Power grid intelligent scheduling method based on multi-energy complementation

Abstract

The invention relates to the technical field of data processing, in particular to a power grid intelligent scheduling method based on multi-energy complementation, and the method comprises the steps: collecting current data, obtaining an extreme point in surrounding data of each current data, obtaining the suspected noise degree of each current data based on the extreme point, and obtaining the suspected noise degree of each current data based on the suspected noise degree. The suspected noise data is corrected according to the positive correlation between the current data and the temperature data collected under the corresponding time sequence, the noise degree of each piece of current data is obtained, according to the noise degree of each piece of current data, an IDR algorithm is used for denoising the current data, and a final current data sequence is obtained; the power load of the key node of the power grid is estimated according to the final current data sequence, so that the power grid system can perform energy scheduling more accurately, the energy utilization rate is improved, and the filtering effect is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology. More specifically, the present invention relates to a power grid intelligent dispatching method based on multi-energy complementarity. Background Art

[0002] A multi-energy complementary power grid refers to a power grid system that utilizes a variety of renewable energy power generation technologies, such as solar energy, wind energy, biomass energy, etc., and integrates multiple renewable energy sources such as solar energy, wind energy, and biomass energy, and converts them appropriately through efficient technical means to achieve complementarity and optimization between different energy forms. Intelligent scheduling can achieve precise control and scheduling of various energy sources, ensuring that various energy sources operate in the best state, thereby maximizing energy utilization. In order to achieve precise scheduling of the power grid, it is necessary to detect the current, voltage and other power data parameters of the key nodes of the power grid in real time to accurately estimate the power load of the key nodes of the power grid, so that the power grid system can perform energy scheduling more accurately and improve energy utilization.

[0003] Yang Gang, Research on the extended method of harmonic analysis for time-domain filtering and reconstruction of remote sensing data, doctoral dissertation of Wuhan University, 2016. In this study, an iterative interpolation data reconstruction algorithm (IDR algorithm) was proposed. The algorithm first calculates a time series data for comparison. The time series data for comparison is obtained by averaging the two points before and after each point of the original data. Then, the difference series is calculated using the comparison time series data and the original time series data. Then, the points with the largest difference in the original time series data are replaced by the points in the comparison time series data to obtain a new time series data. This replacement is only performed when the maximum difference is greater than 0.02. This process is an iterative calculation process. The new time series data is used as the original data for iterative calculation until the convergence condition is reached.

[0004] Since there is a lot of electromagnetic interference around the working environment of the power grid, there is noise in the collected current data. Therefore, the collected current data needs to be denoised. The traditional IDR algorithm only updates one data in each iteration, so the efficiency is extremely low. Moreover, only using the difference as the condition for updating the data may erase the real data value, thereby accumulating errors in the continuous iteration process, resulting in poor final denoising effect, which affects the accuracy and efficiency of subsequent multi-energy complementary power grid intelligent scheduling. Summary of the invention

[0005] In order to solve the problem that the IDR algorithm only updates one data per iteration, which is inefficient, and only using the difference as the condition for updating the data may erase the real data value, resulting in poor final denoising effect, the present invention proposes a power grid intelligent scheduling method based on multi-energy complementarity, which includes the following steps: Collect current data; obtain the suspected noise level of each current data; Get the noise level of each current data , Represents the noise level of the i-th current data; Represents the suspected noise level of the i-th current data; represents the correlation between the surrounding data of the i-th current data and its corresponding temperature data; exp() represents an exponential function with a natural constant as the base; According to the noise level of each current data, the IDR algorithm is used to denoise the current data to obtain a final current data sequence; the power load of the key nodes of the power grid is estimated based on the final current data sequence.

[0006] The innovation of the present invention lies in that, according to the noise level of the current data, an IDR algorithm is used to denoise the current data of key nodes of the power grid to obtain more accurate power grid loads at the key nodes, so that the power grid system can be dispatched more accurately, the dispatching efficiency is improved, and the energy utilization rate of the multi-energy complementary power grid is improved; further, the noise level of the current data is obtained by analyzing the changing characteristics of the current data in the power grid, so that the number of current data updated each time is adaptively obtained according to the noise level. Compared with the method in which the traditional IDR algorithm only updates one current data with the largest difference each time, the processing efficiency of the IDR algorithm is significantly improved. At the same time, by limiting the noise level, the problem of some normally changing data being smoothed can be avoided, thereby improving the denoising effect and efficiency of the IDR algorithm on the power grid current data.

