Wind and light storage capacity optimal configuration method considering power shortage probability and power transmission deviation degree

Abstract

The invention relates to the technical field of new energy power transmission, and particularly discloses a wind and light storage capacity optimal configuration method considering power shortage probability and power transmission deviation, which comprises the following steps: S1, clustering a receiving end power grid load curve by using a K-means method to obtain a typical load scene; s2, calculating a direct current power transmission plan for each typical load scene based on a linear operation mode; s3, constructing a new energy base wind and light storage capacity optimal configuration model considering the power shortage probability and the power transmission deviation degree; and solving to obtain an optimal configuration scheme. The method has the advantages that the correlation between installed capacities and output of different power supplies and direct-current power transmission is quantified, and effective theoretical support is provided for initial planning of a new energy large base.

Description

Technical Field

[0001] The present invention relates to the field of renewable energy power transmission technology, and in particular to a method for optimizing wind, solar and storage capacity configuration by taking into account power shortage probability and power transmission deviation. Background Art

[0002] The safety and stability of new energy bases depend on the coordination between various power sources and energy storage systems. Ensuring a reasonable ratio between wind and solar energy and energy storage capacity, and utilizing the mutual complementarity of different energy sources, is crucial to ensuring the economy and reliability of the system. Therefore, the reasonable allocation of various power sources and energy storage capacity is an important direction for research in the planning stage of new energy bases. With the development of energy Internet and the increase in the proportion of new energy, promoting the construction of new energy transmission bases has a far-reaching impact on promoting the large-scale utilization of clean energy and the development of a green and low-carbon economy. New energy transmission bases can help my country solve the current problem of uneven distribution of new energy to a large extent. For example, Qinghai Province, as one of the national new energy transmission bases, has built and connected to the grid photovoltaic and wind power installed capacity of more than 10 million kilowatts. In particular, many large-scale wind power and photovoltaic hybrid power generation projects in Haixi Prefecture, Qinghai, have become an important part of new energy transmission. Practice has shown that new energy transmission bases have actual results and potential in increasing the proportion of energy structure and optimizing energy layout.

[0003] In the research of new energy transmission bases, the existing technology of evaluation of power transmission capacity of new energy transmission system based on converter grid connection deeply explores the power transmission capacity of clean energy transmission system through converter grid connection, and emphasizes its key role in the stable operation of power grid. The article comprehensively considers various constraints, determines the maximum transmission capacity of new energy, and verifies through example analysis that the synergy of reactive power compensation and dynamic thermal capacity can enhance the transmission capacity. The existing technology of optimizing the configuration strategy of power source combination supporting power transmission of new energy base analyzes the importance of new energy base development and power transmission for achieving clean and low-carbon energy transformation. Based on wide-area long-term meteorological data, the article uses time series production simulation and power expansion optimization model to study the optimization combination strategy of regulating power source and new energy installed capacity. The existing technology of optimizing the medium- and long-term operation mode of large-scale new energy base through ultra-high voltage direct current transmission system introduces a method for optimizing the medium- and long-term operation mode of large-scale new energy base through ultra-high voltage direct current transmission system. By establishing the aggregation operation model of different power station types and the ultra-high voltage direct current non-integer variable operation model, combined with flexible regulation performance and load demand, the maximum transmission and consumption of new energy is achieved. Prior artModel and application of renewable energyaccommodation capacity calculation considering utilization level of inter-provincial tie-line, based on sequential production simulation model, a new accommodation capacity model considering the utilization rate of inter-provincial tie-line is proposed. Four schemes are designed for comparative analysis, and the renewable energy accommodation capacity of China's "Three Norths" region in 2020 is calculated. Prior artMassive multi-source joint outbound and benefitdistribution model based on cooperative gameIn view of the challenge that China's new energy is widely distributed and far from the load center, a dynamic coordinated dispatch model is proposed to combine wind power, photovoltaic and thermal power to optimize the profit of the energy complementary transmission system.

[0004] At present, there are studies on clean energy transmission capacity, new energy base development and power transmission, UHV DC transmission system, renewable energy capacity and energy complementary transmission system. However, few articles have studied the optimal configuration of wind, solar and storage capacity in new energy transmission bases. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method for optimizing the configuration of wind, solar and storage capacity taking into account the probability of power shortage and power transmission deviation.

