Distributed new energy monitoring method and system

Abstract

The invention relates to a distributed new energy monitoring method and system, and belongs to the technical field of power distribution networks. The method comprises the following steps: calculating the similarity between historical data of a data maintenance station and historical data of each data base station corresponding to the data maintenance station in a set time length through a gray correlation algorithm, and selecting the data base station with the maximum similarity to calculate a proportionality coefficient of historical power generation power of the data base station and the data maintenance station in the set time length; the generated power corresponding to a data maintenance station is obtained according to the generated power data corresponding to the data base station and the proportionality coefficient, and the data maintenance station comprises distributed new energy of which the generating capacity data is missing or the uploading frequency does not meet the control requirement of a power grid regulation and control strategy; each data base station corresponding to the data maintenance station comprises each distributed new energy, wherein the geographic distance between the distributed new energy and the data maintenance station meets the condition of a close distance, and the historical data and the corresponding generated power data meet the condition of high-quality data.

Description

Technical Field

[0001] The present invention belongs to the technical field of power distribution network, and in particular relates to a distributed new energy monitoring method and system. Background Art

[0002] In recent years, distributed renewable energy has shown "rapid development in installed capacity". Since distributed renewable energy is usually connected from the end of the distribution network with a low access voltage level, after large-scale access, the randomness and intermittency of renewable energy output will aggravate the fluctuation of grid voltage, causing the flow of the distribution network to change from "single direction" to "bidirectional randomness", affecting the stability of the distribution network and the quality of power supply to users. Because distributed new energy points are numerous and widespread, it is difficult to achieve real-time collection and observation with 100% coverage, so the distribution network has problems such as incomplete measurement and slow information exchange. At present, for most distribution networks, usually only some distributed new energy site data equipped with measurement and collection devices are directly connected to the dispatching system with high data quality, and have "observable" and "measurable" capabilities. Most distributed new energy is connected through the distribution network marketing system. Due to communication channels, costs and other reasons, the corresponding measurement devices are not configured or incomplete, resulting in missing or incomplete measurement data of power generation. Only the power generation sent up at the hourly level can be obtained from the marketing system. Therefore, the power generation data is "observable at the hourly level", which cannot meet the observation requirements of real-time control, resulting in the distributed power supply of 380 / 220V connected to the low-voltage distribution network being in the blind spot of the power grid monitoring, unable to meet the power grid monitoring and power control needs and realize the consumption analysis and main distribution coordinated control of distributed new energy. Summary of the invention

[0003] The purpose of the present invention is to provide a distributed renewable energy monitoring method and system to solve the problem in the prior art that the power generation data of distributed renewable energy is missing or incomplete because the measuring device is not configured or incompletely configured, resulting in blind spots in the power grid's monitoring of the power generation of distributed renewable energy.

[0004] In order to solve the above technical problems, the present invention provides a distributed new energy monitoring method, which includes: calculating the similarity of historical data within a set time period between a data maintenance station and each data reference station corresponding to the data maintenance station through a grey correlation algorithm, selecting the data reference station with the largest similarity to calculate the proportional coefficient of the historical power generation within the set time period between the data reference station and the data maintenance station, and obtaining the power generation power corresponding to the data maintenance station according to the power generation power data and the proportional coefficient corresponding to the data reference station, wherein the data maintenance station includes distributed new energy whose power generation data is missing or the upload frequency does not meet the control requirements of the power grid regulation strategy, and each data reference station corresponding to the data maintenance station includes each distributed new energy whose geographical distance to the data maintenance station meets the short distance condition and whose historical data and corresponding power generation power data meet the high-quality data condition.

[0005] Further, the power generation data corresponding to the data reference station includes at least one of real-time power generation data of the data reference station and predicted power generation data of the data reference station; when the power generation data corresponding to the data reference station includes the real-time power generation data of the data reference station, the power generation corresponding to the data maintenance station includes the real-time power generation data of the data maintenance station; when the power generation data corresponding to the data reference station includes the predicted power generation data of the reference station, the power generation corresponding to the data maintenance station includes the predicted power generation data of the data maintenance station.

[0006] Furthermore, the historical data includes historical meteorological data and historical power generation data. The historical meteorological data includes historical daily average temperature; the historical power generation data includes historical daily power generation.