[0007] Preferably, obtaining the suspected noise level of each current data includes: Obtain surrounding data of each current data; obtain extreme value points in the surrounding data of each current data; ; In the formula, Represents the suspected noise level of the i-th current data; represents the variance of the sampling time intervals between all adjacent extreme value points in the surrounding data of the i-th current data; represents the variance of the current difference between all adjacent extreme points in the surrounding data of the i-th current data; tanh() represents the normalization function; Represents the number of extreme points in the surrounding data of the i-th current data.

[0008] This facilitates the subsequent acquisition of the noise level of each current data based on the suspected noise level.

[0009] Preferably, the step of obtaining surrounding data of each current data includes: The number of sampling moments N is preset, and the current data at the first N sampling moments and the last N sampling moments of the i-th current data are used as the surrounding data of the i-th current data.

[0010] According to the variation characteristics of each current data in its surrounding data, the suspected noise level of each current data can be accurately analyzed.

[0011] Preferably, obtaining the extreme value points in the surrounding data of each current data includes: The difference method is used to obtain the extreme point in the surrounding data of each current data.

[0012] Preferably, obtaining the correlation between the surrounding data of the current data and the corresponding temperature data includes: Obtain temperature data in a sampling time period corresponding to the surrounding data of the i-th current data, and record it as the temperature data corresponding to the surrounding data of the i-th current data; obtain the Pearson correlation coefficient between the surrounding data of the i-th current data and the corresponding temperature data, if the Pearson correlation coefficient is greater than 0, then record the Pearson correlation coefficient between the surrounding data of the i-th current data and the corresponding temperature data as the correlation between the surrounding data of the i-th current data and the corresponding temperature data, otherwise the correlation between the surrounding data of the i-th current data and the corresponding temperature data is 0.

[0013] It is convenient to correct the suspected noise level of each current data according to the correlation and obtain the noise data of each current data.

[0014] Preferably, the method of denoising the current data using an IDR algorithm according to the noise level of each current data to obtain a final current data sequence includes: A difference threshold T and a noise level threshold T1 are preset, and a comparison current data sequence is obtained using an IDR algorithm based on the current data sequence. A difference sequence is obtained based on the current data sequence and the comparison current data sequence. The current data in the difference sequence whose difference is greater than the difference threshold and whose suspected noise level is greater than the noise level threshold is updated using the data in the comparison current data sequence to obtain a new current data sequence. The new current data is then used as the original current data sequence to obtain a difference sequence, and the process stops when the maximum difference in the difference sequence is less than or equal to the difference threshold to obtain a final current data sequence.

[0015] The number of current data updated each time is adaptively obtained according to the noise level of the current data, which significantly improves the processing efficiency of the IDR algorithm and avoids the problem of smoothing some normally changing data, thereby improving the denoising effect and efficiency of the IDR algorithm on the power grid current data.

[0016] Preferably, the collecting current data includes: The current sensors are installed at the key nodes of the power grid to collect current data every two minutes for a total of three hours. The current data at each sampling moment is used as a current data sequence.

[0017] Preferably, the acquisition of the temperature data includes: Temperature sensors are installed at key nodes of the power grid to collect temperature data at sampling intervals of two minutes for a total of three hours. The present invention has the following beneficial effects: the purpose of the present invention is to use an IDR algorithm to denoise the current data of key nodes of a power grid according to the noise level of the current data to obtain more accurate power grid loads at the key nodes, so that the power grid system can be dispatched more accurately, the dispatch efficiency can be improved, and the energy utilization rate of the multi-energy complementary power grid can be improved; further, by analyzing the changing characteristics of the current data in the power grid and calculating its noise level, the number of current data updated each time is adaptively obtained according to the noise level, compared with the method in which the traditional IDR algorithm only updates one current data with the largest difference each time, the processing efficiency of the IDR algorithm is significantly improved, and at the same time, by limiting the noise level, the problem of some normally changing data being smoothed can be avoided, thereby improving the denoising effect and efficiency of the IDR algorithm on the power grid current data. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 It is a flowchart of the steps of a power grid intelligent dispatching method based on multi-energy complementarity in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0020] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0021] See also Figure 1 , which shows a flowchart of a method for intelligent dispatching of a power grid based on multi-energy complementarity provided by an embodiment of the present invention, the method comprising the following steps: S001. Collect current data.

