A distributed power supply perception method based on state estimation

Abstract

The present invention provides a distributed power perception method based on state estimation, which establishes a distributed photovoltaic power database within a perception area based on a SCADA dispatching system; extracts real-time data of grid-connected photovoltaic power from the SCADA platform database, calculates the real-time output of the regional distributed power source S, and draws a perception curve; evaluates the obtained real-time output value of the regional distributed photovoltaic, compares the integrated electricity with the meter electricity data, and outputs a distributed photovoltaic output data perception curve if the error requirement is met; if the requirement is not met, error correction is performed and step two is repeated until the error requirement is met. The distributed photovoltaic real-time output curve output by the method of the present invention can accurately display the real-time output of regional distributed photovoltaics, provide support for the lean scheduling of distributed power sources in the regional power grid, and improve the economic operation of the power grid.

Description

Technical Field

[0001] The present invention belongs to the technical field of distributed power grid connection, and in particular to a distributed power perception method based on state estimation. Background Art

[0002] To achieve the "dual carbon" goals, my country is rapidly increasing the installed capacity of new energy sources, including distributed photovoltaic power sources. To address the large-scale integration of distributed power sources, State Grid Corporation of China is actively promoting the construction of a new power system that can accommodate large-scale distributed power generation. Consequently, the distribution network is transitioning from a one-way distribution network to a new active power network with interactive supply and demand. The continuous influx of distributed photovoltaic power sources on the low-voltage side poses a threat to the security, economy, and stability of the distribution network. This is especially true when the number of photovoltaic sources connected to a single line exceeds the specified protection limit, which also tests the protection of the power grid. Achieving safe, efficient, and open consumption has become a limiting factor in the development of distributed photovoltaic power and a pressing challenge for power grid development.

[0003] At present, the dispatching monitoring system and the distribution system are unable to monitor and analyze the real-time output of distributed power sources globally and on lines. Under the current technical level, the average output of photovoltaic power sources can only be analyzed from the power generation and online power of low-voltage distributed power sources. With the rapid increase in the installed capacity of distributed photovoltaic power sources, the management problems of high-proportion distributed power sources will become increasingly prominent. Summary of the Invention

[0004] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a distributed power supply perception method based on state estimation, which can realize real-time perception of photovoltaic power sources, especially the perception of regional distributed photovoltaic power sources, and perfectly solve the current common "invisible and intangible" problem of distributed photovoltaics.

[0005] The technical solution adopted by the present invention to solve the technical problem is:

[0006] A distributed power supply sensing method based on state estimation is characterized by the following steps:

[0007] Step 1: Establish a distributed photovoltaic power database within the sensing area based on the SCADA dispatching system;

[0008] Step 2: Extract the real-time data of the centralized grid-connected photovoltaic power source from the SCADA platform database, substitute it into formula (1), calculate the real-time output of the regional distributed power source S, and draw the perception curve;

[0009]

[0010] Where: S is the total output of regional distributed photovoltaic power (MW), m: the year of equipment sensing; n: the year of distributed installation (...2014, 2015...m); S i : The installed capacity of distributed photovoltaic power generation by installed year, MW; k2: The attenuation coefficient of photovoltaic panels per year, %; B: The deviation correction function; X: The unit capacity power generation of centralized grid-connected photovoltaic power generation;

[0011] Step 3: Evaluate the real-time output value of regional distributed photovoltaic power obtained by formula (1). Use the state estimation method to correct the large interference points of distributed photovoltaic power data for the values ​​with large fluctuations. Use Kalman filtering technology to solve the problem of large output error in the distributed power perception process. The state estimation algorithm is shown in formula (2).

[0012] S k+1 =f(s k , x k , t k ) (2)

[0013] Where S k+1 represents an estimated value; S k+1 t k+1 The estimated value of the state at time s k represents the actual value; S k t k At this moment, formula (1) is used to find the actual value; k represents the state vector;

[0014] Step 4: Compare the integrated electricity with the meter electricity data. If the error requirements are met, output the distributed photovoltaic output data perception curve. If the requirements are not met, perform error correction and repeat step 2 until the error requirements are met. The integrated electricity is the integral value of the real-time perception data S curve, and the meter electricity data is the meter power generation settlement electricity provided by the marketing collection system.

