Data communication method of distributed power grid-connected controller based on cloud edge collaboration

Abstract

The application relates to the technical field of data communication, in particular to a data communication method of a distributed power supply grid-connected controller based on cloud edge cooperation. The method acquires running time sequence data of each storage battery in a distributed power supply system at a current moment; for any storage battery, the data corresponding to the current moment in the running time sequence data of the storage battery is taken as target data, a local window of the target data is acquired in the running time sequence data, a running mode deviation factor of the storage battery at the current moment is acquired according to a path deviation factor of each data in the local window of the target data; the data transmission priority of each storage battery is acquired according to the running mode deviation factor of each storage battery at the current moment, the target data of each storage battery is transmitted to a cloud computing server, the grid-connected mode of each storage battery in the distributed power supply system is regulated, and the accuracy of the transmission priority of the running data of each storage battery in the distributed power supply system is improved.

Description

Technical Field

[0001] This invention relates to the field of data communication technology, and more specifically to a data communication method for a distributed power grid-connected controller based on cloud-edge collaboration. Background Technology

[0002] In distributed power systems, grid-connected controllers are needed to control the charging and discharging status of batteries. This control is based on operational data such as the charging and discharging power of the batteries. While the grid-connected controller can analyze and process the collected operational data, to ensure the lifespan of the batteries and the quality of power supply from the grid, it needs to process the data after acquiring it. This data determines the priority for transmitting the operational data of each battery to a cloud computing server. The controller then transmits this data according to priority. The cloud computing server, based on the received data, determines the grid connection status of each battery and adjusts the grid connection mode accordingly to ensure efficient grid operation.

[0003] In existing technologies, the transmission priority of operational data for each battery in a distributed power system is determined by first acquiring the operational sequence data of each battery, then analyzing the operational sequence data to obtain outlier indicators for measuring abnormal changes, and finally determining the importance of each battery based on these outlier indicators, and then assigning a transmission priority to each battery's operational data according to its importance. However, in the process of acquiring outlier indicators, the judgment is based on the pattern deviation of a single data point in the operational sequence data. During the operation of the battery, it is impossible to determine whether a single outlier data point is monitoring abnormal data or data indicating a change in the battery's operating state. This results in a significant deviation in the transmission priority obtained through existing connectivity-based data deviation methods, which in turn prevents the cloud computing server from timely adjusting the grid connection mode of batteries that have changed their operating state.

[0004] Therefore, improving the accuracy of the transmission priority of the operating data of each battery in a distributed power system, so as to ensure that the cloud computing server can timely regulate the grid connection mode of the batteries, has become an urgent problem to be solved. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a data communication method for a distributed power grid-connected controller based on cloud-edge collaboration, in order to solve the problem of how to improve the accuracy of the transmission priority of the operating data of each battery in a distributed power system, so as to ensure that the cloud computing server can timely regulate the grid connection mode of the batteries.

[0006] This invention provides a data communication method for a distributed power grid-connected controller based on cloud-edge collaboration, the method comprising the following steps:

[0007] Obtain the runtime sequence data of each battery in the distributed power system at the current moment;

[0008] For any battery, the data corresponding to the current moment in the battery's runtime sequence data is taken as the target data. A local window of the target data is obtained in the runtime sequence data. Based on the data changes between the data in the local window, the path offset factor of each data in the local window is obtained respectively.

[0009] Based on the path offset factor of each data point in the local window of the target data and the data difference between adjacent data points, the operating mode deviation factor of the battery at the current moment is obtained.

[0010] Based on the operating mode deviation factor of each battery at the current moment, the data transmission priority of each battery is obtained. Based on the data transmission priority of each battery, the target data of each battery is transmitted to the cloud computing server. The cloud computing server is used to receive the operating data of each battery transmitted by the grid-connected controller in order to regulate the grid connection mode of each battery in the distributed power system.

[0011] Furthermore, if the target data is the last data in the local window, then obtaining the path offset factor for each data in the local window based on the data changes between data in the local window includes:

[0012] For any data in the local window, obtain the timestamp distance between the data and the target data, take all data from the target data to the data as the neighborhood data of the data, and obtain the path change vector of each neighborhood data.

