An optimization method for coordinated peak shaving of a hydropower cascade and a photovoltaic power station

Abstract

An optimization method for coordinated peak shaving of a hydropower group cascade and a photovoltaic power station, which relates to the field of water-light peak shaving involving small hydropower station clusters. It includes the analysis of the power curve characteristics of distributed photovoltaic power generation and the analysis of the power curve parameters of distributed small hydropower; the peak shaving process and peak shaving energy efficiency analysis of the small hydropower cluster; the construction of a comprehensive evaluation system for water-light peak shaving jointly participated by multiple systems; the optimal peak shaving power output strategy, the power supply power output strategy, and the power ratio under the optimal cost and its calculation method; a water-light peak shaving model with the minimum working cost as the optimization goal is constructed, and through the introduction of a multi-objective evaluation system in combination with specific algorithms, the water-light peak shaving work with a high proportion of hydropower participation is realized, improving the acceptability of the power supply system for photovoltaic power and the flexible allocation degree of photovoltaic power generation electric energy. It can realize the flexible dispatching of hydropower stations and reduce the dependence on traditional energy while considering the load demand of the power system and the volatility of new wind and solar energy.

Description

Technical Field

[0001] This application relates to the technical field of small hydropower peak shaving, and particularly to an optimization method for coordinated peak shaving of a hydropower group cascade and a photovoltaic power station. Background Art

[0002] With the increasing global attention to carbon emissions and the severity of climate change, countries have been increasing their investment and development efforts in clean energy. As a clean and renewable energy source, hydropower has the characteristics of strong peak shaving ability and high stability, and has become an important resource for peak shaving in the power system. Photovoltaic power generation, as another clean energy source, although it has the advantages of environmental protection and renewability, its output is affected by natural conditions such as weather, and its volatility is relatively large, unable to meet the stable demand of the power system.

[0003] Therefore, applying peak shaving technology to a high proportion of hydropower resources and photovoltaic power generation resources can give full play to the peak shaving ability of hydropower, make up for the volatility of photovoltaic power generation, and thus achieve the stable operation of the power system. The background of this technology application aims to solve the instability and reliability problems of clean energy, promote the transformation of the power system towards a clean, low-carbon and sustainable direction, and achieve the sustainable development of energy supply. At the same time, this also helps to reduce the dependence on traditional fossil energy, reduce carbon emissions, and provide important support for building a green energy system. Summary of the Invention

[0004] To solve the above technical problems, this application proposes an optimization method for coordinated peak shaving of a hydropower group cascade and a photovoltaic power station. The specific steps are as follows:

[0005] 1. An optimization method for coordinated peak shaving of a hydropower group cascade and a photovoltaic power station, the specific steps are as follows:

[0006] S1. Analysis of the power characteristics of distributed photovoltaic power stations and hydropower station clusters;

[0007] Assume that the output of wind and light satisfies the normal distribution N(μ,σ2), where the mean of the wind and light output prediction is μ and the variance is σ2; use the Monte Carlo simulation method to generate a large number of wind and light output scenarios that conform to the normal distribution, and the probability of each generated scenario is obtained by dividing 1 by the total number of scenarios; combining the above power characteristics and random sampling method, while retaining the volatility of photovoltaic power generation, as much as possible through the random sampling method to achieve the concrete description of power, and realize the linearization of photovoltaic power generation participating in peak shaving.

[0008] The simulation of hydropower station power focuses on the extraction of hydropower power historical data, and learns the historical data through an artificial neural network model, and then realizes the simulation of hydropower power data. The mathematical expression of the artificial neural network is as follows:

[0009]

[0010]

[0011] wherein; a n represents the nth weight coefficient of the training parameters involved in the artificial neural network; θ (j) n is the memory matrix of the coefficients of each layer in the artificial neural network, used to store the data flowing from the upper layer to the lower layer learning network; g(.) is the hidden layer function, used to calculate the data of each output layer; h θ (.) is the output hydropower of each output layer;

[0012] Then, the photovoltaic power and hydropower curves are simulated by the random sampling method and the artificial neural network method respectively;

[0013] S2. Peak shaving process and peak shaving energy efficiency analysis of small hydropower clusters;

[0014] First, the photovoltaic power generation prediction is carried out; then, the hydropower station makes peak shaving preparation work, including adjusting the reservoir water level and planning the operation mode of the generating units; subsequently, the photovoltaic power station generates electricity according to the predicted light conditions, generating corresponding electric energy; finally, the energy efficiency analysis of the peak shaving process is carried out to evaluate the effect and quality of peak shaving, including peak shaving cost and stability index of the power system;

[0015] S3. Construct a comprehensive evaluation system for water-light peak shaving jointly participated by multiple systems;

