Probability evaluation method for output power of water-wind-light complementary system

Abstract

A probability evaluation method for output power of a water-wind-light complementary system comprises the following steps: step 1, based on the output power of the water-wind-light complementary system in a typical wind-light uncertain scene, a Cornish-Fisher series method is adopted to construct a display mathematical expression of output power quantile; step 2, establishing a sending power probability evaluation model of the water-wind-solar complementary system; and 3, converting the highly nonlinear model into a linear model, and carrying out multi-time iterative solution to obtain a probability evaluation result of the output power of the water-wind-light complementary system. According to the method, the complementary flexibility of the water-wind-light system can be maximized, the upper limit and the lower limit of the output power of each time section of the system are evaluated with certain confidence, and the obtained output power boundary can be used for the system to report a power adjustable range to a power grid and serves as a reference basis for the power grid to formulate a day-ahead power generation dispatching instruction; therefore, the water-wind-light system can satisfy the execution of the power generation plan of the power grid with a large probability on the premise of considering the wind-light prediction error.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid operation control, and in particular to a probabilistic evaluation method for power delivered by a hydro-wind-solar complementary system. Background Art

[0002] As the pace of low-carbon transformation of electricity accelerates, renewable energy power generation such as wind power, photovoltaics, and hydropower has developed rapidly. With hydropower as the leader, a cascaded water, wind, and solar multi-energy complementary system has been formed.

[0003] The bundling and delivery mode of hydropower, wind power and photovoltaic power is an effective means of developing wind and photovoltaic resources that has been widely recognized in recent years. Hydropower has the characteristics of large capacity and flexible operation, while wind and photovoltaic power have the characteristics of large reserves and short infrastructure cycles. Using the existing transmission channels of hydropower to "bundle" and deliver them can give full play to the technical and economic characteristics of different types of power sources. The mode of hydropower as the main power source and wind power and photovoltaic power as the auxiliary power sources enables hydropower units with regulation capabilities to participate in peak regulation and smooth out the output fluctuations of new energy, which helps to alleviate the phenomenon of insufficient electricity during the dry season and alleviates the situation of "difficulty in absorbing" new energy to a certain extent. In the dry season, hydropower units mainly serve as peak regulation capacity to bear peak load or waist load; in the flood season, in order to avoid water abandonment, they mainly bear base load.

[0004] However, wind and solar energy are affected by natural climate conditions, and their output power is highly random, volatile, and intermittent, making it impossible to accurately predict wind and solar power, resulting in strong uncertainty in the bundled power output of the hydro-wind-solar complementary system, which causes a large deviation between the system output power and the power generation plan required by the power grid and is therefore subject to assessment. Therefore, it is necessary to design an effective interaction method between the hydro-wind-solar complementary system and the power grid, which fully utilizes the flexible complementary characteristics of multiple resources of the hydro-wind-solar system while taking into account the uncertainty of wind and solar power and effectively coordinating water and power regulation, and improves the reliability of the system's power supply support capabilities.

[0005] "Quantification and Improvement Method of Hydropower Flexibility under Large-Scale Wind and Solar Grid Connection Conditions" proposes a method to quantify the availability of hydropower flexibility supply capacity and the flexibility demand quantification method in response to the forecast error of new energy. By embedding the constraints of the flexibility supply and demand relationship into the day-ahead dispatch model, the flexibility supply capacity of hydropower can be recognized, so that the system has a large adjustment margin to cope with the randomness of wind and solar power generation. However, the premise of quantifying the flexibility supply capacity of hydropower is that the system receives the power generation dispatch instruction from the power grid, and then obtains it after the internal reservoir dispatch decision, which makes it impossible to achieve effective coordination between reservoir dispatch and power grid dispatch, resulting in waste of resources. In addition, the water-wind-solar complementary system cannot provide a reference basis for the power grid to formulate power generation dispatch instructions in advance, and there is still a risk that the system's output power cannot meet the requirements of the power grid. Summary of the invention