[0006] The object of the present invention is achieved by the following technical solution: a method for optimizing the configuration of wind, solar and storage capacity considering the probability of power shortage and power transmission deviation, comprising the following steps:

[0007] Step S1, clustering the load curve of the receiving power grid using the K-means method to obtain a typical load scenario;

[0008] Step S2, calculating a DC transmission plan for each typical load scenario based on a linear operation mode;

[0009] Step S3: construct a wind, solar and storage capacity optimization configuration model for a new energy base that takes into account the probability of power shortage and power transmission deviation; and solve to obtain the optimal configuration solution.

[0010] Specifically, the specific steps of step S1 include:

[0011] S11, Initialization: First determine the number of categories K to be divided, and randomly select K data points as the initial clustering centers m j (I), I is the number of iterative calculations, initially 1, j = 1, 2, ..., K;

[0012] S12. Assign data points: For each data point x i , calculate its distance d(x i ,m j (I)) and assigns it to the nearest cluster center m j (I) represents the category S j middle:

[0013]

[0014] In the formula, i=(x i1 ,x i2 ,...x iM ); j = (x j1 ,x j2 ,...x jM );

[0015] S13, update cluster center: calculate the average value of the data points in each cluster, and use the average value as the new cluster center.

[0016]

[0017] Where N j represents the number of data objects in the jth class; S j represents the set of all data points belonging to the jth class.

[0018] S14, repeat iteration: repeat step S12 and step S13 until the change of the cluster center is less than the threshold;

[0019] S15. Output clustering results.

[0020] Specifically, the constraints of the DC transmission plan in step S2 are:

[0021]

[0022] Where: T is the number of moments in a typical load scenario, which is 24; u t is the power adjustment state, a 0-1 variable, u t =1 means adjusting the DC channel transmission power during time period t, u t =0 means that the DC channel transmission power is not adjusted during period t; P t HVDC is the planned power of DC transmission in period t; M is the coefficient for implementing the large M method; I is the number of load curves corresponding to the current scenario category; is the value of the i-th load curve in time period t, △t is the time interval of the unit time period, P 1 HVDC and are the planned DC transmission power at the first and Tth moments respectively.

[0023] Specifically, the objective function of calculating the DC transmission plan in step S3 is as follows:

[0024]

[0025] In the formula, T is the number of moments of a typical load scenario, which is 24, and I is the number of load curves corresponding to the current scenario category. is the value of the i-th load curve in time period t.

[0026] Specifically, the wind, solar, and storage capacity optimization configuration model for the new energy base in step S3 takes the minimum investment cost of the three types of power sources, wind, solar, and storage, as the objective function:

[0027] minF(x)=min(C w W+C P P V +C S S+C E E)

[0028] Where: C w is the investment cost of wind power generation; W is the planned capacity of wind power generation; C P is the investment cost of photovoltaic power generation; P V Planning capacity for photovoltaic power generation; C Sis the investment cost of the energy storage power part; S is the planned capacity of the energy storage power part; C E is the investment cost of the energy storage part; E is the planned capacity of the energy storage part.

[0029] Specifically, the constraints of the wind, solar and storage capacity optimization configuration model of the new energy base include:

[0030] Thermal power output constraints:

[0031]

[0032] Where: is the operating status of the thermal power unit in the i-th scenario at time t, 1 represents operation and 0 represents shutdown; P i g,min is the minimum technical output of the i-th unit; is the output of the i-th scenario unit to the energy storage at time t; is the output of the i-th scenario unit to the DC channel at time t; P i g,max The maximum technical output of the i-th scenario unit; T i g,on is the minimum start-up time of the thermal power unit in the i-th scenario; T i g,off is the minimum shutdown time of the thermal power unit in the i-th scenario; r i g,up 、r i g,d are the up and down ramp rates of the thermal power unit in the i-th scenario respectively;

[0033] New energy output constraints:

[0034]

[0035] Where: They are the output of wind and solar power to energy storage in the i-th scenario at time t; are the output of wind and solar power to the DC channel in the i-th scenario at time t; is the maximum output of the scenery of the i-th scene at time t;

[0036] Energy storage operation constraints:

[0037]

[0038] Where: They are the charging and discharging status of the i-th scene at time t, 1 represents operation and 0 represents shutdown; is the energy storage output to the DC channel in the i-th scenario at time t; S is the planned capacity of the power part of the electric energy storage system; is the charging and discharging efficiency of the energy storage system in the i-th scenario; is the current capacity of the energy storage system in the i-th scenario at time t, E is the planned capacity of the energy part of the energy storage system; γ is the minimum limit of the energy part of the energy storage system, which is 0.2;