[0007] Further, calculating the similarity of historical data within a set time period between the data maintenance station and the data reference station corresponding to the data maintenance station by using a grey correlation algorithm includes:

[0008]

[0009] Where x0(1) is the historical daily power generation of the data maintenance station, x0(2) is the historical daily average temperature of the data maintenance station, and x j (1) is the historical daily power generation of the jth data base station, x j (2) is the historical daily average temperature of the jth data base station, ρ is the resolution coefficient, σ j (1) is the data maintenance station x0 and the jth data reference station x corresponding to the data maintenance station j The correlation coefficient of daily power generation, σ j (2) is the data maintenance station x0 and the jth data reference station x corresponding to the data maintenance station j The daily average temperature correlation coefficient is , δ is the similarity of historical data within the set time between the data maintenance station and the data reference station corresponding to the data maintenance station, and N is the number of days corresponding to the set time.

[0010] Furthermore, the method of calculating the proportional coefficient of the historical power generation within the set time period of the data reference station and the data maintenance station includes:

[0011]

[0012] Where P n is the ratio of the power generation of the data base station every 5 minutes to the rated power capacity of the base station, cap is the power capacity of the data maintenance station, W i is the actual power generation of the data maintenance station on the i-th day, K iis the proportional coefficient of the i-th day, K is the proportional coefficient of the historical power generation within the set time period between the data reference station and the data maintenance station, and N is the number of days corresponding to the set time period.

[0013] Furthermore, the method of obtaining the real-time power generation data of the data maintenance station according to the real-time power generation data of the data reference station and the proportional coefficient includes:

[0014]

[0015] Where P 0i实时 is the real-time power generation of the data maintenance station at time i, P Ji实时 is the real-time power generation of the data reference station at time i, and K is the ratio coefficient of the historical power generation of the data reference station and the data maintenance station within the set time period.

[0016] Further, the method of obtaining the predicted power generation data of the data maintenance station according to the predicted power generation data of the data reference station and the proportional coefficient includes:

[0017]

[0018] Where P 0i预测 is the predicted power generation data of the data maintenance station at time i, P Ji预测 is the predicted power generation data of the data reference station at time i, and K is the proportional coefficient of the historical power generation within the set time period between the data reference station and the data maintenance station.

[0019] Furthermore, the method for determining whether the geographical distance between the distributed new energy source and the data maintenance station meets the short distance condition includes:

[0020] If the geographical distance between the distributed new energy and the data maintenance station is less than the set distance threshold, it is determined that the geographical distance between the distributed new energy and the data maintenance station meets the short distance condition.

[0021] Furthermore, the method of judging whether the historical data of distributed renewable energy and the corresponding power generation data meet the conditions of high-quality data includes:

[0022] If the historical data of the distributed new energy and the corresponding power generation data are not missing or the upload frequency does not meet the control requirements, it is determined that the historical data of the distributed new energy and the corresponding power generation data meet the conditions of high-quality data.

[0023] To solve the above technical problems, the present invention further provides a distributed new energy monitoring system, including a processor, and the processor is used to execute a computer program to implement the steps of the distributed new energy monitoring method.

[0024] The beneficial effects of the above technical solution are as follows: the present invention provides a new distributed new energy monitoring method. When the data of the data maintenance station is missing or the upload frequency does not meet the control requirements, the data reference station with the highest similarity between the historical data and the data maintenance station is selected from the data reference stations that are geographically close to the data maintenance station and have higher data quality, so that the power generation data corresponding to the data reference station can be used as a data reference for the data maintenance station, and then the proportional relationship between the historical power generation of the data reference station and the data maintenance station is calculated, which is equivalent to obtaining the ratio of power conversion from the data of the data reference station to the data maintenance station. After that, the power generation data corresponding to the data reference station and the proportional relationship can be used to convert the power generation data corresponding to the data maintenance station. In this way, the data of each time scale missing from the distribution network can be supplemented, the observability and measurability of the distribution network distributed power supply of the distribution network can be improved, and the problem of missing or incomplete power generation data of distributed new energy whose measuring device is not configured or incompletely configured in the prior art, resulting in blind spots in the monitoring of the power generation of distributed new energy by the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic flow chart of a distributed new energy monitoring method according to an embodiment of the distributed new energy monitoring method of the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0027] Distributed new energy monitoring method embodiment:

