Water-light complementation day-ahead optimization scheduling method and system considering source-load collaborative risk
By constructing a robust model of photovoltaic output uncertainty and water-light complementarity recently optimized scheduling model, combined with waveform step-by-step compensation strategy, the problems of source-load coordination difficulty and power abandonment risk in new energy systems are solved, and more stable power is achieved Operation and risk control.
Patent Information
- Application Number
- CN202410956245.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-17
- Publication Date
- 2024-11-08
AI Technical Summary
With the high proportion of new energy sources such as photoelectricity and other complementary systems, the system's power generation power is random and difficult to predict, resulting in an increase in the difficulty of source-load coordination, which is prone to the problem of decreasing synchronization and difficult to control the risk of power abandonment.
Build a robust model of photovoltaic output uncertainty, and establish a water-optical complementarity that considers the source-load coordination risk through analysis of waveform feature fragments and uncertainty scene simulation, combining the hydropower operating environment and the degree of photovoltaic uncertainty conservation. The scheduling model adopts the waveform order compensation strategy with the minimum net load fluctuation as the objective function, and conducts real-time scheduling simulation to avoid the risks of power fluctuations and power abandonment.
It effectively avoids the risks of power fluctuations and power abandonment, gives full play to the flexibility of hydropower regulation, provides more stable operation mode and safe and reliable power support, and meets the future development orientation of multi-energy complementary systems.
Smart Images

Figure CN118920602A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to new energy technology, and in particular to a method and system for optimizing the day-ahead scheduling of water-photovoltaic complementarity considering source-load coordination risks. Background Art
[0002] With the development of renewable energy generation and high-voltage direct current transmission technology such as water, wind, and light, the low-carbon and reliability of the power system can be effectively taken into account by coordinating the complementary development and utilization of various clean energy sources and bundling and grid-connected transmission. However, since photovoltaic output is affected by meteorological conditions and has strong random fluctuations, with the high proportion of photovoltaic and other new energy sources connected to the complementary system, the system power generation will show obvious randomness and unpredictability, further aggravating the difficulty of source-load coordination and being unfavorable to the safe and stable operation of the power system. It is easy to have problems such as reduced source-load synchronization and balancing capacity, and difficult to control the risk of power abandonment. Summary of the invention
[0003] Purpose of the invention: One purpose of the present invention is to provide a method for optimizing the day-ahead scheduling of hydro-photovoltaic complementarity that takes into account the risk of source-load coordination and can effectively avoid power fluctuations and power abandonment risks, and can provide a more stable operation mode and safe and reliable power support in multiple scenarios; another purpose of the present invention is to provide a system corresponding to the method.
[0004] Technical solution: The method for optimizing the day-ahead scheduling of water-solar complementary power considering the source-load coordination risk of the present invention comprises the following steps:
[0005] Construct a robust model of photovoltaic output uncertainty, analyze the characteristic segments of the photovoltaic output sequence waveform and simulate the photovoltaic output uncertainty scenario; including: analyze the spatiotemporal distribution characteristics of the photovoltaic output sequence waveform segments affected by weather disturbances, divide the photovoltaic output sequence waveform segments into multiple initial segments, and identify the initial segments to form the characteristic segments of photovoltaic output; on the basis of the traditional cloud model, couple the qualitative characteristics of the characteristic segments of the photovoltaic output sequence waveform, statistically analyze the prediction deviation information from the two dimensions of the photovoltaic output sequence waveform and output, and obtain the simulation value of the photovoltaic output prediction deviation;
[0006] Construct a day-ahead optimization dispatch model for hydropower-photovoltaic complementarity: Taking the minimum net load fluctuation as the peak-shaving objective function, and taking the hydropower operation environment constraints, the conservative degree of photovoltaic uncertainty and the waveform-order compensation strategy as constraints, a day-ahead optimization dispatch model for hydropower-photovoltaic complementarity considering the source-load coordination risk is constructed;
[0007] Run the day-ahead-real-time scheduling simulation process of the hydro-solar complementary system: Use the hydro-solar complementary day-ahead optimization scheduling model to solve the output process of the hydropower and hydro-solar complementary system, apply the compiled power generation plan to the real-time scheduling simulation process, and obtain the system fluctuation and power abandonment index values under real-time scheduling.
[0008] Furthermore, the process of analyzing the waveform characteristics of the photovoltaic output sequence is as follows:
[0009] The photovoltaic output sequence is divided into multiple initial segments, and the photovoltaic output calculation formula of the initial segment n is:
[0010]
[0011] in, is the photovoltaic output of the initial segment n; P t s is the photovoltaic output in time period t; are the starting and ending time periods of the initial segment n respectively;
[0012] According to the fluctuation duration parameter, the fluctuation intensity parameter of the initial segment of photovoltaic output is calculated, and the expression is:
[0013]
[0014] in, is the fluctuation intensity, which indicates the sharpness of the peak shape of the initial segment n; is the fluctuation duration, indicating the duration of the initial segment n; is the photovoltaic output of the initial segment n in time period t; is the average PV output of all initial segments in time period t; μ is the PV installed capacity; represents the standard deviation of the PV output of the initial segment n;
[0015] The characteristic fragment of photovoltaic output is formed by the principle of merging adjacent initial fragments of the same type and reorganizing different types of segments. The photovoltaic output calculation formula of characteristic fragment i is:
[0016]
[0017] Among them, P i s is the photovoltaic output of feature segment i; J represents clustering through K-means, and the characteristic parameters are optimized according to the historical photovoltaic output sequence The value of .