[0016] Based on the existing water-light peak shaving models and achievements, the depth of hydropower peak shaving participation, the fluctuation range of photovoltaic power, and the minimization of peak shaving scheduling cost are mainly considered to construct a hydropower participation depth model, a water-light peak shaving quality evaluation model with high hydropower participation, and a cost minimization objective function of the water-light peak shaving model with high hydropower participation. And the verification work of the proposed method is carried out in combination with the actual objects of the regional water-light power station group, and a comprehensive evaluation model of water-light peak shaving quality is constructed with the models based on hydropower and photovoltaic collaborative peak shaving, the model based on system optimization, and the model based on comprehensive indicators as the evaluation indicators;

[0017] 1. Hydropower participation depth model

[0018] The hydropower station peak shaving depth model consists of two parts, which include: (1) The hydropower peak shaving depth model calculated based on the maximum and minimum values of the input and output of the hydropower station system; (2) The hydropower station peak shaving capacity ratio. The above two parts respectively represent the proportion of the hydropower station peak shaving power in the rated power of the hydropower station and the proportion of the hydropower station peak shaving power in the total water-light power;

[0019]

[0020] wherein, D h is the hydropower peak shaving depth;

[0021] D h,max and D h,min are the maximum and minimum values of the system output electric power respectively;

[0022] Then there is the peak shaving capacity ratio. The peak shaving capacity ratio is a mathematical model that characterizes the participation of hydropower in the photovoltaic peak shaving operation by comparing the power values before and after photovoltaic peak shaving. Its mathematical expression is as follows:

[0023]

[0024] In the formula, r is the peak shaving capacity ratio;

[0025] P E is the photovoltaic power generation output power before peak shaving;

[0026] P' E represents the photovoltaic power generation output power after peak shaving;

[0027] 2. High-hydropower-participation photovoltaic-hydro peak shaving quality evaluation model;

[0028] To accurately describe the peak-valley changes in the electric power of photovoltaic grid connection, the load changes in 24 dispatching periods are selected, and the system's 24-point peak shaving demand is defined as the absolute difference between the maximum load differences of the latter period and the former period, that is:

[0029]

[0030] In the formula, ΔP LD,t is the peak shaving demand at time t;

[0031] P LD,t and P LD,t+1 are the electric power values of photovoltaic power at times t and t + 1 respectively;

[0032] The above are the peak-valley change characteristics of the power after photovoltaic grid connection. However, since photovoltaic power usually has the characteristic of short-term power mutation, it is necessary to additionally analyze the power ramp value of photovoltaic power generation. For the rising and falling situations of photovoltaic power, the sum of the adjustable capacities of all conventional hydropower units in each period is used to determine the power capacity that the system can use for peak shaving. Further, the power ramp overlimit values at each sampling time node in the peak shaving system can be determined, that is:

[0033]

[0034] Using this index, the ramp value of photovoltaic power generation can be evaluated, and thus the average power value of the photovoltaic-hydro peak shaving system can be deduced, and the smoothness of photovoltaic power can be calculated;

[0035] 3. High-hydropower-participation photovoltaic-hydro peak shaving model cost minimization objective function;

[0036] The cost calculation model is simplified to only calculate the hydropower peak shaving cost, the main grid peak shaving cost, the hydropower station electricity storage / discharge cost, the photovoltaic power generation cost, and the main grid power supply cost. The total cost can be expressed by the following formula:

[0037]

[0038] In the formula, C hydro represents the cost related to hydropower;

[0039] C pv represents the cost related to photovoltaic power;

[0040] C grid represents the cost related to the electricity from the power grid;

[0041] C dis and C chr respectively represent the costs related to the hydropower station discharging extra electric energy and the hydropower station storing redundant electric energy;

[0042] Through the above definition of the cost function, the cost of the water-light peak shaving model is concretely described, providing a mathematical model basis for further carrying out optimal electric energy distribution;

[0043] S4. Calculation method of the optimal peak shaving power output strategy under cost minimization;

[0044] 1. In the constraint model proposed in S3, there are both continuous variables such as P h and Q h , and 0-1 binary variables represented by δ1 and δ2. This makes the optimization model both linear and non-linear, and its expression is as follows.

[0045]

[0046] In the formula, f(x, y) is the objective function, and its decision variable matrixes are x and y;

[0047] x is the 0-1 binary variable matrix, which is the characterization parameter of the hydropower station working mode;

[0048] y is the total set of variable matrixes participating in the optimization, representing the peak shaving power of photovoltaic-hydro.

[0049] C, D, and e are abstract matrixes and vectors, representing the costs and various coefficients in the objective function and constraint conditions.

[0050] 2. For the above two-stage robust model, the C&CG algorithm is used to decompose it into a master problem and a sub-problem for alternating solution.