[0039] Constraints on the rate of abandoned electricity of new energy sources:

[0040]

[0041] Where: μ is the upper limit of the new energy power abandonment rate, which is 0.1;

[0042] Constraints on the proportion of new energy output:

[0043] Set the proportion of renewable energy output to be no less than 50%;

[0044]

[0045] Grid support power constraints:

[0046]

[0047] Where: is the grid support power in time period t in the i-th scenario; ξ is the grid support coefficient, which is defined as the ratio of the grid support power upper limit to the DC transmission planned power; H is the number of calculation cycles of the grid support balance constraint;

[0048] Power shortage probability constraint:

[0049]

[0050] Where: (0) is the N-dimensional reference probability distribution vector, ρ (0) =1 / N, indicating that each scene appears with the same probability, z n =0 indicates full power transmission scenario, z n =1 indicates an incomplete power transmission scenario, i.e., in a scenario, the power transmission power at any time does not meet the DC transmission plan power; α is the power shortage probability; z represents the set of all power transmission scenarios.

[0051] Specifically, it also includes step S4, using actual data to perform case analysis using the proximal strategy optimization algorithm to verify the rationality of the wind, solar and storage capacity optimization configuration model of the new energy base;

[0052] The strategy function of the proximal strategy optimization is:

[0053] π θ (a|s)

[0054] Among them, π θ is the policy network, θ represents the policy parameter, s is the state, and a is the action;

[0055] Use advantage value to measure the relative superiority of an action:

[0056]

[0057] Among them, A t is the advantage value; γ is the discount factor, which is used to weigh the importance of current rewards and future rewards; λ is the smoothing parameter of GAE, which is used to adjust the smoothness of the time difference error; l is the cumulative step size; a t is the immediate reward between states; V θ is the value network, δ t is the time difference at time t;

[0058] The objective function of proximal strategy optimization is:

[0059]

[0060] in, is the expected value of the time step; r t is the action selection probability ratio of the new and old strategies at time step t, represents the ratio of the new and old strategies, ∈ is a small hyperparameter used to control the shear range; π θ is the probability of taking an action under the current state; is the probability of taking an action in the state under the previous policy.

[0061] Specifically, the steps of the proximal strategy optimization algorithm are:

[0062] Step S41, environment modeling: determining the state space, action space and reward mechanism;

[0063] Step S42, initialization: initialization strategy network π θ and value network V θ ;

[0064] Step S43, data collection: collect sample trajectories through the interaction between the current strategy and the environment;

[0065] Step S44, advantage estimation: use the GAE method to calculate the advantage value A of each action t ;

[0066] Step S45, strategy update: calculate the objective function L CLIP (θ), and use the gradient ascent method to update the policy parameters θ:

[0067]

[0068] Among them, α is the learning rate; Indicates that the objective function is gradient θ;

[0069] Step S46, training cycle: repeating steps S43, S44 and S45 until convergence;

[0070] Step S47, model evaluation: evaluating the model performance by calculating indicators such as power shortage probability, power abandonment rate and investment cost;

[0071] Step S48, final strategy export: Once the model converges, export the learned strategy for actual decision optimization.

[0072] The present invention has the following advantages:

[0073] 1. The present invention combines the characteristics of UHV DC transmission and considers the probability of power shortage to propose an optimal configuration model for wind, solar and storage capacity in new energy bases, and quantifies the relationship between the installed capacity and output of different power sources and DC transmission;

[0074] 2. The present invention obtains the influence of confidence probability, wind-solar ratio, investment cost and other aspects on wind-solar storage capacity through actual example analysis, which can provide effective theoretical support for the initial planning of large new energy bases. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 This is a flow chart of the method for optimizing the configuration of wind, solar and storage capacity of the present invention;

[0076] Figure 2 It is a flow chart of the proximal strategy optimization algorithm of the present invention;

[0077] Figure 3 is the normalized wind power generation curve of the present invention;

[0078] Figure 4 is the normalized photovoltaic power generation curve of the present invention;

[0079] Figure 5 is the DC transmission curve of the present invention;

[0080] Figure 6 is the error square sum curve of the present invention;

[0081] Figure 7 is a typical load curve of the present invention;