[0028] A distributed new energy monitoring method of the present embodiment uses distributed new energy whose power generation data is missing or whose upload frequency does not meet the control requirements as a data maintenance station, and uses distributed new energy that is geographically close and has high quality historical data and corresponding power generation data as a data reference station, and selects a data reference station whose historical data is similar to the historical data of the data maintenance station. The proportional relationship between the historical power generation of the data reference station and the data maintenance station is obtained by converting the power generation data corresponding to the data reference station through the power generation data corresponding to the data reference station and the proportional relationship.

[0029] The distributed new energy monitoring method specifically includes:

[0030] The similarity of historical data within a set time period between a data maintenance station and each data reference station corresponding to the data maintenance station is calculated by a grey correlation algorithm, and the data reference station with the greatest similarity is selected to calculate the proportional coefficient of the historical power generation within the set time period between the data reference station and the data maintenance station, and the power generation corresponding to the data maintenance station is obtained according to the power generation data and the proportional coefficient corresponding to the data reference station, wherein the data maintenance station includes distributed new energy sources whose power generation data is missing or whose upload frequency does not meet the control requirements, and each data reference station corresponding to the data maintenance station includes each distributed new energy source whose geographical distance to the data maintenance station meets the condition of close distance and whose historical data and corresponding power generation data meet the condition of high-quality data.

[0031] It can be seen that, when the data of the data maintenance station is missing or the upload frequency does not meet the control requirements, the distributed new energy monitoring method selects the data reference station whose historical data is most similar to the data maintenance station from the data reference stations that are geographically close to the data maintenance station and have higher data quality, so as to use the power generation data corresponding to the data reference station as a data reference for the data maintenance station, and then calculate the proportional relationship between the historical power generation of the data reference station and the data maintenance station, which is equivalent to obtaining the ratio of power conversion from the data of the data reference station to the data maintenance station. After that, the power generation data corresponding to the data reference station and the proportional relationship can be used to convert the power generation data corresponding to the data maintenance station through the power generation data corresponding to the data reference station. In this way, the data of each time scale missing from the distribution network can be supplemented, the observability and measurability of the distribution network distributed power supply of the distribution network can be improved, and the problem of missing or incomplete power generation data of distributed new energy whose measuring device is not configured or incompletely configured in the prior art is solved, resulting in blind spots in the monitoring of the power generation of distributed new energy by the power grid.

[0032] Specifically, in this embodiment, firstly, based on the spatial distribution and grey correlation coefficient analysis, the corresponding data reference station is found to realize knowledge transfer; then, through historical samples, the power conversion ratio coefficient of the station that meets the correlation requirements is calculated to achieve accurate estimation; finally, the forecast and real-time data of the station with good data quality are used to roll-calculate the day-ahead short-term forecast, intraday ultra-short-term forecast, and real-time data of the station with poor data quality, so as to fill in the data of various time scales for the missing distributed new energy in the distribution network and improve the observability and measurability of the distributed power supply of the distribution network. The specific process is as follows: Figure 1 shown.

[0033] In this embodiment, the method for determining whether the geographical distance between the distributed new energy source and the data maintenance station meets the short distance condition includes:

[0034] If the geographical distance between the distributed new energy and the data maintenance station is less than the set distance threshold, it is determined that the geographical distance between the distributed new energy and the data maintenance station meets the short distance condition.

[0035] Methods for judging whether the historical data of distributed renewable energy and the corresponding power generation data meet the conditions of high-quality data include:

[0036] If the historical data of the distributed new energy and the corresponding power generation data are not missing or the upload frequency does not meet the control requirements, it is determined that the historical data of the distributed new energy and the corresponding power generation data meet the conditions of high-quality data.