[0006] The purpose of the present invention is to solve the technical problems that the quantification of the flexible supply capacity of hydropower in the prior art is based on the premise that the system receives the power generation dispatching instructions from the power grid, and then obtains it after the internal reservoir dispatching decision, resulting in the inability to achieve effective coordination between reservoir dispatching and power grid dispatching, thereby causing waste of resources, and the water-wind-solar complementary system cannot provide a reference basis for the power grid to formulate power generation dispatching instructions in advance, and there is still a technical problem that the system power delivery cannot meet the requirements of the power grid. A method for probabilistic evaluation of the power delivery of a water-wind-solar complementary system in the day ahead considering the uncertainty of wind and solar is provided.

[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0008] A probabilistic evaluation method for the power delivered by a hydro-wind-solar hybrid system comprises the following steps:

[0009] Step 1: Based on the power output of the hydro-wind-solar hybrid system under typical wind-solar uncertainty scenarios, the Cornish-Fisher series method is used to construct a mathematical expression for the output power quantile;

[0010] Step 2: Based on the constructed mathematical expression of the transmission power quantile, with the goal of maximizing the upper and lower quantile range of the system's transmission power at each moment under a certain confidence level, a probability assessment model for the transmission power of the hydro-wind-solar hybrid system is established;

[0011] Step 3: Aiming at the nonlinear Cornish-Fisher constraints in the probabilistic assessment model, a small-step iterative solution method based on chain-structured data flow is proposed to transform the highly nonlinear model into a linear model through multiple iterative solutions to obtain the probabilistic assessment results of the power delivered by the hydro-wind-solar hybrid system.

[0012] In step 1, the following sub-steps are used when constructing the explicit mathematical expression for the delivered power quantile:

[0013] Step 1-1: Discretize the uncertainty of wind and solar power output into N based on the scenario method s A typical scene with uncertain scenery;

[0014] Step 1-2: Get N at time t s A one-dimensional array consisting of the transmission power of the hydro-wind-solar hybrid system under the scenario, specifically: Where P 0,t is the system output power at time t, P 0,1,t , P 0,2,t , They are scene 1, scene 2 and scene N at time t respectively. s The system output power under s is the total number of scenes;

[0015] Step 1-3: Use the Cornish-Fisher series method to establish the mathematical relationship between the transmitted power and its quantile.

[0016] In steps 1-3, the fifth-order Cornish-Fisher method of cumulant is used to establish the mathematical relationship between the transmission power and its quantile. The specific expression is as follows:

[0017]

[0018]

[0019]

[0020]

[0021]

[0022]

[0023]

[0024] Among them, P 0,s,t is the system output power under scenario s at time t; μ P0,t , σ P0,t is the mean and standard deviation of the system output power P0 at time t in all scenarios; μ P0,t，3 , μ P0,t,4 , μ P0,t,5 are the normalized 3rd, 4th and 5th order origin moments of the system output power P0 at time t; ξ q is the quantile of the standard Gaussian distribution corresponding to probability q; based on the given probability q, is the normalized quantile of the system output power P0 at time t, F p0,t,q is the denormalized quantile of the system output power P0 at time t with probability q.

[0025] In step 2, the established probability assessment model for the power delivered by the water-wind-solar hybrid system is as follows:

[0026] (1) Objective function

[0027] Based on the displayed mathematical expression of the system power delivery quantile constructed in step 1, the objective function of the probability assessment model of the power delivery of the water-wind-solar hybrid system is to maximize the upper and lower quantile range of the system power delivery at each moment under a certain confidence level:

[0028]

[0029] Among them, T is the total number of scheduling periods; is the probability q at time tmax The denormalized quantile of the system output power under ; is the probability q in period t min The denormalized quantile of the system output power under ;

[0030] (2) Constraints

[0031] 1) Water balance constraints

[0032]