[0013] The cosine similarity between the path change vectors of two adjacent neighboring data is obtained respectively. The cosine similarity is multiplied to obtain the corresponding product result.

[0014] The product result between the timestamp distance and the product result is normalized, and the normalized result is used as the path offset factor of the data.

[0015] Furthermore, obtaining the path change vector for each of the neighborhood data includes:

[0016] Construct the time-series data change curve of the runtime sequence data;

[0017] For any neighborhood data, obtain the adjacent data after the neighborhood data. Based on the two data points corresponding to the adjacent data and the neighborhood data on the time-series data change curve, obtain the vector between the two data points as the path change vector of the neighborhood data.

[0018] Furthermore, obtaining the battery's operating mode deviation factor at the current moment based on the path offset factor of each data point in the local window of the target data and the data difference between adjacent data includes:

[0019] For any data in a local window of the target data, obtain the data change between the previous data and the data, obtain the product between the path offset factor of the data and the data change, and obtain the first summation result of all products based on the product corresponding to each data in the local window of the target data.

[0020] In the runtime sequence data, obtain the local window of each data in the local window of the target data respectively. Based on the local window of each data in the local window of the target data, obtain the second summation result of each data in the local window of the target data respectively, and calculate the average value of all the second summation results.

[0021] The ratio between the first sum and the average value is normalized, and the normalized result is used as the operating mode deviation factor of the battery at the current moment.

[0022] Furthermore, obtaining the data transmission priority of each battery based on its operating mode deviation factor at the current moment includes:

[0023] The operating mode deviation factors of each battery at the current moment are sorted from largest to smallest to obtain the corresponding sorting results. The data transmission priority of each battery is determined based on the sorting results.

[0024] Furthermore, the step of acquiring the runtime sequence data of each battery in the distributed power system at the current moment includes:

[0025] For any battery in a distributed power system, obtain the battery's operating data at the current moment, obtain the battery's historical operating data within a preset period before the current moment, and arrange the historical operating data and the operating data at the current moment in chronological order to obtain the battery's operating sequence data at the current moment.

[0026] The embodiments of the present invention have at least the following beneficial effects:

[0027] This invention acquires the runtime sequence data of each battery in a distributed power system at the current moment. For any battery, the data corresponding to the current moment in the battery's runtime sequence data is taken as target data. A local window of the target data is obtained in the runtime sequence data. Based on the data changes between data in the local window, a path offset factor is acquired for each data in the local window. Based on the path offset factor of each data in the local window of the target data and the data difference between adjacent data, the operating mode deviation factor of the battery at the current moment is acquired. Based on the operating mode deviation factor of each battery at the current moment, the data transmission priority of each battery is acquired. Based on the data transmission priority of each battery, the target data of each battery is transmitted to a cloud computing server. The cloud computing server is used to receive the operating data of each battery transmitted by the grid-connected controller to regulate the grid connection mode of each battery in the distributed power system. By measuring the path deviation factor formed by the real-time operating data of any battery, the deviation of the operating mode at the specified time can be measured during the identification of changes in the battery's operating mode. This is achieved by comparing the local operating data changes with historical operating data changes in the real-time operating data of the battery. This avoids deviations in the priority judgment of transmitting the operating data of each battery in the distributed power system to the cloud computing server, improves the accuracy of the transmission priority of the operating data of each battery in the distributed power system, and ensures that the cloud computing server can timely regulate the grid connection mode of the batteries, thus enabling the grid to operate efficiently. Attached Figure Description

[0028] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1A flowchart illustrating the steps of a data communication method for a distributed power grid-connected controller based on cloud-edge collaboration, as provided in an embodiment of the present invention. Detailed Implementation

[0030] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a data communication method for a distributed power grid-connected controller based on cloud-edge collaboration proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0031] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0032] The following description, in conjunction with the accompanying drawings, details a specific scheme for a data communication method for a distributed power grid-connected controller based on cloud-edge collaboration provided by the present invention.