[0037] Specifically, the distributed power sources in the substations connected to the low voltage at the end of the distribution network are not equipped with measurement devices due to communication channels, costs and other reasons, resulting in missing measurement data. They can only obtain the hourly power generation from the marketing system, but cannot obtain the real-time power generation data. Therefore, in response to the problem of incomplete measurement in the distribution network, the distributed new energy monitoring method of this embodiment takes into account the spatial correlation of meteorological conditions within a certain area, and divides the distributed new energy data maintenance grid area according to the geographical distribution and the actual distribution network frame. The method of dividing the data maintenance grid area includes:

[0038] Distributed renewable energy sources with missing data or whose upload frequency does not meet the control requirements are used as data maintenance stations. According to their geographical distribution, renewable energy stations with good measurement data quality and close geographical relationships are selected as data reference stations corresponding to the data maintenance stations (i.e., each data reference station corresponding to the data maintenance station is selected), and correlation analysis is performed on the data maintenance stations and the data reference stations. The input of the correlation analysis can be historical meteorological data and power generation data obtained from the power sales system, that is, in this embodiment, the historical data used to calculate the similarity includes historical meteorological data and historical power generation data, the historical meteorological data includes the historical daily average temperature; the historical power generation data includes the historical daily power generation.

[0039] The historical data used to calculate the similarity includes historical meteorological data and historical power generation data. The method of calculating the similarity of the historical data within a set time period between the data reference station and the data maintenance station corresponding to the data maintenance station by using a grey correlation algorithm specifically includes:

[0040]

[0041]

[0042] Where x0(1) is the historical daily power generation of the data maintenance station, x0(2) is the historical daily average temperature of the data maintenance station, and x j (1) is the historical daily power generation of the jth data base station, x j (2) is the historical daily average temperature of the jth data base station, ρ is the resolution coefficient, σ j (1) is the correlation coefficient of daily power generation, σ j(2) is the daily average temperature correlation coefficient, δ is the similarity of the historical data within the set time period between the data maintenance station and the data reference station corresponding to the data maintenance station, and N is the number of days corresponding to the set time period. That is, the two vectors x0(1) and x0(2) in the grey correlation algorithm represent the daily power generation and daily average temperature of the data maintenance station respectively; in other embodiments, the historical data used to calculate the similarity can also include other historical data in addition to the historical meteorological data and the historical power generation data, then the above x0(1) and x0(2) can be replaced by other historical data, and only need to be calculated according to the above σ j (1) σ j The data of x0(1) and x0(2) to be substituted into the formula (2) can be replaced; x0(3), ..., x0(m), etc. can also be added adaptively, just by increasing the corresponding j (1) σ j The calculation method of (2) is the same as other formulas. The vector composed of two influencing factors of the jth data base station on the same day is x j =[x j (1),x j (2)] Then the above daily power generation correlation coefficient σ j (1) is the data maintenance station x0 and the jth data reference station x j The correlation coefficient of daily power generation and the correlation coefficient of daily average temperature σ j (2) is the data maintenance station x0 and the jth data reference station x j The correlation coefficient of the daily average temperature; the resolution coefficient ρ is 0.5 in this embodiment.

[0043] In this embodiment, the value of N is 7, that is, starting from the nearest historical day, the daily similarity with the data reference station is calculated one by one on the nearest 7 days:

[0044]

[0045] In the formula, δ j It is the similarity of historical data within a set time period between the data maintenance station on the jth day and the data reference station corresponding to the data maintenance station.

[0046] Then the similarity of the seven days is summed to obtain the similarity of the historical data within the set time period between the data maintenance station and the data reference station corresponding to the data maintenance station:

[0047]

[0048] Where δ is the similarity of historical data within a set time period between the data maintenance station and the data reference station corresponding to the data maintenance station.

[0049] In this embodiment, the method of calculating the proportional coefficient of the historical power generation within the time period set by the data reference station and the data maintenance station includes:

[0050]

[0051] Where P n is the ratio of the power generation of the data base station every 5 minutes to the rated power capacity of the base station, cap is the power capacity of the data maintenance station, W i is the actual power generation of the data maintenance station on the i-th day, K i is the ratio coefficient of the historical power generation of the data base station and the data maintenance station on the i-th day, K is the ratio coefficient of the historical power generation of the data base station and the data maintenance station within the set time, and N is the number of days corresponding to the set time.

[0052] Specifically, the power generation ratio coefficient of the data reference station and the data maintenance station is calculated by regression analysis. The power generation P of the previous 7 days of the data standard station is normalized to obtain a typical distributed renewable energy power generation time series library per unit capacity, as shown in Table 1.