[0033] Where V s,i,t+1 , V s,i,t are the storage capacities of the i-th hydropower station at time t+1 and time t under scenario s; I s,i,t is the inflow flow of the i-th hydropower station at time t under scenario s; Q s,i,t is the outflow of the ith hydropower station at time t under scenario s; △t is the scheduling time interval; τ i-1 is the water flow delay time from the i-1th hydropower station to the i-th hydropower station; is the i-1th hydropower station in time period t-τ after considering water flow lag under scenario s i-1 The flow rate of R s,i,t is the interval flow between power station i-1 and power station i at time t under scenario s; is the power generation flow of the i-th hydropower station at time t under scenario s; is the abandoned water flow of the i-th hydropower station in time period t under scenario s;

[0034] 2) Reservoir water level constraints

[0035]

[0036] In the formula, is the water level in front of the dam of the i-th hydropower station at time t under scenario s; Z i,t , are the lower limit and upper limit of the water level of the ith hydropower station at time t. The lower limit of the water level is usually the dead water level, the lower limit of the dispatching line, etc., and the upper limit of the water level is the flood limit water level, the normal high water level, or the upper limit of the dispatching line, etc. This constraint limits the fluctuation range of the water level;

[0037] 3) Initial and final water level constraints

[0038]

[0039]

[0040] In the formula, They are the initial water level during the dispatch period and the controlled water level at the end of the period; are the scheduling decision water levels of the ith hydropower station at time 1 and time T under scenario s respectively; △Z is the allowable water level deviation at the end of the scheduling period to avoid affecting the water volume in the next scheduling period;

[0041] 4) Outbound flow constraints

[0042]

[0043] In the formula, Q s,i,t is the outflow of the i-th hydropower station in time period t under scenario s; Q i,t , are the lower limit and upper limit of the outflow flow of the ith hydropower station in time period t. The lower limit of the outflow flow is usually determined by the ecological water demand, and the upper limit of the outflow flow is determined by the reservoir's discharge capacity. This constraint ensures the normal range of the reservoir's discharge flow.

[0044] 5) Power station output constraints

[0045]

[0046] In the formula, is the output of the i-th hydropower station in time period t under scenario s; are the lower and upper limits of the output of the ith hydropower station in time period t, respectively;

[0047] 6) Water level-reservoir capacity constraints

[0048]

[0049] In the formula, f i ZV (·) is the nonlinear relationship curve function between water level and storage capacity of the reservoir where power station i is located;

[0050] 7) Tailwater level-discharge flow constraint

[0051]

[0052] In the formula, f i ZQ (·) is the nonlinear relationship curve function between the tailwater level and discharge flow of power station i; is the tailwater level of power station i in time period t under scenario s;

[0053] 8) Unit output constraints

[0054]

[0055] Where P s,i,n,t , P i,n is the output and its upper and lower limits of the nth unit of power station i in time period t under scenario s; us,i,n,t is the start-up and shutdown state variable of the nth unit of power station i in time period t under scenario s. It takes 1 to represent the start-up state and 0 to represent the shutdown state. The output of power station i can be expressed as:

[0056]

[0057] Where N i represents the number of units in power station i;

[0058] 9) Unit power generation flow constraints

[0059]

[0060] In the formula, is the power generation flow and its upper and lower limits of the nth unit of power station i in time period t under scenario s; then the power generation flow of power station i can be expressed as:

[0061]

[0062] 10) Unit vibration zone constraints

[0063]

[0064] In the formula, are the upper and lower limits of the output of the kth vibration zone of the nth unit in power station i;

[0065] 11) Unit start-up and shutdown duration constraints

[0066]

[0067] In the formula, is the startup operation variable of the nth unit in power station i in time period t, and takes 1 to indicate startup operation; is the shutdown operation variable of the nth unit of power station i in time period t, and takes 1 to indicate shutdown operation; are the minimum startup and shutdown durations of the nth unit in power station i, respectively; is the maximum number of startup times of the nth unit of power station i in the scheduling cycle; u s,i,n,t-1 is the start-up and shutdown state variable of the nth unit of power station i in time period t-1 under scenario s, and λ is the time mark variable;

