threshold condition; A deviation power generation and / or deviation power consumption processing unit, used to introduce a demand response mechanism when the deviation power generation of the distributed power source and / or the deviation power consumption of the user meets a threshold condition, and process the deviation power generation and / or deviation power consumption through the demand response mechanism; The unit for collecting and updating relevant data is used to continue collecting and updating relevant data on the power generation of the distributed power source and the power consumption of the user when the deviation power generation of the distributed power source and / or the deviation power consumption of the user do not meet the threshold conditions.

9. A computer device, characterized in that: include: memories, transceivers, processors, and bus systems; Wherein, the memory is used to store programs; The processor is used to execute the program in the memory, including executing the deviation power processing method for distributed transactions according to any one of claims 1 to 7; The bus system is used to connect the memory and the processor so that the memory and the processor can communicate with each other.

10. A computer-readable storage medium, characterized in that: It includes instructions which, when executed on a computer, enable the computer to execute the method for processing deviation electricity quantity in distributed transactions as described in any one of claims 1 to 7.

Images