[0018] Furthermore, the mechanism of simulating the PV output uncertainty scenario is:
[0019] (1) Input the photovoltaic output value of the characteristic segment i sample into the reverse cloud generator and calculate the qualitative characteristics (E i,j ,F i,j ,H i,j ), E i,j 、F i,j , H i,jare the expected value, fluctuation amplitude, and discrete degree of the photovoltaic output prediction deviation, respectively; i is the waveform segment number, and j is the output interval number;
[0020] (2) The qualitative characteristics of the photovoltaic output characteristic fragment samples (E i,j ,F i,j ,H i,j ) is input into the forward cloud generator to generate a cloud cluster (ε' u,i,j ,μ i,j ):
[0021] F′ i,j =randn(1)×H i,j +F i,j
[0022] ε' u,i,j =randn(1)×F′ i,j +E i,j
[0023]
[0024] Among them, u is the sample number; ε' u,i,j is the simulated value of photovoltaic output prediction deviation; μ i,j is the probability distribution of the simulated value of the photovoltaic output prediction deviation; F′ i,j is the fluctuation range of the simulated value of the PV output forecast deviation; randn(1) represents a normal distribution with a mean of 0 and a variance of 1;
[0025] (3) Repeat steps (1) to (2) until a cloud of i×j classification intervals is generated that meets the required number of ε' u,i,j , calculate the simulated photovoltaic output:
[0026] P t se = {P t s +ε' u,i,j |P t s ∈(i,j)}
[0027] Among them, P t se , P t s are respectively the simulated photovoltaic output and the predicted photovoltaic output in time period t; according to P t s The classification interval is used to extract the simulated value of the predicted photovoltaic output and generate the simulated photovoltaic output.
[0028] Furthermore, the peak load regulation objective function expression is:
[0029]
[0030] P t re =L t -P t ,P t =P t h +P t se
[0031] Among them, F is the net load fluctuation, which indicates the relative fluctuation value between the output of the water-solar hybrid system and the grid load; T is the optimized scheduling period; L t , P t re , P t , P t h They are the grid load, residual load, water-solar hybrid system output, and hydropower output in time period t respectively; It is the average value of the remaining load within the optimized dispatch period.
[0032] Furthermore, the load tracking capability and the risk of power abandonment are guaranteed by taking the hydropower operation environment constraints, the conservative degree of photovoltaic uncertainty and the waveform-order compensation strategy as constraints, specifically:
[0033] (1) Hydropower operation environment constraints:
[0034]
[0035]
[0036] Among them, V l,t-1 、V l,t are the storage capacities of the l-level hydropower station at the beginning and end of time period t respectively; For l-1 hydropower station in the period The downstream flow rate; Q l,t is the discharge flow of the l-level hydropower station in time period t; q l,t is the flow rate in front of the dam of the l-level hydropower station in time period t; is the water flow lag time between hydropower stations l and l-1, and △t is the calculation step length. l,0 is the water level at the beginning of the l-level hydropower station; Z l,T The water level at the end of the period of the level l hydropower station; are the control boundaries of the start and end periods of the l-level hydropower station; Z l,t is the water level of the l-level hydropower station in time period t; are the lower and upper limits of the water level of the l-level hydropower station in time period t; Q l,t is the power generation flow of the l-level hydropower station in time period t; are the lower and upper limits of the power generation flow of the l-level hydropower station in time period t respectively; is the output of the l-level hydropower station in time period t; They are the lower and upper limits of the output range of the l-level hydropower station in time period t;
[0037] (2) Conservative degree of photoelectric uncertainty:
[0038] F i,j ={α·F i,j |α∈(0.8,1.2)}
[0039] H i,j ={β·H i,j |β∈(0.8,1.2)}
[0040] Among them, F i,j represents the fluctuation amplitude characteristic set of photovoltaic output, H i,j It represents the discrete degree feature set of photovoltaic output, α and β are robustness factors; the larger α is, the lower the prediction accuracy and the larger the fluctuation range; the larger β is, the higher the disturbance complexity and the more frequent the fluctuation;
[0041] (3) Waveform-order compensation strategy:
[0042] a. Fluctuation control: Constructing the first-order compensation boundary based on the characteristic fragments of the PV output uncertainty scenario
[0043]
[0044] in, is the first-order compensation boundary, consisting of the variable set composition; The set of PV output variables simulated for different characteristic segments, P i is the photovoltaic output variable; t 2i is the time variable; To simulate the photovoltaic output collection;
[0045] b. Power abandonment control: The system output further tracks the grid load through hydropower compensation to form a second-order compensation boundary
[0046] P t ≤G
[0047] Among them, G is the maximum capacity of the transmission channel.
[0048] Furthermore, the real-time scheduling simulation process is specifically as follows:
[0049] The predicted photovoltaic output is input into the photovoltaic output uncertainty robust model, and K groups of simulated photovoltaic outputs are generated according to the values of robust factors α and β to form a photovoltaic uncertainty set;
[0050] The day-ahead optimization dispatch model of hydropower and photovoltaic complementarity is transferred into the simulated photovoltaic output of the kth group as well as the runoff and load information. The dynamic programming algorithm DP is used to solve the output process of the hydropower and hydropower complementary system and prepare the power generation plan.
[0051] The actual photovoltaic output is input into the real-time dispatching process, and the hydropower output is re-corrected by determining the water supply based on electricity. The water storage and supply sequence of hydropower stations at all levels is determined by the discriminant coefficient method to offset and balance the photovoltaic output deviation value, and transmit electricity according to the power generation plan within the regulation capacity and controllable range of the cascade hydropower stations;
[0052] Extract the corrected output of the hydro-photovoltaic complementary system in real-time scheduling, where the corrected output of the hydro-photovoltaic complementary system is the sum of the photovoltaic output and the corrected hydropower output. The output of the hydro-photovoltaic complementary system that exceeds the power generation plan will result in abandoned power.
[0053] The simulated photovoltaic output of group K was fully dispatched, and the output was the reservoir capacity, water level and fluctuation and abandoned power index values of the water-photovoltaic complementary system under real-time dispatching in all scenarios.
[0054] Furthermore, the calculation formula for the power abandonment index is:
[0055]
[0056] Among them, C is the amount of power abandoned; P t h' Corrected hydropower output for time period t.