[0051] 3. Construct the master - sub - bilevel problem model based on the C&CG algorithm decomposition. After decomposition, the mathematical model of the master - sub - bilevel problem is shown as the following formula.

[0052]

[0053] The above formula is the system master problem, where: θ is the auxiliary variable of the micro - grid operation cost in the objective function of the master problem.

[0054]

[0055]

[0056] The above formula is the system master problem, where: θ is the auxiliary variable of the micro - grid operation cost in the objective function of the master problem.

[0057]

[0058]

[0059] Where: n is the number of cascade hydropower stations; t is the time period divided in the planned scheduling period; λ n is the water energy utilization rate of the nth hydropower station, that is, the ability coefficient of converting water energy into electric energy by the generator set; p n,t is the on - grid electricity price of the n hydropower stations at the t - th time period; q n,t is the average power generation water flow of the nth hydropower station at the t - th time period; h n,t is the average head of the nth hydropower station at the t - th time period; m t is the duration of the t - th time period. Further, the constraint conditions of each parameter in the above formula are as follows:

[0060] Water balance constraint:

[0061]

[0062] Where: v n,t and v n,t+1 are the reservoir storage capacities of the nth hydropower station at the t - th time period and the beginning of the t + 1 - th time period.

[0063] Reservoir water storage level constraint:

[0064]

[0065] Where: h n,min and h n,max are the minimum and maximum water storage levels that the reservoir of the nth hydropower station must satisfy during the scheduling period.

[0066] Power generation water flow constraint:

[0067]

[0068] Where: q n,min , q n,max respectively represent the minimum and maximum power generation water flow rates of the nth hydropower station.

[0069] Power generation water consumption constraint:

[0070]

[0071] Where: w n,min , w n,max respectively represent the minimum and maximum power generation water consumption of the nth hydropower station.

[0072] Hydropower station output constraint:

[0073]

[0074] Where: N n,min , N n,max respectively represent the minimum and maximum output powers of the units of the nth hydropower station.

[0075] 4. Linearize the calculation model of power flow parameters. To simplify the calculation, a Distflow DC power flow model with strong adaptability to the optimal power flow model is used to calculate the node voltage parameters of the regional water-light bus. The mathematical expression of the Distflow model is as follows:

[0076]

[0077] Where, P ij and Q ij are the active power and reactive power of the branch respectively;

[0078] v ij and l ij are the square of the amplitudes of the branch voltage and branch current respectively;

[0079] Further calculate the node voltage and node voltage phase angle of each node. The calculation formulas for the two parameters are as follows:

[0080]

[0081]

[0082] Where; δ i is the node voltage drop of the ith node;

[0083] V i represents the node voltage of this node.

[0084] Finally, peak shaving optimization is achieved by using the typical weekly load curve, and the output of the cascade hydropower plants and the wind and solar power plants at each moment is arranged, which can give full play to the load-following ability of the hydropower and solar power plants as much as possible and minimize the fluctuation of the remaining load to the greatest extent.

[0085] Compared with the prior art, the beneficial effects of the present application adopting the above technical solutions are as follows:

[0086] (1) The water inflow conditions and dispatching operation requirements of hydropower plants are different in the flood season, normal water season and dry season, and the complementarity between each hydropower plant and photovoltaic power also varies. The present invention analyzes the peak shaving evaluation indexes, finds the optimal combination of water and light in each period to form an equivalent power plant, and conducts combined peak shaving dispatching with other cascade hydropower plants.

[0087] (2) The coordinated peak shaving optimization operation strategy of the cascade hydropower plants and the wind and solar power plants considering the uncertainty of the output of the wind and solar power plants proposed by the present invention utilizes the good peak shaving ability of the cascade hydropower plants, further extends the cascade structure of the hydropower plants, generalizes its cascade structure to the peak shaving system, enables the power system to accept more output of new energy power generation from wind and solar power, and at the same time reduces the peak shaving pressure of the system. In addition, in the process of solving the optimal solution, the present invention linearizes the nonlinear hydropower conversion function to facilitate model solution. Finally, peak shaving optimization is achieved by using the typical weekly load curve, and the output of the cascade hydropower plants and the wind and solar power plants at each moment is arranged, which can give full play to the load-following ability of the hydropower and solar power plants as much as possible and minimize the fluctuation of the remaining load to the greatest extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 It is a network diagram of IEEE 22 nodes;

[0089] Figure 2 It is a flow chart of water-light peak shaving;

[0090] Figure 3 It is a curve diagram of the photovoltaic power generation before and after peak shaving;

[0091] Figure 4 It is a curve diagram of the peak shaving cost under different operating environments;

[0092] Figure 5 It is the frequency change of the reconnection of the hydropower plant under fault interference;

[0093] Figure 6 It is a bar chart of the peak shaving power ratio including the participation of the hydropower plant. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0094] The following further describes in detail the specific embodiments of the present application in conjunction with the drawings and embodiments. The following embodiments are used to illustrate the present application, but are not used to limit the scope of the present application.