[0068] 12) Unit climbing constraints

[0069] -△P i,n ≤P s,i,n,t+1 -P s,i,n,t ≤△P i,n

[0070] In the formula, △P i,nis the climbing capability of the nth unit in power station i, P s,i,n,t+1 It refers to the output of the nth unit of power station i in time period t+1 under scenario s;

[0071] 13) Unit power generation head constraints

[0072]

[0073] In the formula, H s,i,n,t , is the generating head and head loss of the nth unit of power station i in time period t under scenario s, It refers to the water level in front of the dam of the i-th hydropower station at time t-1 under scenario s;

[0074] 14) Head loss function constraint

[0075]

[0076] In the formula, a i 、b i are the head loss coefficient and loss constant of power station i;

[0077] 15) Unit dynamic characteristics constraints

[0078]

[0079] In the formula, is the output-head-flow nonlinear relationship function of the nth unit of power station i;

[0080] 16) Power balance constraints

[0081]

[0082] Where P 0,s,t is the channel output power in time period t under scenario s; P PV,s,t , are the output and abandoned power of photovoltaic power in time period t under scenario s; P WT,s,t , are the wind power output and abandoned power in time period t under scenario s; P L,s,t is the load in the domain at time period t under scenario s, and I refers to the total number of hydropower stations;

[0083] 17) Restrictions on power and water abandonment

[0084]

[0085] Where Pr{·} is the confidence expression; η is the confidence level;

[0086] 18) Transmission channel constraints

[0087]

[0088] In the formula, is the upper limit of the power transmission channel;

[0089] 19) Nonlinear Cornish-Fisher Constraint

[0090] The specific expression of the Cornish-Fisher constraint is the mathematical relationship between the transmission power and its quantile proposed in steps 1-3. The Cornish-Fisher constraint establishes the mathematical relationship between the system transmission power and its quantile in each scenario.

[0091] In step 3, a small-step iterative solution method based on a chain structure data stream is proposed for the nonlinear Cornish-Fisher constraint in the probability assessment model of the power delivered by the hydro-wind-solar hybrid system. The highly nonlinear model is converted into a linear model for multiple iterative solutions to obtain the probability assessment result of the power delivered by the hydro-wind-solar hybrid system. The method includes the following steps:

[0092] Step 3-1: Initialize the number of iterations k = 0;

[0093] Step 3-2: Obtain the feasible solution of the kth iteration of the power delivered by the hydro-wind-solar hybrid system in the dispatch period under the typical wind-solar uncertainty scenario, and use the Cornish-Fisher series method to calculate the quantile of the power delivered at the kth iteration. max With q min The quantile of the system output power is and

[0094] Step 3-3: Based on the chain structure data flow, derive the partial derivative function of the output power quantile and calculate the partial derivative coefficient of the kth iteration:

[0095]

[0096]

[0097]

[0098]

[0099]

[0100]

[0101]

[0102] in, are the denormalized quantile, normalized quantile, normalized fifth-order origin moment, normalized fourth-order origin moment, normalized third-order origin moment, mean and standard deviation increments of the system output power under probability q at time t at the kth iteration, respectively; is the increment of the system output power under scenario s at time t at the k-th iteration; Substitute the feasible solution of the k-th iteration of the system output power in step 3-2 to obtain the partial derivative coefficients;

[0103] Step 3-4: Set the small step change range of the transmit power. Based on the partial derivative function of the transmit power quantile derived in step 3-3 and the calculation results of the k-th iteration partial derivative coefficient, the following linearized Cornish-Fisher constraint is obtained:

[0104]

[0105]

[0106]

[0107]

[0108]

[0109]

[0110]

[0111]

[0112]