[0057] The system corresponding to the method includes:
[0058] The robust model building unit is used to build a robust model of photovoltaic output uncertainty, analyze the characteristic segments of the photovoltaic output sequence waveform and simulate the photovoltaic output uncertainty scenario; including: analyzing the spatiotemporal distribution characteristics of the photovoltaic output sequence waveform segments affected by weather disturbances, dividing the photovoltaic output sequence waveform segments into multiple initial segments, and identifying the initial segments to form the characteristic segments of photovoltaic output; based on the traditional cloud model, coupling the qualitative characteristics of the characteristic segments of the photovoltaic output sequence waveform, statistically predicting the deviation information from the two dimensions of the photovoltaic output sequence waveform and output, and obtaining the simulated value of the photovoltaic output prediction deviation;
[0059] The dispatch model construction unit is used to construct a day-ahead optimal dispatch model for hydro-photovoltaic complementarity: taking the minimum net load fluctuation as the peak-shaving objective function, and taking the hydropower operation environment constraints, the conservative degree of photovoltaic uncertainty and the waveform-order compensation strategy as constraints, a day-ahead optimal dispatch model for hydro-photovoltaic complementarity considering the source-load coordination risk is constructed;
[0060] The simulation scheduling unit is used to run the day-ahead-real-time scheduling simulation process of the hydro-photovoltaic complementary system: the hydro-photovoltaic complementary day-ahead optimization scheduling model is used to solve the output process of the hydropower and hydro-photovoltaic complementary system, and the compiled power generation plan is applied to the real-time scheduling simulation process to obtain the system fluctuation and abandoned power index values under real-time scheduling.
[0061] An electronic device for storing and executing the method, the device comprising:
[0062] A memory storing executable program code;
[0063] a processor coupled to the memory;
[0064] The processor calls the executable program code stored in the memory to execute the steps of the method for optimizing the day-ahead scheduling of water-solar complementarity considering source-load coordination risks.
[0065] A computer-readable storage medium for storing the method, wherein the computer-readable storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the steps of the method for optimizing the day-ahead scheduling of water-solar complementarity considering source-load coordination risks.
[0066] Beneficial effects: Compared with the prior art, the significant technical effects of the present invention are as follows: (1) Based on the waveform and output of the photovoltaic output sequence, the qualitative characteristics of the photovoltaic output are extracted, and the coupled cloud-generated photovoltaic uncertainty scenario is introduced. The day-ahead optimization scheduling model is established with the minimum net load fluctuation as the objective function. The waveform step-by-step compensation strategy is adopted to ensure the load tracking capability and control the risk of power abandonment, and a power generation plan is compiled and applied to the day-ahead-real-time scheduling simulation operation process, which effectively avoids power fluctuations and power abandonment risks; (2) Considering the impact of photovoltaic volatility and uncertainty on the source-load coordination risk, the flexibility of hydropower regulation is fully utilized, and it has a stronger risk-bearing capacity. The transmission form meets the source-load coordination requirements, which can effectively avoid the risk of power abandonment. At the same time, it is in line with the future development direction of the multi-energy complementary system: it can provide a more stable operation mode and safe and reliable power support in multiple scenarios. BRIEF DESCRIPTION OF THE DRAWINGS
[0067] Figure 1 is a flow chart of the method of the present invention;
[0068] Figure 2 is a flow chart of constructing a robust model of photovoltaic output uncertainty in the method of the present invention;
[0069] Figure 3 It is a PV output characteristic segment division diagram;
[0070] Figure 4 It is a simulation diagram of photovoltaic output scenario;
[0071] Figure 5 A flow chart of constructing a day-ahead optimization scheduling model for water-photovoltaic complementarity considering source-load synergy risk in the method of the present invention;
[0072] Figure 6 It is a flow chart of day-ahead-real-time scheduling simulation;
[0073] Figure 7 This is a simulation comparison chart of the day-ahead and real-time scheduling scenarios. DETAILED DESCRIPTION
[0074] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The following embodiments are used to illustrate the present invention but are not intended to limit the scope of the present invention.
[0075] Embodiment 1
[0076] like Figure 1 As shown, a method for optimizing the day-ahead scheduling of water-solar complementary power considering the source-load coordination risk of the present invention comprises the following steps:
[0077] S1. Construct a robust model of photovoltaic output uncertainty, analyze the characteristic fragments of photovoltaic output sequence waveform and simulate photovoltaic output uncertainty scenarios;
[0078] Firstly, from the perspective of the peak shape and distribution of the photovoltaic output sequence, the spatiotemporal distribution characteristics of the photovoltaic output sequence waveform fragments affected by weather disturbances are analyzed, and the photovoltaic output sequence waveform fragments are divided into multiple initial fragments; the initial fragments are clustered and identified using the fluctuation intensity and fluctuation duration parameters to form the characteristic fragments of the photovoltaic output; then, based on the traditional cloud model, the qualitative characteristics of the characteristic fragments of the photovoltaic output sequence waveform are coupled, and the prediction deviation information is statistically analyzed from the two dimensions of the photovoltaic output sequence waveform and output, to obtain the simulation value of the photovoltaic output prediction deviation;
[0079] A water-photovoltaic complementary system is selected for example analysis. The water-photovoltaic complementary system consists of 6 cascade hydropower stations and 3 distributed photovoltaic power stations, with installed capacities of 9.07 million kW and 10 million kW respectively, and a 10 million kW transmission channel. The 6 cascade hydropower stations are ①, ②, ③, ④, ⑤, and ⑥ from upstream to downstream. Among them, cascade hydropower station No. ② has annual regulation performance and plays a major regulatory role. The basic parameters of each cascade hydropower station are shown in Table 1.
[0080] Table 1 Scale parameters of cascade hydropower stations
[0081]
[0082] In the embodiment, the day-ahead and real-time scheduling of the water-photovoltaic complementary system are based on the scheduling cycle of the day and the calculation step of the hour. The sampling interval of photovoltaic output information is 1 hour, and the predicted photovoltaic output of the next day is calculated by combining weather forecast with long short-term memory network; due to the strong regulation and storage capacity of cascade hydropower stations, the uncertainty of short-term runoff can be ignored. The hydropower boundary adopts the average daily flow of 680m on a certain day in a typical normal water year. 3 / s; Since the grid load demand is much higher than the output of the hydro-photovoltaic complementary system, the volatility and uncertainty of the load have little impact, and the load uses the historical operating data of the day.