[0095] A high hydropower participation hydro-solar peak-shaving model for hydropower station clusters and a solution method thereof, characterized in that the technology is divided into the following steps:

[0096] S1. Analysis of power characteristics of distributed photovoltaic power stations and hydropower station clusters;

[0097] S1.1. Analysis of power curve characteristics of distributed photovoltaic power generation;

[0098] The photovoltaic power generation power P is known PV The linear constraints are satisfied, and the basic mathematical constraints are as follows:

[0099]

[0100] Where: P pv Indicates the power output of the photoelectric device;

[0101] Pmin pv and Pmax pv represent the minimum and maximum values ​​of the photovoltaic power output, respectively;

[0102] Assume that the wind and solar power output satisfies the normal distribution N(μ,σ2), where the predicted mean of the wind and solar power output is μ and the variance is σ2; ​​use the Monte Carlo simulation method to generate a large number of wind and solar power output scenarios that conform to the normal distribution, and the probability of each generated scenario is obtained by dividing 1 by the total number of scenarios; combining the above power characteristics and random sampling methods, the present invention retains the volatility of photovoltaic power generation while achieving a concrete characterization of power through random sampling methods as much as possible, thereby realizing the linearization of photovoltaic power generation participating in peak regulation.

[0103] The power simulation of hydropower stations focuses on the extraction of historical data of hydropower power, and the use of artificial neural network models to learn historical data, and then realize the simulation of hydropower power data. The mathematical expression of artificial neural network is as follows:

[0104]

[0105]

[0106] Where: a n Represents the nth weight coefficient of the training parameters in the artificial neural network; θ (j) n is the memory matrix of the coefficients of each layer in the artificial neural network, which is used to save the data of the previous layer flowing into the next layer of learning network; g(.) is the hidden layer function, which is used to calculate the data of each output layer; h θ (.) is the output hydropower power for each output layer;

[0107] S1.2, Parameter analysis of distributed small hydropower power curve;

[0108] Furthermore, the light and hydropower power curves are simulated by random sampling method and artificial neural network method respectively.

[0109] S2. Peak shaving process and peak shaving energy efficiency analysis of small hydropower clusters;

[0110] The peak shaving process of hydropower clusters and photovoltaics usually includes the following steps:

[0111] First, conduct photovoltaic power generation prediction; use methods such as weather forecasting to predict the power generation of photovoltaic power plants in the future for a period of time.

[0112] Then, hydropower stations make peak shaving preparation work, including adjusting reservoir water levels and planning the operation modes of generating units;

[0113] Subsequently, the photovoltaic power station generates electricity according to the predicted light conditions, generating corresponding electric energy; the hydropower station adjusts the reservoir water level and the output of generating units in real time according to the actual changes in photovoltaic power generation to provide necessary peak shaving support; the hydropower station and the photovoltaic power station coordinate and interact through the information communication system to ensure the effective implementation of peak shaving work.

[0114] Finally, conduct energy efficiency analysis on the peak shaving process, evaluate the effect and quality of peak shaving, including indicators such as peak shaving cost and the stability of the power system. This kind of peak shaving energy efficiency analysis mainly involves aspects such as peak shaving effect evaluation, peak shaving cost analysis, and power system stability analysis, comprehensively evaluating the energy efficiency of hydropower cluster peak shaving and photovoltaics, providing a scientific basis and optimization direction for peak shaving work.

[0115] S3. Build a comprehensive evaluation system for water-light peak shaving jointly participated by multiple systems;

[0116] In the traditional evaluation system, a single evaluation index is often used to measure the quality of water-light peak shaving. However, with the increasing proportion of photovoltaic power generation in energy supply year by year, the single evaluation system can no longer meet the needs in some cases of extreme photovoltaic power output. Therefore, building a comprehensive evaluation system for water-light peak shaving jointly participated by multiple systems has become an important research direction at present.

[0117] Considering that there are many hydropower stations distributed in the real-scene area, and a large number of wind and solar power generation devices are built along the hydropower stations, the renewable energy microgrid in this area generally shows the comprehensiveness of wind, light, and water renewable energies. In addition, the geographical and climatic characteristics of many mountains and rains make the local hydropower stations show the characteristics of small-scale and decentralized distribution. It has profound practical significance to study the characteristics of multi-energy complementarity of wind, light, and water and propose the optimal power consumption plan for distributed small hydropower stations.