[0082] Figure 8 This is a typical DC transmission planning curve of the present invention. DETAILED DESCRIPTION

[0083] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the embodiments described are only part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0084] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0085] It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0086] The present invention is further described below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following description. Figures 1 to 8 As shown, the method for optimizing the configuration of wind, solar and storage capacity considering the probability of power shortage and power transmission deviation includes the following steps: Step S1, clustering the load curve of the receiving power grid using the K-means method to obtain a typical load scenario;

[0087] Step S2, calculating a DC transmission plan for each typical load scenario based on a linear operation mode;

[0088] Step S3: construct a wind, solar and storage capacity optimization configuration model for a new energy base that takes into account the probability of power shortage and power transmission deviation; and solve to obtain the optimal configuration solution.

[0089] Furthermore, the elbow point method is first used to determine the number of categories K. The elbow point method is used to determine the number of categories K. Its basic idea is to evaluate the clustering effect by calculating the sum of squares of clustering errors (SSE) under different numbers of categories, and select the K value that makes the SSE decline trend significantly slower as the optimal number of categories. The sum of squares of clustering errors can be obtained by the following formula:

[0090]

[0091] Then, the K-means method is used to cluster the load curve of the receiving power grid. The specific steps of the clustering process include:

[0092] S11, Initialization: First determine the number of categories K to be divided, and randomly select K data points as the initial clustering centers m j (I), I is the number of iterative calculations, initially 1, j = 1, 2, ..., K;

[0093] S12. Assign data points: For each data point x i , calculate its distance d(x i ,m j (I)) and assigns it to the nearest cluster center m j (I) represents the category S j middle:

[0094]

[0095] In the formula, i=(x i1 ,x i2 ,...x iM ); j = (x j1 ,x j2 ,...x jM ); This step is mainly to divide the data points into the categories represented by the most similar cluster centers;

[0096] S13, update cluster center: calculate the average value of the data points in each cluster, and use the average value as the new cluster center.

[0097]

[0098] Where N j represents the number of data objects in the jth class; S j Represents the set of all data points belonging to the jth class; this step improves the accuracy of clustering by updating the cluster centers.

[0099] S14, repeat iteration: repeat step S12 and step S13 until the change of the cluster center is less than the threshold value; in each iteration, the allocation of data points and the update of cluster centers will continue until the stop condition is reached;

[0100] S15, outputting the clustering result; the final clustering result is determined based on the updated cluster centers and the distribution of data points, each data point is assigned to a category, and the cluster centers represent the characteristics of each category;

[0101] Furthermore, the "straight-line" operation mode of the DC transmission plan refers to an operation mode in which the transmission of current is carried out along a straight path in the DC transmission system. The "straight-line" operation mode is currently the main mode adopted by my country's ultra-high voltage DC transmission. In essence, it is an efficient and stable DC power transmission method, which uses the technical advantages of DC transmission to directly transmit electric energy from resource-rich areas to areas with high power demand. This mode is particularly suitable for power transmission across long distances and regions, which can minimize energy loss on the way and improve the efficiency and stability of power transmission.

[0102] In this mode, in order to ensure the safety and stability of the DC transmission channel, its transmission power remains fixed most of the time. Only in rare cases will the power of the transmission channel be switched and adjusted through automated equipment, and usually the power adjustment will not exceed twice a day. Therefore, for each typical load scenario, the power set in the DC transmission plan should comply with the following constraints:

[0103]

[0104] Where: T is the number of moments in a typical load scenario, which is 24; u t is the power adjustment state, a 0-1 variable, u t =1 means adjusting the DC channel transmission power during time period t, u t =0 means that the DC channel transmission power is not adjusted during period t; P t HVDC is the planned power of DC transmission in period t; M is the coefficient for implementing the large M method; I is the number of load curves corresponding to the current scenario category; is the value of the i-th load curve in time period t, △t is the time interval of the unit time period, P 1 HVDC and are the planned DC transmission powers at the first and Tth moments respectively.

[0105] Furthermore, the objective function of calculating the DC transmission plan in step S3 is as follows:

[0106]

[0107] In the formula, T is the number of moments of a typical load scenario, which is 24, and I is the number of load curves corresponding to the current scenario category. is the value of the i-th load curve in time period t.