[0022] In the embodiment of the present invention, a sampling moment is taken every two minutes, and the data is collected for a total of three hours. The current sensor and the temperature sensor are installed on the key nodes of the power grid (such as the transformer), and the current data and the temperature data are collected each time; the current data at each sampling moment is taken as the current data sequence, and the temperature data at each sampling moment is taken as the temperature data sequence.

[0023] S002. Obtain extreme value points in surrounding data of each current data, and obtain the suspected noise level of each current data based on the extreme value points.

[0024] It should be noted that the IDR algorithm first calculates a time series data for comparison. The time series data for comparison is obtained by averaging the two points before and after each point of the original data. Then, the difference series is calculated using the comparison time series data and the original time series data. Then, the points with the largest difference in the original time series data are replaced by the points in the comparison time series data to obtain a new time series data. This replacement is performed only when the maximum difference is greater than 0.02. This process is an iterative calculation process. The new time series data is used as the original data for iterative calculation until the convergence condition is reached. It can be seen that the traditional IDR algorithm only updates one data per iteration, so the efficiency is extremely low, and only using the difference as the condition for updating the data may erase the real data value, thereby accumulating errors in the continuous iteration process, resulting in poor final denoising effect.

[0025] Therefore, the present invention obtains the noise level of each current data by analyzing the changing characteristics of the current data, and thus adaptively obtains the number of each updated data according to the noise level of each current data, thereby increasing the restriction conditions for updating the data and improving the efficiency and accuracy of the IDR algorithm.

[0026] It should be further explained that due to the presence of a lot of electromagnetic interference around the working environment of the power grid, there is noise in the collected current data. Such noise manifests itself as random fluctuations in the current data in its local range. Therefore, the random volatility of the current data in its local range can be used to calculate its suspected noise level. The stronger the random volatility, the stronger the suspected noise level.

[0027] The random volatility of current data in its local range manifests itself in two aspects, namely, the randomness of fluctuation time and the randomness of fluctuation amplitude. Therefore, we can first find the extreme points of current data in its local range, because the appearance of extreme points means that the current data has fluctuated up and down. Then, the consistency of the sampling time interval between all two adjacent extreme points can be used to characterize the randomness of fluctuation time. The better the consistency, the worse the randomness of fluctuation time. The consistency of the current difference between all two adjacent extreme points can be used to characterize the randomness of fluctuation amplitude. The better the consistency, the worse the randomness of fluctuation amplitude.

[0028] In the embodiment of the present invention, the number of sampling moments N is preset, and the current data at the N sampling moments before and the N sampling moments after the i-th current data are used as the surrounding data of the i-th current data; in this embodiment, the number of sampling moments N=40 is preset, and in other embodiments, the implementer can preset the value of the number of sampling moments N according to the specific implementation situation. It should be noted that if the number of sampling moments before or after the i-th current data does not meet N, it can be discarded.

[0029] The difference method is used to obtain the extreme point in the surrounding data of each current data.

[0030] Get the suspected noise level for each current data: ; In the formula, Represents the suspected noise level of the i-th current data; represents the variance of the sampling time intervals between all adjacent extreme value points in the surrounding data of the i-th current data; represents the variance of the current difference between all adjacent extreme points in the surrounding data of the i-th current data; tanh() represents the normalization function; represents the number of extreme value points in the surrounding data of the i-th current data; if , indicating that the surrounding data of the i-th current data changes smoothly, so the suspected noise level of the i-th current data is 0; if At this time, the larger the variance of the sampling time intervals between all adjacent extreme points in the surrounding data of the i-th current data, the greater the time fluctuation of the surrounding data of the i-th current data is, the greater the suspected noise level of the i-th current data is; the larger the variance of the current difference between all adjacent extreme points in the surrounding data of the i-th current data is, the greater the numerical fluctuation amplitude of the surrounding data of the i-th current data is, the greater the suspected noise level of the i-th current data is.