[0118] This invention synthesizes existing water-light peak shaving models and achievements, focuses on aspects such as the depth of hydropower peak shaving participation, the fluctuation range of photovoltaic power, and the minimization of peak shaving scheduling costs, and verifies the proposed method in combination with the actual object of an actual case of a water-light power station group.

[0119] Current evaluation criteria for the quality of single water-light peak shaving include: 1. A model based on the collaborative peak shaving of hydropower and photovoltaic power; 2. A model based on power prediction; 3. A model based on system optimization; 4. A model based on comprehensive indicators; 5. A model based on measured data. Considering the power curve characteristics of real hydropower stations and photovoltaic power stations, this invention selects the first, third, and fourth evaluation indicators among them to construct a comprehensive evaluation model for the quality of water-light peak shaving.

[0120] S3.1. A model for the depth of hydropower participation in a hydropower peak shaving system with high hydropower participation;

[0121] Distributed small hydropower stations usually exhibit characteristics such as wide distribution, large quantity, and stable power output, which makes the participation ratio of hydropower stations in peak shaving and energy supply work usually relatively high. To sum up, increasing the participation ratio of hydropower in peak shaving work as much as possible is beneficial to the carbon reduction goal of the overall system and the reduction of the total system cost.

[0122] Furthermore, a model for the depth of hydropower participation is constructed, and a hydropower peak shaving depth model as shown in the following formula is introduced.

[0123] The hydropower peak shaving depth model consists of two parts. It includes: (1) A hydropower peak shaving depth model calculated based on the maximum and minimum values of the input and output of the hydropower station system; (2) The peak shaving capacity ratio of the hydropower station. The above two parts respectively represent the proportion of the peak shaving power of the hydropower station in the rated power of the hydropower station and the proportion of the peak shaving power of the hydropower station in the total water-light power.

[0124]

[0125] In the formula, D h is the depth of hydropower peak shaving;

[0126] D h,max and D h,min are respectively the maximum and minimum values of the output electric power of the system.

[0127] Then there is the peak shaving capacity ratio. The peak shaving capacity ratio is a mathematical model that characterizes the participation degree of hydropower in photovoltaic peak shaving work by comparing the power values before and after photovoltaic peak shaving, and its mathematical expression is as follows.

[0128]

[0129] In the formula, r is the peak shaving capacity ratio;

[0130] P Eis the photovoltaic power generation output power before peak shaving;

[0131] P' E represents the photovoltaic power generation output power after peak shaving.

[0132] S3.2, Hydro-power-involved water-light peak shaving quality assessment model;

[0133] The maximum peak-valley difference of the system net load after photovoltaic grid connection has a strong impact on the stability of the power system. Therefore, minimizing the impact brought by photovoltaic grid connection is of great significance to the stability of the distribution system.

[0134] To accurately describe the peak-valley changes of the electric power of photovoltaic grid connection, the present invention selects the load changes in 24 dispatching periods, and defines the system's 24-point peak shaving demand as the absolute difference between the maximum load differences of the latter period and the former period, that is:

[0135]

[0136] In the formula, ΔP LD,t is the peak shaving demand at time t;

[0137] P LD,t and P LD,t+1 are the electric power values of the photovoltaic power at times t and t + 1 respectively;

[0138] The above are the peak-valley change characteristics of the power after photovoltaic grid connection. However, due to the short-time power mutation characteristics of photovoltaic power, it is necessary to additionally analyze the power ramp value of photovoltaic power generation. For the rising and falling situations of photovoltaic power, the sum of the adjustable capacities of all conventional hydro-generator units in each period is used to determine the power capacity that the system can use for peak shaving. Further, the power ramp over-limit values at each sampling time node in the peak shaving system can be determined, that is:

[0139]

[0140] Using this index, the ramp value of photovoltaic power generation can be evaluated, and thus the average power value of the water-light peak shaving system can be deduced, and the smoothness of the photovoltaic power can be calculated.

[0141] S3.3, Cost minimization objective function of the hydro-power-involved water-light peak shaving model;

[0142] Considering that both continuous variables such as P h and Q h and 0-1 binary variables represented by δ1 and δ2 exist in the constraint model proposed in this paper, which makes the optimization model both linear and non-linear, and its expression is as follows.

[0143]

[0144] In the formula, f(x, y) is the objective function, and its decision variable matrices are x and y;

[0145] x is a 0-1 binary variable matrix, which is the characterization parameter of the hydropower station operation mode;

[0146] y is the total set of variable matrices participating in the optimization, representing the peak shaving power of photovoltaic-hydro.

[0147] C, D, and e are abstract matrices and vectors, representing the costs and various coefficients in the objective function and constraints.