[0083] like Figure 2 As shown, the specific steps include:
[0084] S11. Analyze the characteristic fragments of the photovoltaic output sequence waveform according to the photovoltaic output. The specific process is as follows:
[0085] Combining the three typical meteorological conditions of sunny, cloudy and rainy, from the perspective of the peak shape and distribution of the photovoltaic output sequence, the temporal and spatial distribution characteristics of the photovoltaic output sequence waveform fragments affected by weather disturbances are analyzed. According to the intensity and discreteness of the fluctuation, it can be divided into three characteristic types: smooth fluctuation, peak fluctuation, and sawtooth fluctuation. Read the peak and valley positions of the photovoltaic output, record the time period information of adjacent peaks and valleys, and divide the photovoltaic output sequence into multiple initial fragments based on the basic principle of dividing a single initial fragment by adjacent peaks and valleys. The photovoltaic output calculation formula for the initial fragment n is:
[0086]
[0087] in, is the photovoltaic output of the initial segment n, kW; P t s is the photovoltaic output in time period t, kW; is the starting and ending time period of the initial segment n, h;
[0088] According to the fluctuation duration parameter, the fluctuation intensity parameter of the initial segment of photovoltaic output is calculated, and the expression is:
[0089]
[0090] in, is the fluctuation intensity, indicating the sharpness of the peak shape of the initial segment n, h -1 ; is the fluctuation duration, indicating the duration of the initial segment n, h; is the PV output of the initial segment n in time period t, kW; is the average PV output of all initial segments in time period t; μ is the PV installed capacity, kW; represents the standard deviation of the PV output of the initial segment n;
[0091] The characteristic fragment of photovoltaic output is formed by the principle of merging adjacent initial fragments of the same type and reorganizing different types of segments. The photovoltaic output calculation formula of characteristic fragment i is:
[0092]
[0093] Among them, P i s is the photovoltaic output of feature segment i, kW; J represents clustering implemented by K-means, and the characteristic parameters are optimized according to the historical photovoltaic output sequence The value of .
[0094] S12, performing uncertainty scenario simulation based on photovoltaic output characteristic fragments to generate simulated photovoltaic output, specifically including:
[0095] Step 1: Input the photovoltaic output value of the characteristic segment i sample into the reverse cloud generator and calculate the qualitative characteristics (E i,j ,F i,j ,H i,j ), E i,j 、F i,j , H i,j are the expected value, fluctuation amplitude, and discrete degree of the photovoltaic output prediction deviation, respectively; i is the waveform segment number, and j is the output interval number;
[0096] Step 2: The qualitative features of the photovoltaic output feature fragment samples (E i,j ,F i,j ,H i,j ) is input into the forward cloud generator to generate a cloud cluster (ε' u,i,j ,μ i,j ):
[0097] F′ i,j =randn(1)×H i,j +F i,j (4)
[0098] ε' u,i,j =randn(1)×F′ i,j +E i,j (5)
[0099]
[0100] Among them, u is the sample number, ε' u,i,j is the simulated value of photovoltaic output prediction deviation, kW; μ i,j is the probability distribution of the simulated value of the photovoltaic output prediction deviation; F′ i,jis the fluctuation range of the simulated value of the PV output forecast deviation; randn(1) represents a normal distribution with a mean of 0 and a variance of 1;
[0101] Step 3: Repeat steps 1 to 2 until the number of clouds in i×j classification intervals is enough to meet the specific demand. u,i,j , calculate the simulated photovoltaic output:
[0102] P t se = {P t s +ε′ u,i,j |P t s ∈(i,j)} (7)
[0103] Among them, P t se , P t s They are respectively the simulated photovoltaic output and the predicted photovoltaic output in time period t, kW; according to P t s The classification interval is used to extract the simulated value of the predicted photovoltaic output and generate the simulated photovoltaic output.
[0104] like Figure 3 As shown in the figure, the characteristic segments of photovoltaic output are divided according to the photovoltaic output sequence, and the waveform parameters of the three types of characteristic segments of photovoltaic output are The cluster centers are: [0.73, 0.92, 0.36], [10.57, 4.13, 6.22]; the output interval is set to 300,000 kW. According to the measured and predicted photovoltaic output information, the entropy and super entropy values of the output prediction deviation of the characteristic fragments of the photovoltaic output sequence are statistically analyzed.
[0105] According to the qualitative characteristics of the above photovoltaic output characteristic fragments, the forward cloud generator is input to generate 1000 sets of prediction deviation simulation values to form a photovoltaic output set with a 90% guarantee rate. Figure 4 As shown in the figure, both the predicted and measured PV outputs are inside the set, and the first-order compensation boundary takes the outer boundary of the set as the compensation object, forming a stepped boundary with 2 output variables and 6 time variables.
[0106] S2. Construct a day-ahead optimization scheduling model for hydro-photovoltaic complementarity considering source-load coordination risks;
[0107] The optimal scheduling of hydropower-photovoltaic complementarity takes into account the constraints of the hydropower operation environment, the conservative degree of photovoltaic uncertainty, and the waveform-order compensation strategy. The minimum net load fluctuation is taken as the peak-shaving objective function, and a hydropower-photovoltaic complementarity day-ahead optimal scheduling model considering the source-load coordination risk is constructed. Figure 5 As shown, the following steps are included:
[0108] S21. The day-ahead optimization scheduling model of hydro-photovoltaic complementary takes the minimum net load fluctuation as the objective function:
[0109]
[0110] P t re =L t -P t ,P t =P t h +P t se (9)
[0111] Among them, F is the net load fluctuation, which represents the relative fluctuation value between the output of the water-solar hybrid system and the grid load, kW; T is the optimized scheduling period; L t , P t re , P t , P t h They are grid load, residual load, water-solar hybrid system output, and hydropower output in time period t, kW; It is the average value of the remaining load within the optimized dispatch period.