[0053] Table 1

[0054] Serial number time Power (kW) 1 2024 / 03 / 01 00:00:00 0.0 2 2024 / 03 / 01 00:05:00 0.00 3 2024 / 03 / 01 00:10:00 0.00 ...... ...... ..... 108 2024 / 03 / 01 09:00:00 0.15 ...... ...... ..... 144 2024 / 03 / 01 12:00:00 0.89 ...... ...... ..... 180 2024 / 03 / 01 15:00:00 0.43 ...... ...... ..... 287 2024 / 03 / 01 23:50:00 0.00 288 2024 / 03 / 01 23:55:00 0.00

[0055] The proportional coefficient of power conversion from the data base station data to the data maintenance station data is calculated by integrating the daily power generation value of the data maintenance station in the previous 7 days and the fixed interval unit capacity power generation power of the data base station (that is, by substituting the data in Table 1 into the above K i The calculation formula is used to calculate the proportional coefficient of power conversion from the data base station data to the data maintenance station data, that is, the proportional coefficient of the historical power generation between the data base station and the data maintenance station within the set time period).

[0056] In this embodiment, the power generation data corresponding to the data reference station includes at least one of the real-time power generation data of the data reference station and the predicted power generation data of the data reference station; when the power generation data corresponding to the data reference station includes the real-time power generation data of the data reference station, the power generation data corresponding to the data maintenance station includes the real-time power generation data of the data maintenance station; when the power generation data corresponding to the data reference station includes the predicted power generation data of the reference station, the power generation data corresponding to the data maintenance station includes the predicted power generation data of the data maintenance station.

[0057] Methods for obtaining the real-time power generation data of the data maintenance station based on the real-time power generation data of the data reference station and the proportional coefficient include:

[0058]

[0059] Where P 0i实时 is the real-time power generation of the data maintenance station at time i, P Ji实时 is the real-time power generation of the data reference station at time i, and K is the ratio coefficient of the historical power generation of the data reference station and the data maintenance station within the set time period.

[0060] Methods for obtaining the predicted power generation data of the data maintenance station based on the predicted power generation data and the proportional coefficient of the data reference station include:

[0061]

[0062] Where P 0i预测 is the predicted power generation data of the data maintenance station at time i, P Ji预测 is the predicted power generation data of the data reference station at time i, and K is the proportional coefficient of the historical power generation within the set time period between the data reference station and the data maintenance station.

[0063] The use of digital technology to solve the problem of observable and measurable low-voltage distributed renewable energy can save huge investments, avoid a lot of manual operation and maintenance work, improve the automation and intelligence level of the system, have broad application scenarios, and improve the monitoring capabilities of the distribution network compared to installing a large number of measurement and collection devices.

[0064] Distributed new energy monitoring system implementation example:

[0065] The present invention also provides a distributed new energy monitoring system, including a processor, the processor is used to execute a computer program to implement the steps of the distributed new energy monitoring method. The specific process, principle and effect of the distributed new energy monitoring system have been described in detail in the embodiment of the distributed new energy monitoring method, and will not be repeated here.

[0066] Specific implementation methods are given above, but the present invention is not limited to the described implementation methods. The basic idea of ​​the present invention lies in the above basic scheme. For ordinary technicians in this field, it does not take creative work to design various deformed models, formulas, and parameters according to the teachings of the present invention. Changes, modifications, substitutions, and variations of the implementation methods without departing from the principles and spirit of the present invention still fall within the scope of protection of the present invention.

Claims

1. A distributed new energy monitoring method, characterized in that: The method comprises: calculating the similarity of historical data within a set time period between a data maintenance station and each data reference station corresponding to the data maintenance station by a grey correlation algorithm, selecting the data reference station with the greatest similarity to calculate the proportional coefficient of historical power generation within the set time period between the data reference station and the data maintenance station, and obtaining the power generation corresponding to the data maintenance station according to the power generation data corresponding to the data reference station and the proportional coefficient, wherein the data maintenance station comprises distributed new energy sources whose power generation data is missing or whose transmission frequency does not meet the control requirements of the power grid regulation strategy, and each data reference station corresponding to the data maintenance station comprises distributed new energy sources whose geographical distance to the data maintenance station meets the condition of close distance and whose historical data and corresponding power generation data meet the condition of high-quality data.