[0148] For the above two-stage robust model, in this chapter, the C&CG algorithm is used to decompose it into a master problem and a sub-problem for alternating solution.

[0149] S4. Calculation method for the optimal peak shaving power output strategy with minimum cost;

[0150] Based on the above minimum cost model, a master-sub double-layer problem model decomposed by the C&CG algorithm is further constructed. The mathematical model of the master-sub double-layer problem after decomposition is shown in the following formula.

[0151]

[0152] The above formula is the system master problem. In the formula: θ is the auxiliary variable of the microgrid operation cost in the master problem objective function.

[0153]

[0154]

[0155] The above formula is the system master problem. In the formula: θ is the auxiliary variable of the microgrid operation cost in the master problem objective function.

[0156] Furthermore, the present invention linearly constructs the power flow parameter calculation model. The actual hydropower station is a standard IEEE model. However, due to the large number of hydropower stations in practice, the bus connection specifications of hydropower stations are extremely complex. To simplify the calculation, the present invention adopts the Distflow DC power flow model with strong adaptability to the optimal power flow model to calculate the node voltage parameters of the water-light grid-connected bus. The mathematical expression of the Distflow model is shown as follows:

[0157]

[0158] In the formula, P ij and Q ij are the branch active power and branch reactive power respectively;

[0159] v ij and l ijThey are respectively the squares of the magnitudes of the branch voltage and the branch current;

[0160] Further calculate the voltage of each node and the phase angle of each node voltage. The calculation formulas for the two parameters are as follows:

[0161]

[0162]

[0163] In the formula; δ i is the voltage drop of the node voltage of the i-th node;

[0164] V i represents the node voltage of this node;

[0165] Embodiment:

[0166] To verify the feasibility and effectiveness of the practice of the present invention, based on the hydropower generation of the hydropower station group in Longnan area, an experimental power grid improved based on the IEEE22 standard power grid is constructed as the research object.

[0167] Secondly, through the random sampling method and the artificial neural network method, the present invention respectively realizes power simulation of photovoltaic power and hydropower, and generates typical photovoltaic and hydropower power curves. And, in order to select the optimal hydropower grid connection node, it is necessary to analyze the power fluctuation situation of the system after the hydropower station is connected to the grid. Through comparative analysis, the present invention determines nodes 1, 2, and 3 as the grid connection nodes of the hydropower station. For the grid connection nodes of the photovoltaic power station, by comparing the characteristics of the system power flow distribution after the photovoltaic power station is connected to the grid, the present invention selects nodes 5 and 6 to connect the photovoltaic power station. The network structure after the combined operation of the hydropower and photovoltaic power stations is as Figure 1 shown. Using this model can not only simplify the distribution network model, but also realize the large-scale modeling of the IEEE22 nodes through the unified characterization of the distribution network by this model, ensuring the modeling accuracy.

[0168] Since additional considerations such as the peak shaving ratio of the hydropower station and the cost minimization goal are required when the power station conducts the combined operation of hydropower and photovoltaic power, it is necessary to carry out optimal power flow optimization work. Based on the peak shaving dispatching system in the previous S3 and S4, the work flow block diagram of the present invention for carrying out the combined operation of hydropower and photovoltaic power is as Figure 2 shown. Figure 2 The peak shaving process steps in it include the steps:

[0169] 1. Carry out sampling work on the local combined hydropower and photovoltaic power generation data based on the data of local hydropower stations and photovoltaic power stations, and generate a typical daily output power sequence of the combined hydropower and photovoltaic power station;

[0170] 2. Unit commitment simulation, and carry out the calculation of the power flow model of the grid connection experimental model;

[0171] 3. Calculate the expectations of peak shaving demand, peak shaving supply, and peak shaving balance evaluation indicators;

[0172] 4. Determine whether the result converges;

[0173] 5. Output the peak shaving power and obtain the peak shaving balance evaluation result.

[0174] Then, the present invention solves the optimal peak shaving and scheduling scheme of the water-light power under the requirement of cost minimization through the CPLEX solver. The photovoltaic power generation curve obtained after peak shaving by the hydropower station is as Figure 3 shown. It can be seen from the above figure that after the implementation of the combined water-light peak shaving work, the photovoltaic power curve has been smoothed to a large extent, and the goal of peak shaving and valley filling of the photovoltaic power curve has been achieved to a large extent. In addition, the bar chart of the allocation schemes of various powers obtained by the present invention is as Figure 4 shown.

[0175] Finally, in order to verify the effectiveness of the method proposed by the present invention, the present invention calculates the peak shaving scheme of the method proposed by the present invention, and at the same time conducts comparative experiments on some current mainstream methods in the academic circle. Considering the current mainstream methods comprehensively, the present invention selects two methods, namely the traditional linear optimization method and the heuristic optimization method, for comparison. The photovoltaic power generation curves and peak shaving costs after peak shaving obtained by various methods are as Figure 5 shown. In addition, the present invention also compares the peak shaving optimization response times of various methods. The response times of various methods are shown in Table 1.