[0112] S22. The day-ahead optimization dispatch model for hydro-solar hybrid avoids power fluctuations and power abandonment risks under different forecast levels through constraint functions, including:
[0113] (1) Hydropower operation environment constraints:
[0114]
[0115] Among them, V l,t-1 、V l,t are the storage capacities of the l-level hydropower station at the beginning and end of time period t, m 3 ; For l-1 hydropower station in the period The downstream flow rate; Q l,t is the discharge flow of the l-level hydropower station in time period t; q l,t is the flow rate in front of the dam of the l-level hydropower station in time period t, m 3 / s; is the water flow lag time between hydropower stations l and l-1, △t is the calculation step length, h. l,0 is the water level at the beginning of the l-level hydropower station; Z l,T The water level at the end of the period of the level l hydropower station; is the control boundary of the start and end period of the l-level hydropower station, m; Z l,t is the water level of the l-level hydropower station in time period t; is the lower and upper limits of the water level of the l-level hydropower station in time period t, m; Q l,t is the power generation flow of the l-level hydropower station in time period t; is the lower and upper limits of the power generation flow of the l-level hydropower station in time period t, m 3 / s; is the output of the l-level hydropower station in time period t; is the lower and upper limits of the output range of the l-level hydropower station in time period t, kW;
[0116] (2) Conservative degree of photoelectric uncertainty:
[0117] F i,j ={α·F i,j |α∈(0.8,1.2)} (15)
[0118] H i,j ={β·H i,j |β∈(0.8,1.2)} (16)
[0119] Among them, F i,j represents the fluctuation amplitude characteristic set of photovoltaic output, H i,j It represents the discrete degree feature set of photovoltaic output, where α and β are robust factors, ranging from 0.9 to 1.1. The larger the α, the lower the prediction accuracy and the larger the fluctuation range; the larger the β, the higher the disturbance complexity and the more frequent the fluctuation.
[0120] (3) Waveform-order compensation strategy:
[0121] a. Fluctuation control: Construct the first-order compensation boundary based on the characteristic fragments of the PV output uncertainty scenario:
[0122]
[0123] in, is the first-order compensation boundary, consisting of the variable set Composition, kW; The set of PV output variables simulated for different characteristic segments, P i is the photovoltaic output variable, kW; t 2i is the time variable, h; is the simulated photovoltaic output set, kW;
[0124] b. Power abandonment control: The output of the hydropower-solar complementary system further tracks the grid load through hydropower compensation to form a second-order compensation boundary:
[0125] P t ≤G (19)
[0126] Where G is the maximum capacity of the transmission channel, kW.
[0127] S3. The day-ahead-real-time dispatch simulation process uses the day-ahead optimization dispatch model of hydropower and photovoltaic complementarity to solve the output process of the hydropower and hydropower complementary system. The compiled power generation plan is applied to the real-time dispatch simulation process to obtain the fluctuation amount and abandoned power index values of the hydropower complementary system under real-time dispatch.
[0128] like Figure 6 As shown in the figure, the power generation plan compiled by the day-ahead dispatch needs to be applied to the real dispatch scenario, and then analyze the impact of the power generation plan on the actual operation of the hydro-photovoltaic complementary system. The specific process is as follows:
[0129] Step 1: Input the predicted PV output into the PV output uncertainty robust model, generate K groups of simulated PV outputs according to the values of robust factors α and β, and form a photovoltaic uncertainty set;
[0130] Step 2: Transfer the day-ahead optimization scheduling model of hydropower and photovoltaic complementarity, input the kth group of simulated photovoltaic output and runoff and load information, use the dynamic programming algorithm (DP) to solve the output process of hydropower and hydropower complementary system, and prepare the power generation plan;
[0131] Step 3: Input the actual PV output into the real-time dispatching process, re-correct the hydropower output by “determining water supply based on electricity”, and use the discriminant coefficient method widely used in load distribution to determine the water storage and supply order of hydropower stations at all levels, offset and balance the PV output deviation, and deliver electricity according to the power generation plan within the regulation capacity and controllable range of the cascade hydropower stations;
[0132] Step 4: Extract the corrected output of the hydro-photovoltaic complementary system in real-time scheduling, where the corrected output of the hydro-photovoltaic complementary system is the sum of the photovoltaic output and the corrected hydropower output. The output of the hydro-photovoltaic complementary system that exceeds the power generation plan forms the abandoned power, and the abandoned power index is calculated:
[0133]
[0134] Where C is the amount of power wasted, kW; P t h' Corrected hydropower output for time period t, kW;
[0135] Step 5: The simulated photovoltaic output of group K is fully dispatched, and the output is the reservoir capacity, water level and fluctuation and power abandonment index values of the water-photovoltaic complementary system under real-time dispatching in all scenarios.
[0136] Conventional methods and the present invention are introduced to solve example scenarios and analyze the effects. The objective function of the conventional method is consistent with that of the present invention, but the waveform order compensation strategy is not considered, and the predicted photovoltaic output or simulated photovoltaic output is introduced, and a deterministic scenario optimization method is adopted; to ensure the reliability of the comparison effect, both methods are solved using dynamic programming algorithms and day-ahead-real-time scheduling.
[0137] Conventional method and the present invention day-ahead scheduling are as follows Figure 7 As shown in (a) and (b), the hydropower output is from bottom to top in the order of ①, ②, ③, ④, ⑤, and ⑥ hydropower station output processes. In the power generation plan compiled by the conventional method, the fluctuation of the objective function value is 713,000 kW, while the fluctuation of the objective function value in the present invention is 1,357,000 kW, which can effectively guarantee the load tracking capability and improve the risk bearing capacity of the water-solar complementary system.