[0108] Furthermore, the optimal configuration model for wind, solar and storage capacity of the new energy base in step S3 takes the minimum investment cost of the three types of power sources of wind, solar and storage as the objective function:

[0109] minF(x)=min(C w W+C P P V +C S S+C E E)

[0110] Where: C w is the investment cost of wind power generation; W is the planned capacity of wind power generation; C P is the investment cost of photovoltaic power generation; P V Planning capacity for photovoltaic power generation; C S is the investment cost of the energy storage power part; S is the planned capacity of the energy storage power part; C E is the investment cost of the energy storage part; E is the planned capacity of the energy storage part.

[0111] Furthermore, the constraints of the wind, solar and storage capacity optimization configuration model of the new energy base include:

[0112] Thermal power output constraints:

[0113]

[0114] Where: is the operating status of the thermal power unit in the i-th scenario at time t, 1 represents operation and 0 represents shutdown; P i g,min is the minimum technical output of the i-th unit; is the output of the i-th scenario unit to the energy storage at time t; is the output of the i-th scenario unit to the DC channel at time t; P i g,max The maximum technical output of the i-th scenario unit; T i g,on is the minimum start-up time of the thermal power unit in the i-th scenario; T i g,off is the minimum shutdown time of the thermal power unit in the i-th scenario; r i g,up 、r i g,d are the up and down ramp rates of the thermal power unit in the i-th scenario respectively;

[0115] New energy output constraints:

[0116]

[0117] Where: They are the output of wind and solar power to energy storage in the i-th scenario at time t; are the output of wind and solar power to the DC channel in the i-th scenario at time t; is the maximum output of the scenery of the i-th scene at time t;

[0118] Energy storage operation constraints:

[0119]

[0120] Where: They are the charging and discharging status of the i-th scene at time t, 1 represents operation and 0 represents shutdown; is the energy storage output to the DC channel in the i-th scenario at time t; S is the planned capacity of the power part of the electric energy storage system; is the charging and discharging efficiency of the energy storage system in the i-th scenario; is the current capacity of the energy storage system in the i-th scenario at time t, E is the planned capacity of the energy part of the energy storage system; γ is the minimum limit of the energy part of the energy storage system, which is 0.2;

[0121] Constraints on the rate of abandoned electricity of new energy sources:

[0122]

[0123] Where: μ is the upper limit of the new energy power abandonment rate, which is 0.1;

[0124] Constraints on the proportion of new energy output:

[0125] Set the proportion of renewable energy output to be no less than 50%;

[0126]

[0127] Grid support power constraints:

[0128]

[0129] Where: is the grid support power in time period t in the i-th scenario; ξ is the grid support coefficient, which is defined as the ratio of the grid support power upper limit to the DC transmission plan power; H is the number of calculation cycle periods for the grid support balance constraint, which is 168 in the present invention, i.e. one week is the calculation cycle;

[0130] Power shortage probability constraint:

[0131]

[0132] Where: (0) is the N-dimensional reference probability distribution vector, ρ (0) =1 / N, indicating that each scene appears with the same probability, z n =0 indicates full power transmission scenario, z n =1 indicates an incomplete power transmission scenario, that is, in a scenario, the power transmission power at any time does not meet the DC transmission plan power; α is the power shortage probability, and z represents the set of all power transmission scenarios.

[0133] Furthermore, it also includes step S4, using the proximal policy optimization algorithm to perform case analysis through actual data to verify the rationality of the wind, solar and storage capacity optimization configuration model of the new energy base; Proximal Policy Optimization (PPO) is an advanced reinforcement learning algorithm designed for processing large-scale and continuous action spaces, and is particularly suitable for the 8760-hour long-term optimization configuration problem in this patent. It effectively balances exploration and utilization by simultaneously training the policy network (Actor) and the value network (Critic), ensuring the stability of learning in a dynamic environment. PPO limits the policy update range through a shear mechanism to prevent over-adjustment and avoid the risk of policy collapse;

[0134] The strategy function of the proximal strategy optimization is:

[0135] π θ (a|s)

[0136] Among them, π θ is the policy network, θ represents the policy parameter, s is the state, and a is the action;

[0137] Use advantage value to measure the relative superiority of an action:

[0138]

[0139] Among them, A t is the advantage value; γ is the discount factor, which is used to weigh the importance of current rewards and future rewards; λ is the smoothing parameter of GAE, which is used to adjust the smoothness of the time difference error; l is the cumulative step size; a t is the immediate reward between states; V θ is the value network, δ t is the time difference at time t;

[0140] The objective function of proximal strategy optimization is:

[0141]

[0142] in, is the expected value of the time step; rt is the action selection probability ratio of the new and old strategies at time step t, represents the ratio of the new and old strategies, ∈ is a small hyperparameter used to control the shear range; π θ is the probability of taking an action under the current state; is the probability of taking an action in the state under the previous policy.