[0033] The specific scenario addressed by this invention is as follows: During the regulation of the grid-connected mode of multiple batteries in a distributed power system, the grid-connected controller of the distributed power system collects the operating data of each battery, analyzes the data transmission priority of each battery based on the collected operating data, and realizes data communication between the grid-connected controller and the cloud computing server based on the data transmission priority. Thus, after the cloud computing server receives the operating data of each battery, it promptly regulates the grid-connected mode of each battery in the distributed power system based on the received operating data to ensure the efficient operation of the power grid.

[0034] Please see Figure 1 The diagram illustrates a flowchart of a data communication method for a distributed power grid-connected controller based on cloud-edge collaboration, according to an embodiment of the present invention. The method includes the following steps:

[0035] Step S101: Obtain the runtime sequence data of each battery in the distributed power system at the current moment.

[0036] In a distributed power system, there are multiple distributed power units, and each distributed power unit is a power system of one type of energy. Therefore, the distributed power system in this embodiment of the invention includes power systems of multiple types of energy, such as photovoltaic power generation and wind power generation. For each type of energy, after the distributed power unit completes power generation, it is necessary to store electrical energy through batteries, and then connect the distributed power system to the power grid through a grid-connected controller.

[0037] In the operation of a distributed power system, in order to ensure the normal operation of the batteries and extend their service life, it is necessary to monitor the operating data of each battery, so as to determine the operating status of the batteries in a timely manner and adjust the grid connection mode of the batteries.

[0038] In this embodiment of the invention, the operating data of each battery in the distributed power system is collected in real time, and the collected operating data of each battery is synchronized to the grid-connected controller. To ensure the accuracy of the distributed power system during the grid-connected mode regulation of the batteries, a cloud-edge collaborative approach is required. Specifically, the grid-connected mode regulation of each battery is achieved through collaboration between the cloud computing server and the grid-connected controller. The grid-connected controller transmits the operating data of each battery to the cloud computing server. The cloud computing server evaluates the grid-connected mode based on the received operating data and then adjusts the grid-connected mode of each battery according to the evaluation results.

[0039] However, in the coordination process between the cloud computing server and the grid-connected controller, to ensure the accuracy of the grid connection mode control for all batteries, it is necessary to determine the data transmission priority of each battery before the grid-connected controller transmits the operational data of each battery to the cloud computing server. This priority is then used to enable data communication between the cloud computing server and the grid-connected controller. Therefore, for the control of the grid connection mode of each battery at the current moment, before the grid-connected controller transmits the operational data of each battery to the cloud computing server, the operating sequence data of each battery in the distributed power system at the current moment must be obtained.

[0040] Preferably, the step of acquiring the runtime sequence data of each battery in the distributed power system at the current moment includes:

[0041] For any battery in a distributed power system, obtain the battery's operating data at the current moment, obtain the battery's historical operating data within a preset period before the current moment, and arrange the historical operating data and the operating data at the current moment in chronological order to obtain the battery's operating sequence data at the current moment.

[0042] It should be noted that the operating data of a battery includes, but is not limited to, the battery's SOC value, charging power, or discharging power. The SOC value reflects the battery's remaining capacity and is defined numerically as the ratio of the remaining capacity to the battery's total capacity, usually expressed as a percentage.

[0043] For example, for any battery in a distributed power system, the charging power of that battery at the current moment and its historical charging power for the previous week are obtained. The charging power and historical charging power are then sorted according to the collection time to obtain the time-series data of the battery's charging power at the current moment. There is no restriction on the sampling frequency of the battery's charging power; for example, it can be collected every 5 seconds or every hour.

[0044] Step S102: For any battery, take the data corresponding to the current moment in the battery's runtime sequence data as the target data, obtain a local window of the target data in the runtime sequence data, and obtain the path offset factor of each data in the local window according to the data changes between the data in the local window.