[0138] In the actual execution of power generation plan, under the scenario of robust factor = 1, the conventional method of real-time scheduling is shown in Figure 7 In (c), due to the misalignment between the peak periods of measured and predicted photovoltaic output, power abandonment occurred during the stage of high photovoltaic penetration. The reasons for the abandonment are: the photovoltaic output is at its peak at 13h, which directly exceeds the power generation plan and cannot be fully absorbed and connected to the grid, resulting in power abandonment; and hydropower is subject to the minimum output constraint from 12 to 16h, resulting in continuous power abandonment. During the actual operation, the system abandoned 3.59 million kW of power, accounting for 2.9% of the power loss;
[0139] Correspondingly, the real-time scheduling of the present invention is shown in Figure 7 In (d), waveform-ordered compensation constraints are imposed in two characteristic segments from 12 to 14 hours and from 16 to 17 hours: first, hydropower output is used to smooth the volatility of photovoltaic power under different characteristic segments, improve the system transmission stability, and effectively ensure the load tracking capability; second, based on the waveform uncertainty characteristics, reasonable capacity is reserved for hydropower output adjustment and balance to avoid power abandonment caused by the output of the hydropower complementary system exceeding the channel. In actual operation, the system power abandonment is as low as 12,400 kW.
[0140] Through the day-ahead-real-time dispatch process, the remaining 10 groups of PV output uncertainty scenarios are calculated, and the key information such as fluctuation and power abandonment index of the two methods when the robustness factor is 0.9 and 1.1 is continuously output. The evaluation indexes of all scenarios are averaged for comparative analysis. See Table 2 for specific information.
[0141] As the robustness factor increases, the photovoltaic output prediction deviation intensifies and the risk of photovoltaic output uncertainty rises. The maximum amount of power abandonment during the operation of the hydro-photovoltaic complementary system of the present invention can be reduced from 3.73 million kW to 32,000 kW, which gives full play to the flexibility of hydropower regulation and can almost completely avoid the risk of power abandonment, providing a more reasonable risk assessment and application guidance strategy for system operation and decision-making.
[0142] Table 2 Effect comparison analysis
[0143]
[0144] Therefore, the present invention fully considers the impact of the flexibility of hydropower regulation capacity and the volatility and uncertainty of photovoltaic output waveform sequence on source-load coordination risk, so that the hydropower complementary system has a stronger risk-bearing capacity, meets the source-load coordination requirements, and avoids power fluctuations and power abandonment risks under different prediction levels.
[0145] Embodiment 2
[0146] A day-ahead optimal dispatching system for hydro-photovoltaic complementarity considering source-load coordination risk, comprising:
[0147] A robust model building unit is used to build a robust model of photovoltaic output uncertainty, analyze photovoltaic output sequence waveform feature segments and simulate photovoltaic output uncertainty scenarios; it includes a waveform feature segment analysis unit and a scenario simulation unit;
[0148] The waveform feature segment analysis unit is used to analyze the waveform feature segments of the photovoltaic output sequence: from the perspective of the peak shape and distribution of the photovoltaic output sequence, the spatiotemporal distribution characteristics of the photovoltaic output sequence waveform segments affected by weather disturbances are analyzed, and they can be divided into three feature types according to the intensity and discreteness of the fluctuations: smooth fluctuations, peak fluctuations, and sawtooth fluctuations. Read the peak and valley positions of the photovoltaic output, record the time period information of adjacent peaks and valleys, and divide the photovoltaic output sequence into multiple initial segments based on the adjacent peaks and valleys as the basic criterion for dividing a single initial segment. Then, the fluctuation intensity and fluctuation duration parameters are used to cluster and identify the initial segments. The characteristic segments of the photovoltaic output are formed through the principle of merging adjacent initial segments of the same type and reorganizing different types of segments;
[0149] The scenario simulation unit is used to simulate the uncertainty scenario of photovoltaic output: based on the traditional cloud model, the qualitative characteristics of the characteristic fragments of the photovoltaic output sequence waveform are coupled, and the prediction deviation information is statistically analyzed from the two dimensions of the waveform and output of the photovoltaic output sequence to obtain the simulation value of the photovoltaic output prediction deviation;
[0150] The dispatch model construction unit is used to construct a day-ahead optimal dispatch model for hydro-photovoltaic complementarity: taking the minimum net load fluctuation as the peak-shaving objective function, and taking the hydropower operation environment constraints, the conservative degree of photovoltaic uncertainty and the waveform-order compensation strategy as constraints, a day-ahead optimal dispatch model for hydro-photovoltaic complementarity considering the source-load coordination risk is constructed;
[0151] The simulation scheduling unit is used to run the day-ahead-real-time scheduling simulation process of the hydro-photovoltaic complementary system: the hydro-photovoltaic complementary day-ahead optimization scheduling model is used to solve the output process of the hydropower and hydro-photovoltaic complementary system, and the compiled power generation plan is applied to the real-time scheduling simulation process to obtain the fluctuation amount and abandoned power index values of the hydro-photovoltaic complementary system under real-time scheduling.
[0152] Embodiment 3
[0153] An electronic device, comprising:
[0154] A memory storing executable program code;
[0155] a processor coupled to the memory;
[0156] The processor calls the executable program code stored in the memory to execute the steps of the above-mentioned method for optimizing the day-ahead scheduling of water-solar complementarity considering source-load coordination risks.
[0157] The device may include: a memory storing executable program code; a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the steps in the method described in the first embodiment.
[0158] The memory may include computer system readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory. The device may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the memory may be used to read and write non-removable, non-volatile magnetic media (commonly referred to as a "hard drive"). A program / utility having a set (at least one) of program modules may be stored in, for example, the memory, such program modules including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include the implementation of a network environment. The program modules typically perform the functions and / or methods in the embodiments described herein.
[0159] The processor executes various functional applications and data processing by running the program stored in the memory, such as implementing the method provided in the first embodiment of the present invention.
[0160] Embodiment 4
[0161] A computer-readable storage medium stores computer instructions, which, when called, are used to execute the steps of the above-mentioned method for optimizing the day-ahead scheduling of water-photovoltaic complementarity considering source-load coordination risks.