[0143] Furthermore, the steps of the proximal strategy optimization algorithm are:

[0144] Step S41, environment modeling: determining the state space, action space and reward mechanism;

[0145] Step S42, initialization: initialization strategy network π θ and value network V θ ;

[0146] Step S43, data collection: collect sample trajectories through the interaction between the current strategy and the environment;

[0147] Step S44, advantage estimation: use the GAE method to calculate the advantage value A of each action t ;

[0148] Step S45, strategy update: calculate the objective function L CLIP (θ), and use the gradient ascent method to update the policy parameters θ:

[0149]

[0150] Among them, α is the learning rate; Indicates that the objective function is gradient θ;

[0151] Step S46, training cycle: repeating steps S43, S44 and S45 until convergence;

[0152] Step S47, model evaluation: evaluating the model performance by calculating indicators such as power shortage probability, power abandonment rate and investment cost;

[0153] Step S48, final strategy export: Once the model converges, export the learned strategy for actual decision optimization.

[0154] An example analysis is conducted using actual data from a new energy base in Northwest China. The installed capacity of each power source, DC transmission capacity, investment cost of each power source and other related parameters of the base are shown in the following table.

[0155] Table 1 Example parameters

[0156]

[0157] The typical output curves of wind power generation, photovoltaic power generation and DC transmission curves are selected from the base, such as Figures 3 to 5 At the same time, simulation scenarios are constructed based on historical data. In order to reflect the seasonal fluctuations of wind and solar power generation, an 8760-hour continuous operation model is used to ensure the continuity between each scenario, reflecting the coherence of the time series.

[0158] (1) Calculation of DC transmission plan

[0159] First, the number of categories of the DC transmission curve is obtained by the elbow point method, and the maximum value of K is taken as 6 to calculate the error sum of squares under different K values, as shown in Figure 6 It can be seen that when the K value exceeds 4, the square value of the error does not change significantly when the K value is increased, so the elbow point is selected as 4.

[0160] Set the number of categories to 4, and use the K-means calculation method to cluster the DC transmission curves to obtain four typical curves as shown below;

[0161] Combined with the obtained typical load curve, the optimization operation problem is constructed in the "linear" operation mode to calculate the corresponding DC transmission plan curve.

[0162] (2) Planning wind, solar and storage capacity considering power shortage probability and power transmission deviation

[0163] In order to analyze the impact of different power shortage probabilities on the planned wind, solar and storage capacity, the confidence probabilities are taken as 1 and 0.95 respectively. At the same time, since the large M method is used to implement the power shortage probability constraint, which is different from the M of the general large M method, the M value here represents the allowable power transmission deviation in the incomplete power transmission scenario. Therefore, considering the impact of different M values ​​on the planned capacity, M is taken as 10 and 100 respectively. The optimization results are shown in the following table.

[0164] Table 2 Optimization calculation results considering power outage probability

[0165]

[0166] It can be seen that when the power shortage probability is 0.95, the total system cost is reduced compared with full power transmission, and when M = 100, the total system cost decreases significantly. M = 100 is equivalent to the maximum allowable power transmission deviation of 100MW in the incomplete power transmission scenario, which accounts for 1% of the overall DC power transmission capacity of 8000MW. This is an acceptable range. Therefore, the subsequent optimization calculation takes the power shortage probability as 0.05 and M as 100.

[0167] In order to analyze the impact of the planned capacity ratio of wind and solar power on the planned wind, solar and storage capacity, the planned capacity ratio of wind and solar power is set to 2:1, 1:1 and 1:2 respectively. The results are shown in the following table. When the planned capacity ratio of photovoltaic power generation continues to increase, the system cost continues to rise. This is because photovoltaic power generation is more volatile and requires a larger capacity of energy storage to support DC power transmission.