[0031] S003. According to the positive correlation between the current data and the temperature data collected at the corresponding time sequence, the suspected noise data of each current data is corrected to obtain the noise level of each current data.

[0032] It should be noted that the above method preliminarily calculates the suspected noise level of each current data. Since the current data may change dramatically in a local range due to factors such as a sudden increase in the load of the power grid during the operation of the power grid, and such drastic changes are similar to data changes affected by noise, if the above method is directly used to obtain the suspected noise level of each current data, the normally changing data points may be detected as noise. Therefore, it is necessary to correct the suspected noise level of each current data. According to Joule's law, heat is generated when current passes through a conductor, and the amount of heat is positively correlated with the current and the power-on time, that is, the rate of change of the current data and the temperature data collected under the time series is positively correlated. Therefore, this positive correlation can be used to correct the suspected noise level of the current data, wherein the more positive the correlation is between the rate of change of the current data and the temperature data collected under the corresponding time series, the lower the suspected noise level of the current data.

[0033] In an embodiment of the present invention, temperature data in a sampling time period corresponding to the surrounding data of the i-th current data is obtained, and recorded as the temperature data corresponding to the surrounding data of the i-th current data; the Pearson correlation coefficient between the surrounding data of the i-th current data and the temperature data corresponding to it is obtained, and if the Pearson correlation coefficient is greater than 0, the Pearson correlation coefficient between the surrounding data of the i-th current data and the temperature data corresponding to it is recorded as the correlation between the surrounding data of the i-th current data and the temperature data corresponding to it, otherwise the correlation between the surrounding data of the i-th current data and the temperature data corresponding to it is 0.

[0034] Get the noise level of each current data: ; In the formula, Represents the noise level of the i-th current data; Represents the suspected noise level of the i-th current data; Represents the correlation between the surrounding data of the i-th current data and its corresponding temperature data; exp() represents an exponential function with a natural constant as the base; if the correlation between the surrounding data of the i-th current data and its corresponding temperature data is greater, it means that the surrounding data of the i-th current data and its corresponding temperature data are positively correlated, and the noise level of the i-th current data is smaller.

[0035] S004. According to the noise level of each current data, the IDR algorithm is used to denoise the current data to obtain the final current data sequence. The power load of the key nodes of the power grid is estimated based on the final current data sequence, so that the power grid system can perform energy scheduling more accurately and improve energy utilization.

[0036] It should be noted that the known IDR algorithm is an algorithm that gradually improves the estimated value of the data through iteration to approach the true value. In time series data denoising, the algorithm can identify and remove the noise components in the data while retaining the original characteristics of the data. However, the traditional IDR algorithm only updates one data per iteration, so the efficiency is undoubtedly relatively low. Moreover, for normal data changes in the current data, if only the difference is updated, it will undoubtedly cause the algorithm to have problems of over-noising or under-noising of the current data, thereby affecting the accuracy and efficiency of the subsequent multi-energy complementary power grid intelligent scheduling. Therefore, according to the noise level of each current data mentioned above, in the process of denoising the current data sequence using the IDR algorithm, the current data can be updated in batches each time it is updated, and the specific number of updates can be determined according to the noise level of the current data.

[0037] In an embodiment of the present invention, a difference threshold T and a noise level threshold T1 are preset, and a comparison current data sequence is obtained based on the current data sequence using an IDR algorithm. A difference sequence is obtained based on the current data sequence and the comparison current data sequence, and current data in the difference sequence whose difference is greater than the difference threshold and whose suspected noise level is greater than the noise level threshold is updated using data in the comparison current data sequence to obtain a new current data sequence. The new current data is then used as the original current data sequence to obtain a difference sequence, and the process stops when the maximum difference in the difference sequence is less than or equal to the difference threshold to obtain a final current data sequence.

[0038] In the embodiment of the present invention, the preset difference threshold value T=0.03 and the noise level threshold value T1=0.8. In other embodiments, the implementer may set the difference threshold value and the noise level threshold value according to the specific implementation situation.