[0162] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, a device or a device or used in combination with it.
[0163] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.
[0164] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
[0165] Computer program code for performing the operation of the present invention may be written in one or more programming languages or combinations thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0166] Of course, the computer executable instructions of a storage medium including computer executable instructions provided by an embodiment of the present invention are not limited to the above method operations, and can also execute related operations in the method provided by any embodiment of the present invention.
Claims
1. A method for optimizing the day-ahead scheduling of water-solar complementary power considering the risk of source-load coordination, characterized in that: The following steps are involved: Construct a robust model of photovoltaic output uncertainty, analyze the characteristic segments of the photovoltaic output sequence waveform and simulate the photovoltaic output uncertainty scenario; including: analyze the spatiotemporal distribution characteristics of the photovoltaic output sequence waveform segments affected by weather disturbances, divide the photovoltaic output sequence waveform segments into multiple initial segments, and identify the initial segments to form the characteristic segments of photovoltaic output; on the basis of the traditional cloud model, couple the qualitative characteristics of the characteristic segments of the photovoltaic output sequence waveform, statistically analyze the prediction deviation information from the two dimensions of the photovoltaic output sequence waveform and output, and obtain the simulation value of the photovoltaic output prediction deviation; Construct a day-ahead optimization dispatch model for hydropower-photovoltaic complementarity: Taking the minimum net load fluctuation as the peak-shaving objective function, and taking the hydropower operation environment constraints, the conservative degree of photovoltaic uncertainty and the waveform-order compensation strategy as constraints, a day-ahead optimization dispatch model for hydropower-photovoltaic complementarity considering the source-load coordination risk is constructed; Run the day-ahead-real-time scheduling simulation process of the hydro-solar complementary system: Use the hydro-solar complementary day-ahead optimization scheduling model to solve the output process of the hydropower and hydro-solar complementary system, apply the compiled power generation plan to the real-time scheduling simulation process, and obtain the system fluctuation and power abandonment index values under real-time scheduling.
2. The method for optimizing the day-ahead scheduling of water-solar complementary power considering source-load coordination risk according to claim 1, characterized in that: The specific process of analyzing the waveform characteristics of photovoltaic output sequence is as follows: The photovoltaic output sequence is divided into multiple initial segments, and the photovoltaic output calculation formula of the initial segment n is: in, is the photovoltaic output of the initial segment n; P t s is the photovoltaic output during time period t; are the starting and ending time periods of the initial segment n respectively; According to the fluctuation duration parameter, the fluctuation intensity parameter of the initial segment of photovoltaic output is calculated, and the expression is: in, is the fluctuation intensity, which indicates the sharpness of the peak shape of the initial segment n; is the fluctuation duration, indicating the duration of the initial segment n; is the photovoltaic output of the initial segment n in time period t; is the average PV output of all initial segments in time period t; μ is the PV installed capacity; represents the standard deviation of the PV output of the initial segment n; The characteristic fragment of photovoltaic output is formed by the principle of merging adjacent initial fragments of the same type and reorganizing different types of segments. The photovoltaic output calculation formula of characteristic fragment i is: Among them, P i s is the photovoltaic output of feature segment i; J represents clustering through K-means, and the characteristic parameters are optimized according to the historical photovoltaic output sequence The value of .
3. The method for optimizing the day-ahead scheduling of water-solar complementary power considering source-load coordination risk according to claim 1, characterized in that: The mechanism of simulating the PV output uncertainty scenario is: (1) Input the photovoltaic output value of the characteristic segment i sample into the reverse cloud generator and calculate the qualitative characteristics (E i,j ,F i,j ,H i,j ), E i,j 、F i,j , H i,j are the expected value, fluctuation amplitude, and discrete degree of the photovoltaic output prediction deviation, respectively; i is the waveform segment number, and j is the output interval number; (2) The qualitative characteristics of the photovoltaic output characteristic fragment samples (E i,j ,F i,j ,H i,j ) is input into the forward cloud generator to generate a cloud cluster (ε' u,i,j ,μ i,j ): F i ' ,j =randn(1)×H i,j +F i,j ε' u,i,j = randn(1) × F i ' ,j +E i,j Among them, u is the sample number; ε' u,i,j is the simulated value of photovoltaic output prediction deviation; μ i,j is the probability distribution of the simulated value of the photovoltaic output prediction deviation; F′ i,j is the fluctuation range of the simulated value of the PV output forecast deviation; randn(1) represents a normal distribution with a mean of 0 and a variance of 1; (3) Repeat steps (1) to (2) until a cloud of i×j classification intervals is generated that meets the required number of ε' u,i,j , calculate the simulated photovoltaic output: P t se ={P t s +ε' u,i,j |P t s ∈(i,j)} Among them, P t se , P t s are respectively the simulated photovoltaic output and the predicted photovoltaic output in time period t; according to P t s The classification interval is used to extract the simulated value of the predicted photovoltaic output and generate the simulated photovoltaic output.
4. The method for optimizing the day-ahead scheduling of water-solar complementary power considering source-load coordination risk according to claim 1, characterized in that: The peak load objective function expression is: P t re =L t -P t ,P t =P t h +P t se Among them, F is the net load fluctuation, which indicates the relative fluctuation value between the output of the water-solar hybrid system and the grid load; T is the optimized scheduling period; L t , P t re , P t , P t h They are the grid load, residual load, water-solar hybrid system output, and hydropower output in time period t respectively; It is the average value of the remaining load within the optimized dispatch period.