[0045] In this embodiment of the invention, considering that when the operating state of the battery changes abnormally, the battery's runtime sequence data will show continuous data deviation, if only the outlier obtained by conventional outlier analysis methods (such as the outlier algorithm based on connectivity) is used to determine whether the battery's runtime data at each moment is deviated, it will be insensitive to the situation where the battery's operating state changes abnormally continuously, thus failing to guarantee the accuracy of the data transmission priority of each battery. Therefore, it is necessary to first obtain the weight of each local data in the runtime sequence data corresponding to the current moment in the analysis of the battery's operating mode deviation state at the current moment based on the local data of the data at the current moment.

[0046] Taking a battery as an example, the data corresponding to the current time in the runtime sequence data of that battery is taken as the target data. For example, the data corresponding to the current time t in the runtime sequence data of the i-th battery is the target data X. i,t After determining the target data, a local window of the target data is obtained from the runtime sequence data. In this embodiment of the invention, the length of the local window is set to 20, and the target data is the last data in the local window. That is, the local window of the target data is composed of a total of 20 data in the runtime sequence data, including the target data.

[0047] After determining the local window of the target data, the path offset factor of each data in the local window is obtained according to the data changes between the data in the local window. The path offset factor is used to characterize the weight of the data. The larger the path offset factor, the greater the weight, which can provide more effective information when analyzing the deviation of the battery's operating mode at the current moment.

[0048] Preferably, the step of obtaining the path offset factor for each data item in the local window based on the data changes between data items in the local window includes:

[0049] (1) For any data in the local window, obtain the timestamp distance between the data and the target data, take all data from the target data to the data as the neighborhood data of the data, and obtain the path change vector of each neighborhood data.

[0050] In this embodiment of the invention, for the target data X corresponding to the i-th battery at the current time t... i,t The j-th data point in the local window is obtained by comparing the j-th data point with the target data X, since each data point in the runtime sequence data corresponds to a sampling time. i,t The timestamp distance between two data points is the sampling time interval between them.

[0051] Furthermore, the j-th data in the runtime sequence data is mapped to the target data X. i,t All data are considered as the neighborhood data of the j-th data. For example, the neighborhood data of the j-th data are the (j+1)-th, (j+2)-th, (j+3)-th, ...-th data in the runtime sequence data (that is, the target data X corresponding to the current time step in the runtime sequence data). i,t There are a total of Nj neighborhood data, where N is the total number of data contained in the runtime sequence data.

[0052] After determining the neighborhood data of the j-th data, the path change vector of each neighborhood is obtained, wherein obtaining the path change vector of each neighborhood data includes:

[0053] Construct the time-series data change curve of the runtime sequence data;

[0054] For any neighborhood data, obtain the adjacent data after the neighborhood data. Based on the two data points corresponding to the adjacent data and the neighborhood data on the time-series data change curve, obtain the vector between the two data points as the path change vector of the neighborhood data.

[0055] For example, if any neighboring data of the j-th data is the (j+1)-th data in the runtime sequence data, then the path change vector of the (j+1)-th data is the vector obtained by pointing from the coordinate point corresponding to the (j+2)-th data on the time sequence data change curve of the runtime sequence data to the coordinate point corresponding to the (j+1)-th data, and is denoted as the path change vector of the (j+1)-th data.

[0056] It should be noted that the time-series data change curve for the runtime data is constructed by using the data acquisition time as the horizontal axis and the runtime data as the vertical axis. The specific construction method is existing technology and will not be elaborated here.

[0057] (2) Obtain the cosine similarity between the path change vectors of two adjacent neighboring data respectively, and perform a product operation on all cosine similarities to obtain the corresponding product result.

[0058] Based on the method for calculating the cosine similarity between two vectors, the cosine similarity between the path change vectors of each pair of adjacent neighboring data is obtained, and all cosine similarities are multiplied to obtain the corresponding product result. The multiplication operation is a prior art technique and will not be elaborated upon here.

[0059] (3) The product result between the timestamp distance and the product result is normalized, and the normalized result is used as the path offset factor of the data.