[0015] The distributed photovoltaic real-time output curve generated by this method accurately displays the real-time output of regional distributed photovoltaics, supporting the lean scheduling of distributed power sources in regional power grids and improving the economic operation of power grids. Furthermore, rationally arranging power grid operation can effectively increase the proportion of clean energy in the grid and promote the development of new distribution networks with a high proportion of clean energy access.

[0016] This method uses the photovoltaic power source connected to the SCADA system in the divided area as the reference object, avoiding the influence of multiple variables such as light, temperature and other environmental variables on the output perception accuracy of distributed power sources.

[0017] In step 1, a distributed photovoltaic power database is established as a graphic model database based on the SCADA dispatching system, which can meet the integrity, accuracy and real-time update of the database data.

[0018] In step 2, the output perception of distributed photovoltaic power sources in the region divided by weather and the output perception of distributed photovoltaic power sources in the region divided by lines can be calculated, but the function is not limited to this.

[0019] The error correction method described in step 4 is a method for correcting the fluctuation error between the electricity consumption on the settlement day and the curve-integrated electricity consumption.

[0020] The advantages and positive effects of the present invention are:

[0021] This invention utilizes a distributed photovoltaic power generation sensing method based on state estimation to achieve real-time sensing of photovoltaic power sources, effectively resolving the current "invisible and intangible" problem of distributed photovoltaic systems. This method is particularly effective in areas where distributed power generation systems lack the ability to upload real-time status data or experience significant errors in uploaded data. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 4 is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0024] The distributed photovoltaic power supply sensing method based on state estimation of the present invention includes three steps:

[0025] Step 1: Establish a distributed photovoltaic power database within the sensing area based on the SCADA dispatching system. The distributed photovoltaic power database includes classification and statistics of grid-connected distributed power sources, so that each data is labeled based on the principle of data retrieval. Distributed power user labels include household name, household number, installed capacity, access mode, installation year, etc.

[0026] Step 2: Use the real-time data of the centralized photovoltaic power station connected to the region as the reference data. In order to improve the accuracy of distributed photovoltaic output perception, the method divides the distributed photovoltaic users into regions according to the principle of similar meteorological conditions, and selects the real-time data of the centralized photovoltaic power station in each meteorological region as the basic theoretical data for perception. The reference point in this method is that the real-time data has taken into account the weather factors of the day. Therefore, in the perception of distributed power sources, there is no need to separately consider the impact of weather factors such as light, temperature, and humidity on the output of photovoltaic power sources by the perceived equipment, which reduces the control variables. In order to improve accuracy, factors that affect the efficiency of photovoltaic power sources, such as grid-connected voltage level, power station scale, and user nature, are considered in the perception process to achieve real-time perception of the photovoltaic power status of the entire region or line by line, and to grasp the contribution of the distributed power sources of the entire regional network to the entire network. The formula is expressed as follows:

[0027]

[0028] Where:

[0029] m: year of the device perception year;

[0030] n: year of distributed power installation (…2014, 2015…m);

[0031] S i : The installed capacity of distributed photovoltaic power generation by installed capacity year (unit: MW);

[0032] k2: annual attenuation coefficient of photovoltaic panels (%);

[0033] B: bias correction function;

[0034] X: Unit capacity power generation of centralized grid-connected photovoltaic, where: p: real-time active power of the reference power station (MW); S j : Reference centralized grid-connected photovoltaic installed capacity in the region (MW).

[0035] Step 3: This distributed photovoltaic power generation sensing method uses a reference simulation and state estimation approach. Because this method uses real-time data from centralized, grid-connected photovoltaic power plants as a reference for describing the output of distributed photovoltaic power plants, distributed photovoltaic output sensing relies heavily on this reference power plant. Power fluctuations in the reference plant, or maintenance at a centralized power plant within the sensing area, can cause fluctuations in the sensed data of the distributed power generation. Therefore, this sensing method incorporates a state estimation algorithm and a power correction method to further refine the state sensing results.