5. The method for optimizing the day-ahead scheduling of water-solar complementary power considering the source-load coordination risk according to claim 1, characterized in that: The constraints of the hydropower operation environment, the conservative degree of photovoltaic uncertainty and the waveform-order compensation strategy are used to ensure the load tracking capability and control the risk of power abandonment. Specifically: (1) Hydropower operation environment constraints: Among them, V l,t-1 、V l,t are the storage capacities of the l-level hydropower station at the beginning and end of time period t respectively; For l-1 hydropower station in the period The downstream flow rate; Q l,t is the discharge flow of the l-level hydropower station in time period t; q l,t is the flow rate in front of the dam of the l-level hydropower station in time period t; is the water flow lag time between hydropower stations l and l-1, and △t is the calculation step length. l,0 is the water level at the beginning of the l-level hydropower station; Z l,T The water level at the end of the period of the l-level hydropower station; are the control boundaries of the start and end periods of the l-level hydropower station; Z l,t is the water level of the l-level hydropower station in time period t; are the lower and upper limits of the water level of the l-level hydropower station in time period t; Q l,t is the power generation flow of the l-level hydropower station in time period t; are the lower and upper limits of the power generation flow of the l-level hydropower station in time period t respectively; is the output of the l-level hydropower station in time period t; They are the lower and upper limits of the output range of the l-level hydropower station in time period t; (2) Conservative degree of photoelectric uncertainty: F i,j ={α·F i,j |α∈(0.8,1.2)} H i,j ={β·H i,j |β∈(0.8,1.2)} Among them, F i,j represents the fluctuation amplitude characteristic set of photovoltaic output, H i,j It represents the discrete degree feature set of photovoltaic output, α and β are robustness factors; the larger α is, the lower the prediction accuracy and the larger the fluctuation range; the larger β is, the higher the disturbance complexity and the more frequent the fluctuation; (3) Waveform-order compensation strategy: a. Fluctuation control: Constructing the first-order compensation boundary based on the characteristic fragments of the PV output uncertainty scenario in, is the first-order compensation boundary, consisting of the variable set composition; The set of PV output variables simulated for different characteristic segments, P i is the photovoltaic output variable; t 2i is the time variable; To simulate the photovoltaic output collection; b. Power abandonment control: The system output further tracks the grid load through hydropower compensation to form a second-order compensation boundary P t ≤G Among them, G is the maximum capacity of the transmission channel.
6. The method for optimizing the day-ahead scheduling of water-solar complementary power considering source-load coordination risk according to claim 1, characterized in that: The real-time scheduling simulation process is as follows: The predicted photovoltaic output is input into the photovoltaic output uncertainty robust model, and K groups of simulated photovoltaic outputs are generated according to the values of robust factors α and β to form a photovoltaic uncertainty set; The day-ahead optimization dispatch model of hydropower and photovoltaic complementarity is transferred into the simulated photovoltaic output of the kth group as well as the runoff and load information. The dynamic programming algorithm DP is used to solve the output process of the hydropower and hydropower complementary system and prepare the power generation plan. The actual photovoltaic output is input into the real-time dispatching process, and the hydropower output is re-corrected by determining the water supply based on electricity. The water storage and supply sequence of hydropower stations at all levels is determined by the discriminant coefficient method to offset and balance the photovoltaic output deviation value, and transmit electricity according to the power generation plan within the regulation capacity and controllable range of the cascade hydropower stations; Extract the corrected output of the hydro-photovoltaic complementary system in real-time scheduling, where the corrected output of the hydro-photovoltaic complementary system is the sum of the photovoltaic output and the corrected hydropower output. The output of the hydro-photovoltaic complementary system that exceeds the power generation plan will result in abandoned power. The simulated photovoltaic output of group K was fully dispatched, and the output was the reservoir capacity, water level and fluctuation and abandoned power index values of the water-photovoltaic complementary system under real-time dispatching in all scenarios.
7. The method for optimizing the day-ahead scheduling of water-solar complementary power considering the source-load coordination risk according to claim 6, characterized in that: The calculation formula of the abandoned power index is: Among them, C is the amount of power abandoned; P t h' Corrected hydropower output for time period t.
8. A day-ahead optimization scheduling system for water-solar complementary power generation considering source-load synergy risk, characterized in that: include: The robust model building unit is used to build a robust model of photovoltaic output uncertainty, analyze the characteristic segments of the photovoltaic output sequence waveform and simulate the photovoltaic output uncertainty scenario; including: analyzing the spatiotemporal distribution characteristics of the photovoltaic output sequence waveform segments affected by weather disturbances, dividing the photovoltaic output sequence waveform segments into multiple initial segments, and identifying the initial segments to form the characteristic segments of photovoltaic output; based on the traditional cloud model, coupling the qualitative characteristics of the characteristic segments of the photovoltaic output sequence waveform, statistically predicting the deviation information from the two dimensions of the photovoltaic output sequence waveform and output, and obtaining the simulated value of the photovoltaic output prediction deviation; The dispatch model construction unit is used to construct a day-ahead optimal dispatch model for hydro-photovoltaic complementarity: taking the minimum net load fluctuation as the peak-shaving objective function, and taking the hydropower operation environment constraints, the conservative degree of photovoltaic uncertainty and the waveform-order compensation strategy as constraints, a day-ahead optimal dispatch model for hydro-photovoltaic complementarity considering the source-load coordination risk is constructed; The simulation scheduling unit is used to run the day-ahead-real-time scheduling simulation process of the hydro-photovoltaic complementary system: the hydro-photovoltaic complementary day-ahead optimization scheduling model is used to solve the output process of the hydropower and hydro-photovoltaic complementary system, and the compiled power generation plan is applied to the real-time scheduling simulation process to obtain the system fluctuation and abandoned power index values under real-time scheduling.
9. An electronic device, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the steps of a method for day-ahead optimization scheduling of water-photovoltaic complementarity considering source-load coordination risks as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, which, when called, are used to execute the steps of a method for optimizing the day-ahead scheduling of water-photovoltaic complementarity considering source-load coordination risks as described in any one of claims 1 to 7.
Citation Information
Cited By
Grid-connected power fluctuation active stabilization control method and system for multi-source new energy
CN120377317A
Active control method and system for smoothing grid-connected power fluctuations for multi-source renewable energy
CN120377317B
Wind-light-water storage double-layer robust optimization scheduling strategy construction method and system
CN121216626A
A method and system for constructing a double-layer robust optimization scheduling strategy of wind-solar-water storage
CN121216626B