[0039] The power load at key nodes of the power grid is estimated based on the final current data sequence. The system then automatically adjusts the input and output of various energy sources based on information such as the grid load and energy production, thereby achieving optimal utilization of energy and ensuring stable operation of the power grid based on multi-energy complementarity.

[0040] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for intelligent dispatching of power grid based on multi-energy complementarity, characterized in that: include: Collect current data; Obtain the suspected noise level of each current data; Get the noise level of each current data , Represents the noise level of the i-th current data; Represents the suspected noise level of the i-th current data; Represents the correlation between the surrounding data of the i-th current data and its corresponding temperature data; exp() represents an exponential function with a natural constant as the base; According to the noise level of each current data, the IDR algorithm is used to denoise the current data to obtain the final current data sequence; The power load at key nodes of the power grid is estimated based on the final current data sequence.

2. The method for intelligent dispatching of a power grid based on multi-energy complementarity according to claim 1 is characterized in that: The obtaining of the suspected noise level of each current data includes: Obtain surrounding data of each current data; obtain extreme value points in the surrounding data of each current data; ； In the formula, Represents the suspected noise level of the i-th current data; represents the variance of the sampling time intervals between all adjacent extreme value points in the surrounding data of the i-th current data; represents the variance of the current difference between all adjacent extreme points in the surrounding data of the i-th current data; tanh() represents the normalization function; Represents the number of extreme points in the surrounding data of the i-th current data.

3. A method for intelligent dispatching of a power grid based on multi-energy complementarity according to claim 2, characterized in that: The step of obtaining surrounding data of each current data comprises: The number of sampling moments N is preset, and the current data at the first N sampling moments and the last N sampling moments of the i-th current data are used as the surrounding data of the i-th current data.

4. The method for intelligent dispatching of a power grid based on multi-energy complementarity according to claim 2 is characterized in that: The step of obtaining the extreme value points in the surrounding data of each current data comprises: The difference method is used to obtain the extreme point in the surrounding data of each current data.

5. The method for intelligent dispatching of a power grid based on multi-energy complementarity according to claim 1, characterized in that: The acquisition of the correlation between the surrounding data of the current data and the corresponding temperature data includes: Obtain temperature data in a sampling time period corresponding to the surrounding data of the i-th current data, and record it as the temperature data corresponding to the surrounding data of the i-th current data; obtain the Pearson correlation coefficient between the surrounding data of the i-th current data and the corresponding temperature data, if the Pearson correlation coefficient is greater than 0, then record the Pearson correlation coefficient between the surrounding data of the i-th current data and the corresponding temperature data as the correlation between the surrounding data of the i-th current data and the corresponding temperature data, otherwise the correlation between the surrounding data of the i-th current data and the corresponding temperature data is 0.

6. The method for intelligent dispatching of a power grid based on multi-energy complementarity according to claim 1, characterized in that: According to the noise level of each current data, the current data is denoised using the IDR algorithm to obtain a final current data sequence, including: A difference threshold T and a noise level threshold T1 are preset, and a comparison current data sequence is obtained using an IDR algorithm based on the current data sequence. A difference sequence is obtained based on the current data sequence and the comparison current data sequence. The current data in the difference sequence whose difference is greater than the difference threshold and whose suspected noise level is greater than the noise level threshold is updated using the data in the comparison current data sequence to obtain a new current data sequence. The new current data is then used as the original current data sequence to obtain a difference sequence, and the process stops when the maximum difference in the difference sequence is less than or equal to the difference threshold to obtain a final current data sequence.

7. The method for intelligent dispatching of a power grid based on multi-energy complementarity according to claim 1, characterized in that: The collecting current data comprises: The current sensors are installed at the key nodes of the power grid to collect current data every two minutes for a total of three hours. The current data at each sampling moment is used as a current data sequence.

8. The method for intelligent dispatching of a power grid based on multi-energy complementarity according to claim 5, characterized in that: The acquisition of the temperature data includes: Temperature sensors are installed at key nodes of the power grid to collect temperature data at sampling intervals of two minutes for a total of three hours.

Images