[0060] In one embodiment, the i-th battery corresponds to the target data X at the current time t. i,t The expression for calculating the path offset factor of any data in a local window is:

[0061]

[0062] in, This indicates that the target data X corresponding to the i-th battery at the current time t is... i,t The path offset factor of the j-th data in the local window, softmax() represents the normalized exponential function, d j,t Represents target data X i,t The j-th data in the local window and the target data X i,t The timestamp distance between them, Nj represents the target data X i,t The total number of neighboring data of the j-th data in the local window, which is also the number of data points from the j-th data in the local window to the target data X. i,t The total number of data points between them, where cos() represents the cosine similarity function. Indicates from target data X i,t The path change vector of the (o-1)th data point forward is also the path change vector of the neighboring data of the j-th data point in the local window. Indicates from target data X i,t The path change vector of the o-th data point forward. Represents target data X i,t The product of the path change vectors of each data point in the local window up to the j-th data point.

[0063] It should be noted that for the i-th battery at the current time t, the target data X is... i,t The path offset factor of the j-th data in the local window is essentially a factor of the target data X. i,tThe path distance weighting of data within a local window during outlier factor measurement aims to influence subsequent calculations of the target data X. i,t When calculating local link distances, the focus is on data with continuously changing patterns, thus reflecting the target data X in the subsequent calculation of the deviation factor of the operating pattern. i,t The difference between the corresponding data changes and the historical operational data changes, therefore, it is necessary to first determine the target data X. i,t The path change of each data point is determined within a local window. For the j-th data point X in the window... i,t,j It needs to be done through X i,t-1 →X i,t The corresponding data change vector is used to continuously judge forward, thereby determining X. i,t,j Corresponding X i,t,j-1 →X i,t,j The path deviation factor of the path change in the target data X i,t Within a local window, the purpose of performing a series of cosine similarity products from the path change vector corresponding to the last data point forward is to obtain the target data X. i,t As a standard, its previous consecutive data are compared with the target data X. i,t The judgment is made based on paths with similar path change vectors, that is, using X... i,t-1 →X i,t As a standard, anything similar to this path is considered to be related to the target data X. i,t Those with the same operating mode have a higher corresponding weight, and through To target data X i,t The length of the multiplication process within a local window is limited, that is, the distance from the target data X. i,t The farther away, the lower the corresponding weight.

[0064] Thus, by using the method for obtaining the path offset factor, it is possible to obtain the path offset factor of each data point in a local window of the target data for any battery.

[0065] Step S103: Based on the path offset factor of each data in the local window of the target data and the data difference between adjacent data, obtain the battery's operating mode deviation factor at the current moment.

[0066] In this embodiment of the invention, after obtaining the path offset factor of each data in the local window of the target data, the continuous data change pattern is highlighted during the evaluation of the operating mode deviation factor of the target data. This focuses the data offset on the continuous pattern change of the battery's operating data, making the analysis of abnormal changes in the battery's operation at the current moment more accurate. Therefore, the operating mode deviation factor of the battery at the current moment is obtained based on the path offset factor of each data in the local window of the target data and the data difference between adjacent data.

[0067] Preferably, obtaining the battery's operating mode deviation factor at the current moment based on the path offset factor of each data point in the local window of the target data and the data difference between adjacent data includes:

[0068] For any data in a local window of the target data, obtain the data change between the previous data and the data, obtain the product between the path offset factor of the data and the data change, and obtain the first summation result of all products based on the product corresponding to each data in the local window of the target data.

[0069] In the runtime sequence data, obtain the local window of each data in the local window of the target data respectively. Based on the local window of each data in the local window of the target data, obtain the second summation result of each data in the local window of the target data respectively, and calculate the average value of all the second summation results.

[0070] The ratio between the first sum and the average value is normalized, and the normalized result is used as the operating mode deviation factor of the battery at the current moment.