[0176] It can be seen from Table 1 that the method proposed by the present invention can carry out peak shaving work on the operating environment with large peak-valley differences in a short time to ensure the stability of the grid-connected bus. However, the peak shaving of the present invention will have a small negative impact on the system frequency.

[0177] Table 1 Comparison table of peak shaving response times

[0178]

[0179] Figure 6 are the various power ratio quantities in the water-light peak shaving work calculated by the method proposed by the present invention. From Figure 6 the ratio of various power quantities, it can be seen that the proportion of the hydropower station participating in peak shaving is relatively high, and the goal of high-proportion hydropower peak shaving has been achieved to a certain extent. In addition, the hydropower station also takes into account the responsibilities of energy storage equipment and power generation equipment at the same time. It can be seen from some of the negative power generation data that the hydropower station has a good effect on absorbing redundant electric energy and supplementing the shortage of electric energy.

[0180] By combining Table 1 with Figure 5 、 Figure 6 the grid-connected peak shaving results shown, the feasibility and reliability of the present invention are verified. The peak shaving result of the present invention can carry out peak shaving and valley filling work on the photovoltaic power generation more accurately and efficiently. ThroughFigure 6 The peak shaving power configuration scheme also verifies that the method proposed by the present invention can take into account other sub-goals while optimizing the main goal.

[0181] The above-described embodiments are merely descriptions of the preferred embodiments of the present application, and do not limit the scope of the present application. Without departing from the design spirit of the present application, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present application shall fall within the protection scope determined by the claims of the present application.

[0182] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles applied. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the embodiments of the present disclosure.