[0113] Among them, δ is the small step change range limit of the output power; are the denormalized quantiles of the power delivered by the wind-water hybrid system with probability q at time t at the k-th iteration and the k+1-th iteration respectively; are the normalized quantiles of the power delivered by the wind-water hybrid system with probability q at time t at the k-th iteration and the k+1-th iteration respectively; are the normalized fifth-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the normalized fourth-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the normalized third-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the standard deviations of the power delivered by the hydro-wind-solar hybrid system at time t in the k-th iteration and the k+1-th iteration respectively; are the mean values ​​of the power delivered by the water-wind-solar hybrid system at time t in the k-th iteration and the k+1-th iteration respectively; it should be noted that the other nonlinear constraints in the model, such as water level-storage capacity constraint, unit vibration zone constraint, unit dynamic characteristic constraint, etc., can be linearized using existing technologies, and the present invention will not make too many explanations;

[0114] Step 3-5: Process and solve the probability assessment model of the hydro-wind-solar hybrid system output power at the kth iteration, and determine whether the model converges;

[0115] Step 3-6: When the model iteration is completed, the obtained That is the probabilistic assessment result of the power delivered by the water, wind and solar systems.

[0116] In step 3-5, determine whether the model has converged. The specific steps are:

[0117] Step 3-5-1: Replace all nonlinear Cornish-Fisher constraints in the probability assessment model of the power output of the hydro-wind-solar hybrid system with the linearized Cornish-Fisher constraints obtained in step 3-4, and All are regarded as decision variables, then the probability evaluation model of the power delivered by the hydro-wind-solar hybrid system becomes a mixed integer linear programming model;

[0118] Step 3-5-2: To gradually increase the probability q during the iteration process max With q min The quantile range under the condition is as follows: Add the following constraints to the mixed integer linear programming model obtained in step 3-5-1:

[0119]

[0120]

[0121] in, They are respectively the probability q at time t at the kth iteration max With q min The quantile of the power delivered by the lower system is a known quantity and has been obtained in step 3-2; They represent the probability q at time t in the k+1th iteration. max With q min The quantile of the power delivered by the wind-solar hybrid system in the upper and lower water level. Both are variables to be determined in the model of the kth iteration;

[0122] Step 3-5-3: Use the Gurobi solver to solve the model obtained in step 3-5-2 to obtain the power delivered by the hydro-wind-solar hybrid system under scenario s at time t at the k+1th iteration. To avoid the error caused by linearization, we further use Update the actual quantile using the mathematical relationship between the transmitted power and its quantile proposed in steps 1-3

[0123] Step 3-5-4: Determine whether the model has converged iteratively. If any of the following two conditions is met, the model converges and the iterative solution ends; otherwise, set k = k + 1 and return to step 3-2;

[0124] Condition 1: If the probability q max With q min The absolute value deviation of the transmitted power quantile results obtained in two consecutive iterations is less than the allowable error ε, that is,

[0125] Condition 2: The number of iterations k reaches the maximum number of iterations K.

[0126] A small-step iterative solution method based on a chain structure data stream, the method converts a highly nonlinear model into a linear model for multiple iterative solutions to obtain a probabilistic evaluation result of the power delivered by a water-wind-solar hybrid system; specifically, the method comprises the following steps:

[0127] Step S1: Initialize the number of iterations k=0;

[0128] Step S2: Obtain the feasible solution of the kth iteration of the power delivered by the hydro-wind-solar hybrid system in the scheduling period under the typical wind-solar uncertainty scenario, and use the Cornish-Fisher series method to calculate the quantile of the power delivered at the kth iteration. max With q min The quantile of the system output power is and

[0129] Step S3: Based on the data flow of the chain structure, derive the partial derivative function of the output power quantile and calculate the partial derivative coefficient of the kth iteration:

[0130]

[0131]

[0132]

[0133]

[0134]

[0135]

[0136]

[0137] in, are the denormalized quantile, normalized quantile, normalized fifth-order origin moment, normalized fourth-order origin moment, normalized third-order origin moment, mean and standard deviation increments of the system output power under probability q at time t at the kth iteration, respectively; is the increment of the system power delivery under scenario s at time t at the k-th iteration; the k-th iteration feasible solution of the system power delivery in step S2 is substituted into the system power delivery, and the partial derivative coefficients can be obtained;