[0036] The state estimation algorithm is shown in formula (2);

[0037] S k+1 =f(s k , xk , t k ) (2)

[0038] Where S k+1 represents an estimated value; S k+1 t k+1 The estimated value of the state at time s k represents the actual value; S k t k At this moment, formula (1) is used to find the actual value; k represents the state vector;

[0039] From the perspective of data accuracy and perception smoothness, it is advisable to use the previous day's and similar data to estimate the state of the distributed photovoltaic power station output. The state estimation algorithm in this perception method adopts the weighted least squares state estimation algorithm, which weakens the dependence on the centralized grid-connected photovoltaic power station data in the power perception process and more accurately perceives the output of distributed photovoltaic power. The electricity metering device at the distributed photovoltaic power grid point is the grid-connected electricity data collection point. Compared with the active power data uploaded by the metering device, the accuracy of the electricity data is higher than the active power data uploaded by the metering device to the marketing system. In order to improve the perception accuracy of the distributed photovoltaic power output, the electricity correction method is introduced in the perception method. The electricity data in units of days or months is used to correct the active output of the distributed photovoltaic power. The perception accuracy after correction can already more accurately reflect the distributed power supply as shown in formula (3):

[0040]

[0041] W 计量 : Distributed photovoltaic power meter (kWh);

[0042] W 积分 : is the integral value of the distributed photovoltaic perception curve (kWh);

[0043] Step: The number of data points for state perception.

[0044] This sensing method can also be used for photovoltaic power sensing in power lines and power supply areas.

[0045] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A distributed power supply perception method based on state estimation, characterized in that: Here are the steps: Step 1: Establish a distributed photovoltaic power database within the sensing area based on the SCADA dispatching system; Step 2: Extract the real-time data of the centralized grid-connected photovoltaic power source from the SCADA platform database, substitute it into formula (1), calculate the real-time output of the regional distributed power source S, and draw the perception curve; Where: S: total output of regional distributed photovoltaic power, in MW; m: year of the device perception year; n: year of distributed installation; S i : The installed capacity of distributed photovoltaic power generation by installed year, MW; k2: The attenuation coefficient of photovoltaic panels per year, %; B: The deviation correction function; X: The unit capacity power generation of centralized grid-connected photovoltaic power generation; Step 3: Evaluate the real-time output value of regional distributed photovoltaic power obtained by formula (1), and use the state estimation algorithm to correct the large interference points of distributed photovoltaic power data for the values ​​with large fluctuations. The state estimation algorithm is shown in formula (2); S k+1 =f(s k ,x k ,t k ) (2) Where S k+1 represents an estimated value; S k+1 t k+1 The estimated value of the state at time s k represents the actual value; S k t k At this moment, formula (1) is used to find the actual value; k represents the state vector; Step 4: Compare the integrated electricity with the meter electricity data. If the error requirements are met, output the distributed photovoltaic output data perception curve. If the requirements are not met, perform error correction and repeat step 2 until the error requirements are met. The integrated electricity is the integral value of the real-time perception data S curve, and the meter electricity data is the meter power generation settlement electricity provided by the marketing collection system.

2. The method according to claim 1, characterized in that Step 2: Distributed photovoltaic users are divided into regions according to the principle of similar meteorological conditions. In each meteorological region, real-time data of power generation from centralized photovoltaic power stations are selected as the basic theoretical data for perception.

3. The method according to claim 2, characterized in that Step 2: Consider the grid-connected voltage level, power station size, and user nature during the perception process.

4. The method according to claim 3, characterized in that The state estimation algorithm adopts the weighted least squares state estimation algorithm.

5. The method according to claim 4, characterized in that The power correction method is introduced into the perception method, and the power data in units of days or months are used to correct the active output of distributed photovoltaic power sources. The corrected perception accuracy can already reflect the distributed power source more accurately, as shown in formula (3): W 计量 : Distributed photovoltaic power meter measures kWh; W 积分 : is the integral value kWh of the distributed photovoltaic perception curve; Step: The number of data points for state perception.

Images