[0071] In one embodiment, the formula for calculating the operating mode deviation factor of any battery at the current time t is:

[0072]

[0073] Where, ε i,t Let represent the deviation factor of the operating mode of the i-th battery at the current time t, Norm() represent the normalization function, and M represent the total number of data points contained in the local window. This indicates that the target data X corresponding to the i-th battery at the current time t is... i,t The path offset factor of the j-th data in the local window, X i,t,j-1 This indicates that the target data X corresponding to the i-th battery at the current time t is... i,t The (j-1)th data in the local window, X i,t,jThis indicates that the target data X corresponding to the i-th battery at the current time t is... i,t The j-th data in the local window, (X i,t,j-1 →X i,t,j ) represents the target data X corresponding to the i-th battery at the current time t. i,t The amount of data change from the (j-1)th data point to the jth data point in the local window. This represents the target data X corresponding to the i-th battery at the current time t. i,t When the y-th data in a local window is the target data, the path offset factor X corresponding to the j-th data in the local window. i,y,j-1 This represents the target data X corresponding to the i-th battery at the current time t. i,t When the y-th data in a local window is the target data, the corresponding data in the (j-1)-th local window is X. i,y,j This represents the target data X corresponding to the i-th battery at the current time t. i,t When the y-th data in a local window is the target data, the corresponding j-th data in the local window, (X i,y,j-1 →X i,y,j ) represents the target data X corresponding to the i-th battery at the current time t. i,t When the y-th data in a local window is the target data, the change in data from the (j-1)-th data to the j-th data in the corresponding local window.

[0074] It should be noted that, based on existing connectivity-based outlier factor algorithms, this method uses the path offset factor of each data point within a local window of the target data to weight the data changes corresponding to each path during the calculation of the average link distance ratio. This approach prioritizes continuous data pattern changes within the runtime sequence data during the comparison process, rather than changes in the pattern corresponding to a single data point within a local window. This aims to optimize the connectivity-based outlier factor. Therefore, when the target data X... i,t The greater the deviation from the average link distance within its local window, the more it indicates that the target data X... i,t If the deviation is from the local window, it means the battery's operating state at the current time t is different from before, indicating an abnormality in the battery's operating state at the current time t. Therefore, the deviation factor ε for the battery's operating mode at the current time t is... i,t The larger.

[0075] Step S104: Based on the deviation factor of the operating mode of each battery at the current moment, obtain the data transmission priority of each battery, and transmit the target data of each battery to the cloud computing server according to the data transmission priority of each battery.

[0076] In this embodiment of the invention, according to the method in steps S102 to S103, the operating mode deviation factor of each battery in the distributed power system at the current moment can be obtained. Then, based on the operating mode deviation factor of each battery at the current moment, the data transmission priority of each battery is obtained. Specifically, the operating mode deviation factors of each battery at the current moment are sorted from largest to smallest to obtain the corresponding sorting result. The data transmission priority of each battery is determined based on the sorting result. The larger the operating mode deviation factor, the more it indicates that the target data at the current moment t has experienced a data deviation in the battery's operating sequence data, that is, the abnormal change in the battery's operating state at the current moment t, and the more necessary it is to adjust its grid connection mode.

[0077] After determining the data transmission priority of each battery at the current moment, the target data of each battery at the current moment is transmitted to the cloud computing server according to the data transmission priority of each battery at the current moment. The cloud computing server is used to receive the operating data of each battery transmitted by the grid-connected controller in order to regulate the grid connection mode of each battery in the distributed power system. Therefore, after the cloud computing server receives the target data of each battery at the current moment, the cloud computing server obtains the decision result corresponding to the regulation of the grid connection mode of the battery through the grid connection mode regulation model of the battery, and then regulates the grid connection mode of the corresponding battery in a timely manner according to the decision result to ensure the efficient operation of the power grid.

[0078] It should be noted that the specific process by which the cloud computing server adjusts the grid connection mode of each battery based on the received operating data of each battery is not the focus of this invention, and therefore will not be described in detail in the embodiments of this invention.