[0168] Table 3 Optimization calculation results considering the planned wind and solar capacity ratio

[0169]

[0170] In order to analyze the impact of the support coefficient on the planned wind, solar and storage capacity, the support coefficient is set to 2.5%, 5% and 10% respectively. The results are shown in the following table. When the support coefficient continues to increase, the system cost continues to decrease, which shows that the support power transmission of other lines can largely solve the uncertainty problem of wind and solar power generation. At the same time, when the support coefficient is 10%, the planned wind and solar capacity ratio approaches 1:1, which means that in the absence of grid support, the uncertainty of photovoltaic power generation is largely weakened by wind power generation, proving that the complementarity of wind and solar power generation is obvious in new energy bases.

[0171] Table 4 Optimization results considering support coefficient

[0172]

[0173]

[0174] The above is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any technician familiar with the art can make many possible changes and modifications to the technical solution of the present invention by using the above-mentioned technical content without departing from the scope of the technical solution of the present invention, or modify it into an equivalent embodiment of equivalent changes. Therefore, any changes, modifications, equivalent changes and modifications made to the above embodiments based on the technology of the present invention without departing from the content of the technical solution of the present invention belong to the protection scope of the present technical solution.

Claims

1. A method for optimizing the configuration of wind, solar and storage capacity considering the probability of power shortage and power transmission deviation, characterized by: The following steps are involved: Step S1, clustering the load curve of the receiving power grid using the K-means method to obtain a typical load scenario; Step S2, calculating a DC transmission plan for each typical load scenario based on a linear operation mode; Step S3: construct a wind, solar and storage capacity optimization configuration model for a new energy base that takes into account the probability of power shortage and power transmission deviation; and solve to obtain the optimal configuration solution.

2. The method for optimizing wind, solar and storage capacity configuration considering power shortage probability and power transmission deviation according to claim 1 is characterized by: The specific steps of step S1 include: S11, Initialization: First determine the number of categories K to be divided, and randomly select K data points as the initial clustering centers m j (I), I is the number of iterative calculations, initially 1, j = 1, 2, ..., K; S12. Assign data points: For each data point x i , calculate its distance d(x i ,m j (I)) and assigns it to the nearest cluster center m j (I) represents the category S j middle: In the formula, i=(x i1 ,x i2 ,...x iM ); j = (x j1 ,x j2 ,...x jM ); S13, update cluster center: calculate the average value of the data points in each cluster, and use the average value as the new cluster center. Where N j represents the number of data objects in the jth class; S j represents the set of all data points belonging to the jth class; S14, repeat iteration: repeat step S12 and step S13 until the change of the cluster center is less than the threshold; S15. Output clustering results.

3. The method for optimizing wind, solar and storage capacity configuration considering power shortage probability and power transmission deviation according to claim 1 is characterized in that: The constraints of the DC transmission plan in step S2 are: Where: T is the number of moments in a typical load scenario, which is 24; u t is the power adjustment state, a 0-1 variable, u t =1 means adjusting the DC channel transmission power during time period t, u t =0 means that the DC channel transmission power is not adjusted during period t; P t HVDC is the planned power of DC transmission in period t; M is the coefficient for implementing the large M method; I is the number of load curves corresponding to the current scenario category; is the value of the i-th load curve in time period t, △t is the time interval of the unit time period, P1 HVDC and are the planned DC transmission powers at the first and Tth moments respectively.

4. The method for optimizing wind, solar and storage capacity configuration considering power shortage probability and power transmission deviation according to claim 3 is characterized by: The objective function of calculating the DC transmission plan in step S3 is as follows: In the formula, T is the number of moments of a typical load scenario, which is 24, and I is the number of load curves corresponding to the current scenario category. is the value of the i-th load curve in time period t.

5. The method for optimizing wind, solar and storage capacity configuration considering power shortage probability and power transmission deviation according to claim 1 is characterized in that: The wind, solar, and storage capacity optimization configuration model of the new energy base in step S3 takes the minimum investment cost of the three types of power sources, wind, solar, and storage, as the objective function: minF(x)=min(C w W+C P P V +C S S+C E E) Where: C w is the investment cost of wind power generation; W is the planned capacity of wind power generation; C P is the investment cost of photovoltaic power generation; P V Planning capacity for photovoltaic power generation; C S is the investment cost of the energy storage power part; S is the planned capacity of the energy storage power part; C E is the investment cost of the energy storage part; E is the planned capacity of the energy storage part.