Claims

1. An optimization method for coordinated peak load regulation of hydropower cascade and photovoltaic power stations, the specific steps are as follows: S1. Analysis of power characteristics of distributed photovoltaic power stations and hydropower station clusters; Assume that the output of wind and solar power satisfies the normal distribution N(μ,σ2), where the mean of the wind and solar power output prediction is μ and the variance is σ2; ​​use the Monte Carlo simulation method to generate a large number of wind and solar power output scenarios that conform to the normal distribution, and the probability of each generated scenario is obtained by dividing 1 by the total number of scenarios; combine the above power characteristics and random sampling methods, while retaining the volatility of photovoltaic power generation, try to achieve the concrete characterization of power through random sampling methods, and realize the linearization of photovoltaic power generation participating in peak regulation. The power simulation of hydropower stations focuses on the extraction of historical data of hydropower power, and learns the historical data through an artificial neural network model, and then realizes the simulation of hydropower power data. The mathematical expression of the artificial neural network is as follows: where; a n represents the n-th weight coefficient of the training parameters involved in the artificial neural network; θ (j) n is the memory matrix of the coefficients of each layer in the artificial neural network, used to store the data flowing from the previous layer into the next layer of the learning network; g(.) is the hidden layer function, used to calculate the data of each output layer; h θ (.) is the output hydropower of each output layer; Then the photovoltaic power and hydropower power curves are simulated respectively by random sampling method and artificial neural network method; S2. Small hydropower cluster peak-shaving process and peak-shaving energy efficiency analysis; First, the photovoltaic power generation power is predicted; then, the hydropower station prepares for peak load regulation, including adjusting the reservoir water level and planning the operation mode of the generator set; then, the photovoltaic power station generates electricity according to the predicted light conditions and generates corresponding electricity; finally, the peak load regulation process is analyzed for energy efficiency, and the effect and quality of peak load regulation are evaluated, including the peak load regulation cost and the stability indicators of the power system; S3. Construct a comprehensive evaluation system for water-light peak regulation involving multiple systems; Based on the existing hydropower and photovoltaic peak-shaving models and achievements, the depth of hydropower peak-shaving participation, the fluctuation amplitude of photovoltaic power, and the minimization of peak-shaving scheduling costs were considered in particular, and a hydropower participation depth model, a hydropower peak-shaving quality assessment model with high hydropower participation, and a hydropower peak-shaving model cost minimization objective function with high hydropower participation were constructed. The proposed method was verified in combination with the actual objects of the regional hydropower and photovoltaic power station groups, and a comprehensive evaluation model for hydropower peak-shaving quality was constructed using the model based on the coordinated peak-shaving of hydropower and photovoltaic power, the model based on system optimization, and the model based on comprehensive indicators as evaluation indicators. The deep model for hydropower participation, and the peak shaving deep model of the hydropower station consists of two parts, including: (1) Hydropower peak-shaving depth model based on the maximum and minimum values ​​of the hydropower station system input and output; (2) Hydropower station peak-shaving capacity ratio; the above two parts represent the proportion of the hydropower station peak-shaving power in the rated power of the hydropower station and the proportion of the hydropower station peak-shaving power in the total hydropower and solar power power respectively; Where D h is the depth of hydropower peak regulation; D h,max and D h,min are the maximum and minimum values of the system output electric power respectively; Then there is the peak-shaving capacity ratio. The peak-shaving capacity ratio is a mathematical model that characterizes the participation of hydropower in photovoltaic peak-shaving by comparing the power values ​​before and after photovoltaic peak-shaving. Its mathematical expression is as follows: Where r is the peak load capacity ratio; P E is the photovoltaic power generation output before peak shaving; P' E Indicates the output power of photovoltaic power generation after peak shaving; A quality assessment model for hydropower peak regulation with high hydropower participation; In order to accurately describe the peak-to-valley changes in the power of photovoltaic grid-connected power, the load changes in 24 scheduling periods are selected, and the 24-point peak-shaving demand of the system is defined as the absolute difference between the maximum load difference between the latter period and the previous period, that is: where ΔP LD,t is the peak shaving demand at time t; P LD,t and P LD,t+1 are the electric power values of the optoelectronic power at times t and t + 1, respectively; The above are the characteristics of the power peak-valley variation after photovoltaic power is connected to the grid. However, since photovoltaic power usually has the characteristic of short-term power mutation, it is necessary to additionally analyze the power ramp value of photovoltaic power generation. For the rising and falling conditions of photovoltaic power, the sum of the adjustable capacities of all conventional hydropower units at each time period is used to determine the power capacity that the system can use for peak shaving. Further, the power ramp over-limit values at each sampling time node in the peak shaving system can be determined, which is: Using this index, the ramp value of photovoltaic power generation can be evaluated, and thus the average power value of the photovoltaic-hydro peak shaving system can be deduced, and the smoothness of photovoltaic power can be calculated; The cost minimization objective function of the photovoltaic-hydro peak shaving model with high hydropower participation; The cost calculation model is simplified to only calculate the hydropower peak shaving cost, the main grid peak shaving cost, the hydropower station energy storage / discharge cost, the photovoltaic power generation cost, and the main grid power supply cost. The total cost can be expressed by the following formula: Where C hydro represents the cost related to hydropower; C pv Represents optoelectronic related costs; C grid represents the power-related costs from the power grid; C dis and C chr respectively represent the relevant costs of the hydropower station for discharging additional electric energy and storing redundant electric energy; Through the above definition of the cost function, the cost of the photovoltaic-hydro peak shaving model is concretely described, providing a mathematical model basis for further carrying out optimal power distribution; S4. The calculation method of the optimal peak shaving power output strategy under cost minimization; Both continuous variables such as P h and Q h and binary variables represented by δ1 and δ2 exist in the constraint model proposed in S3. This makes the optimization model both linear and non-linear, and its expression is as follows: In the formula, f(x,y) is the objective function, and its decision variable matrices are x and y; x is a 0-1 binary variable matrix, which is the characterization parameter of the hydropower station working mode; y is the total set of variable matrices participating in the optimization, representing the peak shaving power of photovoltaic-hydro; C, D, and e are abstract matrices and vectors, representing the costs and various coefficients in the objective function and constraint conditions; For the above two-stage robust model, the C&CG algorithm is used to decompose it into a master problem and a sub-problem, which is convenient for alternating solution; Construct a master-slave two-layer problem model based on the decomposition of the C&CG algorithm. The mathematical model of the decomposed master-slave two-layer problem is shown in the following formula; The above formula is the system master problem. In the formula: θ is the auxiliary variable of the microgrid operation cost in the master problem objective function; The power flow parameter calculation model is linearly constructed. For simplicity of calculation, the Distflow DC power flow model with strong adaptability to the optimal power flow model is used to calculate the regional photovoltaic-hydro bus node voltage parameters. The mathematical expression of the Distflow model is shown as follows: Wherein, P ij and Q ij are respectively the active power of the branch and the reactive power of the branch; v ij and l ij are respectively the squares of the magnitudes of the branch voltage and the branch current; Further calculate the voltage of each node and the phase angle of each node voltage. The calculation formulas of the two parameters are shown as follows: where; δ i is the node voltage drop of the i-th node; V i represents the node voltage of this node Finally, the peak shaving optimization is realized by using the typical weekly load curve, and the output of the cascade hydropower group and the wind-solar power station at each moment is arranged, which can give full play to the load following ability of the photovoltaic-hydro power station as much as possible and minimize the fluctuation of the remaining load to the greatest extent.

Images