2. The distributed new energy monitoring method according to claim 1, characterized in that: The power generation data corresponding to the data reference station includes at least one of the real-time power generation data of the data reference station and the predicted power generation data of the data reference station; when the power generation data corresponding to the data reference station includes the real-time power generation data of the data reference station, the power generation corresponding to the data maintenance station includes the real-time power generation data of the data maintenance station; when the power generation data corresponding to the data reference station includes the predicted power generation data of the reference station, the power generation corresponding to the data maintenance station includes the predicted power generation data of the data maintenance station.

3. The distributed new energy monitoring method according to claim 1 or 2, characterized in that: The historical data includes historical meteorological data and historical power generation data. The historical meteorological data includes historical daily average temperature; the historical power generation data includes historical daily power generation.

4. The distributed new energy monitoring method according to claim 3 is characterized in that: The similarity of historical data within a set time period between a data maintenance station and a data reference station corresponding to the data maintenance station is calculated by using a grey correlation algorithm, including: Where x0(1) is the historical daily power generation of the data maintenance station, x0(2) is the historical daily average temperature of the data maintenance station, and x j (1) is the historical daily power generation of the jth data base station, x j (2) is the historical daily average temperature of the jth data base station, ρ is the resolution coefficient, σ j (1) is the data maintenance station x0 and the jth data reference station x corresponding to the data maintenance station j The correlation coefficient of daily power generation, σ j (2) is the data maintenance station x0 and the jth data reference station x corresponding to the data maintenance station j The daily average temperature correlation coefficient is , δ is the similarity of historical data within the set time between the data maintenance station and the data reference station corresponding to the data maintenance station, and N is the number of days corresponding to the set time.

5. The distributed new energy monitoring method according to claim 1 or 2, characterized in that: The method of calculating the proportionality coefficient of the historical power generation of the data reference station and the data maintenance station within a set time period includes: Where P n is the ratio of the power generation of the data base station every 5 minutes to the rated power capacity of the base station, cap is the power capacity of the data maintenance station, W i is the actual power generation of the data maintenance station on the i-th day, K i is the proportional coefficient of the i-th day, K is the proportional coefficient of the historical power generation within the set time period between the data reference station and the data maintenance station, and N is the number of days corresponding to the set time period.

6. The distributed new energy monitoring method according to claim 2, characterized in that: Methods for obtaining the real-time power generation data of the data maintenance station based on the real-time power generation data of the data reference station and the proportional coefficient include: Where P 0i实时 is the real-time power generation of the data maintenance station at time i, P Ji实时 is the real-time power generation of the data reference station at time i, and K is the ratio coefficient of the historical power generation of the data reference station and the data maintenance station within the set time period.

7. The distributed new energy monitoring method according to claim 2, characterized in that: Methods for obtaining the predicted power generation data of the data maintenance station based on the predicted power generation data of the data reference station and the proportional coefficient include: Where P 0i预测 is the predicted power generation data of the data maintenance station at time i, P Ji预测 is the predicted power generation data of the data reference station at time i, and K is the proportional coefficient of the historical power generation within the set time period between the data reference station and the data maintenance station.

8. The distributed new energy monitoring method according to claim 1 or 2, characterized in that: Methods for determining whether the geographical distance between the distributed renewable energy source and the data maintenance station meets the short distance condition include: If the geographical distance between the distributed new energy and the data maintenance station is less than the set distance threshold, it is determined that the geographical distance between the distributed new energy and the data maintenance station meets the short distance condition.

9. The distributed new energy monitoring method according to claim 3, characterized in that: Methods for judging whether the historical data of distributed renewable energy and the corresponding power generation data meet the conditions of high-quality data include: If the historical data of the distributed new energy and the corresponding power generation data are not missing or the upload frequency does not meet the control requirements, it is determined that the historical data of the distributed new energy and the corresponding power generation data meet the conditions of high-quality data.

10. A distributed new energy monitoring system, comprising a processor, characterized in that: The processor is used to execute a computer program to implement the steps of the distributed new energy monitoring method according to any one of claims 1 to 9.

Images