6. The method for optimizing wind, solar and storage capacity configuration considering power shortage probability and power transmission deviation according to claim 5 is characterized by: The constraints of the wind, solar and storage capacity optimization configuration model for the new energy base include: Thermal power output constraints: Where: is the operating status of the thermal power unit in the i-th scenario at time t, 1 represents operation and 0 represents shutdown; P i g,min is the minimum technical output of the i-th unit; is the output of the i-th scenario unit to the energy storage at time t; is the output of the i-th scenario unit to the DC channel at time t; P i g,max The maximum technical output of the i-th scenario unit; T i g,on is the minimum start-up time of the thermal power unit in the i-th scenario; T i g,off is the minimum shutdown time of the thermal power unit in the i-th scenario; r i g,up 、r i g,d are the up and down ramp rates of the thermal power unit in the i-th scenario respectively; New energy output constraints: Where: They are the output of wind and solar power to energy storage in the i-th scenario at time t; are the output of wind and solar power to the DC channel in the i-th scenario at time t; is the maximum output of the scenery of the i-th scene at time t; Energy storage operation constraints: Where: They are the charging and discharging status of the i-th scene at time t, 1 represents operation and 0 represents shutdown; is the energy storage output to the DC channel in the i-th scenario at time t; S is the planned capacity of the power part of the electric energy storage system; is the charging and discharging efficiency of the energy storage system in the i-th scenario; is the current capacity of the energy storage system in the i-th scenario at time t, E is the planned capacity of the energy part of the energy storage system; γ is the minimum limit of the energy part of the energy storage system, which is 0.2; Constraints on the rate of abandoned electricity of new energy sources: Where: μ is the upper limit of the new energy power abandonment rate, which is 0.1; Constraints on the proportion of new energy output: Set the proportion of renewable energy output to be no less than 50%; Grid support power constraints: Where: is the grid support power in time period t in the i-th scenario; ξ is the grid support coefficient, which is defined as the ratio of the grid support power upper limit to the DC transmission planned power; H is the number of calculation cycles of the grid support balance constraint; Power shortage probability constraint: Where: (0) is the N-dimensional reference probability distribution vector, ρ (0) =1 / N, indicating that each scene appears with the same probability, z n =0 indicates full power transmission scenario, z n =1 indicates an incomplete power transmission scenario, that is, in a scenario, the power transmission power at any time does not meet the DC transmission plan power; z represents the set of all power transmission scenarios, and α is the power shortage probability.

7. The method for optimizing wind, solar and storage capacity configuration considering power shortage probability and power transmission deviation according to claim 1 is characterized in that: The step S4 is also included, using actual data to perform case analysis using the proximal strategy optimization algorithm to verify the rationality of the wind, solar and storage capacity optimization configuration model of the new energy base; The strategy function of the proximal strategy optimization is: p θ (a|s) Among them, π θ is the policy network, θ represents the policy parameter, s is the state, and a is the action; Use advantage value to measure the relative superiority of an action: Among them, A t is the advantage value; γ is the discount factor, which is used to weigh the importance of current rewards and future rewards; λ is the smoothing parameter of GAE, which is used to adjust the smoothness of the time difference error; l is the cumulative step size; a t is the immediate reward between states; V θ is the value network, δ t is the time difference at time t; The objective function of proximal strategy optimization is: in, is the expected value of the time step; r t is the action selection probability ratio of the new and old strategies at time step t, represents the ratio of the new and old strategies, ∈ is a small hyperparameter used to control the shear range; π θ is the probability of taking an action under the current state; is the probability of taking an action in the state under the previous policy.

8. The method for optimizing wind, solar and storage capacity configuration considering power shortage probability and power transmission deviation according to claim 7 is characterized in that: The steps of the proximal strategy optimization algorithm are: Step S41, environment modeling: determining the state space, action space and reward mechanism; Step S42, initialization: initialization strategy network π θ and value network V θ ; Step S43, data collection: collect sample trajectories through the interaction between the current strategy and the environment; Step S44, advantage estimation: use the GAE method to calculate the advantage value A of each action t ; Step S45, strategy update: calculate the objective function L CLIP (θ), and use the gradient ascent method to update the policy parameters θ: Among them, α is the learning rate; Indicates that the objective function is gradient θ; Step S46, training cycle: repeating steps S43, S44 and S45 until convergence; Step S47, model evaluation: evaluating the model performance by calculating indicators such as power shortage probability, power abandonment rate and investment cost; Step S48, final strategy export: Once the model converges, export the learned strategy for actual decision optimization.

Images