[0079] In summary, the embodiments of the present invention acquire the runtime sequence data of each battery in the distributed power system at the current moment. For any battery, the data corresponding to the current moment in the battery's runtime sequence data is taken as the target data. A local window of the target data is obtained in the runtime sequence data. Based on the data changes between data in the local window, the path offset factor of each data in the local window is obtained. Based on the path offset factor of each data in the local window of the target data and the data difference between adjacent data, the operating mode deviation factor of the battery at the current moment is obtained. Based on the operating mode deviation factor of each battery at the current moment, the data transmission priority of each battery is obtained. Based on the data transmission priority of each battery, the target data of each battery is transmitted to the cloud computing server. The cloud computing server is used to receive the operating data of each battery transmitted by the grid-connected controller to regulate the grid connection mode of each battery in the distributed power system. By measuring the path deviation factor formed by the real-time operating data of any battery, the deviation of the operating mode at the specified time can be measured during the identification of changes in the battery's operating mode. This is achieved by comparing the local operating data changes with historical operating data changes in the real-time operating data of the battery. This avoids deviations in the priority judgment of transmitting the operating data of each battery in the distributed power system to the cloud computing server, improves the accuracy of the transmission priority of the operating data of each battery in the distributed power system, and ensures that the cloud computing server can timely regulate the grid connection mode of the batteries, thus enabling the grid to operate efficiently.

[0080] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0081] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0082] 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 within the protection scope of the present invention.

Claims

1. A data communication method for a distributed power grid-connected controller based on cloud-edge collaboration, characterized in that, The data communication method includes: Obtain the runtime sequence data of each battery in the distributed power system at the current moment; For any battery, the data corresponding to the current moment in the battery's runtime sequence data is taken as the target data. A local window of the target data is obtained in the runtime sequence data. Based on the data changes between the data in the local window, the path offset factor of each data in the local window is obtained respectively. Based on the path offset factor of each data point in the local window of the target data and the data difference between adjacent data points, the operating mode deviation factor of the battery at the current moment is obtained. Based on the deviation factor of the operating mode of each battery at the current moment, the data transmission priority of each battery is obtained. Based on the data transmission priority of each battery, the target data of each battery is transmitted to the cloud computing server. The cloud computing server is used to receive the operating data of each battery transmitted by the grid-connected controller in order to regulate the grid connection mode of each battery in the distributed power system. If the target data is the last data in the local window, then the step of obtaining the path offset factor for each data in the local window based on the data changes between data in the local window includes: For any data in the local window, obtain the timestamp distance between the data and the target data, take all data from the target data to the data as the neighborhood data of the data, and obtain the path change vector of each neighborhood data. The cosine similarity between the path change vectors of two adjacent neighboring data is obtained respectively. The cosine similarity is multiplied to obtain the corresponding product result. The product result between the timestamp distance and the product result is normalized, and the normalized result is used as the path offset factor of the data. The step of obtaining the battery's operating mode deviation factor at the current moment based on the path offset factor of each data point in the local window of the target data and the data difference between adjacent data includes: For any data in a local window of the target data, obtain the data change between the previous data and the data, obtain the product between the path offset factor of the data and the data change, and obtain the first summation result of all products based on the product corresponding to each data in the local window of the target data. In the runtime sequence data, obtain the local window of each data in the local window of the target data respectively. Based on the local window of each data in the local window of the target data, obtain the first summation result of each data in the local window of the target data respectively, and calculate the average value of all the first summation results. The ratio between the first sum and the average value is normalized, and the normalized result is used as the operating mode deviation factor of the battery at the current moment.

2. The data communication method as described in claim 1, characterized in that, The step of obtaining the path change vector for each of the neighborhood data includes: Construct the time-series data change curve of the runtime sequence data; For any neighborhood data, obtain the adjacent data after the neighborhood data. Based on the two data points corresponding to the adjacent data and the neighborhood data on the time-series data change curve, obtain the vector between the two data points as the path change vector of the neighborhood data.

3. The data communication method as described in claim 1, characterized in that, The step of obtaining the data transmission priority of each battery based on the operating mode deviation factor of each battery at the current moment includes: The operating mode deviation factors of each battery at the current moment are sorted from largest to smallest to obtain the corresponding sorting results. The data transmission priority of each battery is determined based on the sorting results.

4. The data communication method as described in claim 1, characterized in that, The step of acquiring the runtime sequence data of each battery in the distributed power system at the current moment includes: For any battery in a distributed power system, obtain the battery's operating data at the current moment, obtain the battery's historical operating data within a preset period before the current moment, and arrange the historical operating data and the operating data at the current moment in chronological order to obtain the battery's operating sequence data at the current moment.

Images