[0138] Step S4: Set the small step change range of the transmitted power. Based on the partial derivative function of the transmitted power quantile derived in step S3 and the calculation results of the k-th iteration partial derivative coefficient, the following linearized Cornish-Fisher constraint is obtained:

[0139]

[0140]

[0141]

[0142]

[0143]

[0144]

[0145]

[0146]

[0147]

[0148] Among them, δ is the small step change range limit of the output power; are the denormalized quantiles of the power delivered by the wind-water hybrid system with probability q at time t at the k-th iteration and the k+1-th iteration respectively; are the normalized quantiles of the power delivered by the wind-water hybrid system with probability q at time t at the k-th iteration and the k+1-th iteration respectively; are the normalized fifth-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the normalized fourth-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the normalized third-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the standard deviations of the power delivered by the hydro-wind-solar hybrid system at time t in the k-th iteration and the k+1-th iteration respectively; are the mean values ​​of the power delivered by the water-wind-solar hybrid system at time t in the k-th iteration and the k+1-th iteration respectively; it should be noted that the other nonlinear constraints in the model, such as water level-storage capacity constraint, unit vibration zone constraint, unit dynamic characteristic constraint, etc., can be linearized using existing technologies, and the present invention will not make too many explanations;

[0149] Step S5: Process and solve the probability assessment model of the power delivered by the water-wind-solar hybrid system at the kth iteration, and determine whether the model converges;

[0150] Step S6: When the model iteration solution is completed, the obtained That is the probabilistic assessment result of the power delivered by the water, wind and solar systems.

[0151] In step S5, it is determined whether the model has converged. The specific steps are as follows:

[0152] Step S5-1: Replace all nonlinear Cornish-Fisher constraints in the probability assessment model of the power delivered by the hydro-wind-solar hybrid system with the linearized Cornish-Fisher constraints obtained in step S4, and All are regarded as decision variables, then the probability evaluation model of the power delivered by the hydro-wind-solar hybrid system becomes a mixed integer linear programming model;

[0153] Step S5-2: To gradually increase the probability q in the iterative process max With q min The quantile range under the condition is as follows:

[0154]

[0155]

[0156] in, They are respectively the probability q at time t at the kth iteration max With q min The quantile of the power delivered by the lower system is a known quantity and has been obtained in step S2; They represent the probability q at time t in the k+1th iteration. max With qmin The quantile of the power delivered by the wind-solar hybrid system in the upper and lower water level. Both are variables to be determined in the model of the kth iteration;

[0157] Step S5-3: Use the Gurobi solver to solve the model obtained in step S5-2 to obtain the power delivered by the water-wind-solar hybrid system under the scenario s at time t at the k+1th iteration: To avoid the error caused by linearization, we further use Update the actual quantile by the mathematical relationship between the delivered power and its quantile

[0158] Step S5-4: Determine whether the model has converged iteratively. If any of the following two conditions is met, the model has converged and the iterative solution ends; otherwise, set k=k+1 and return to step S2;

[0159] Condition 1: If the probability q max With q min The absolute value deviation of the transmitted power quantile results obtained in two consecutive iterations is less than the allowable error ε, that is,

[0160] Condition 2: The number of iterations k reaches the maximum number of iterations K.

[0161] An electronic device comprises a memory and a processor, wherein the processor and the memory are connected via a communication bus, the memory is used to store a computer program, and when the computer program is executed by the processor, the method according to claim 1 to claim 8 is implemented.

[0162] Compared with the prior art, the present invention has the following technical effects:

[0163] 1) The present invention can solve the technical problem of the fuzzy interaction mode between the water-wind-solar complementary system and the power grid in the prior art. Compared with the prior art, the present invention realizes the probability evaluation of the power delivered by the water-wind-solar complementary system under the premise of considering the uncertainty of wind and solar and the regulation capacity of the reservoir. The evaluation result can be used to report to the power grid dispatching department as the basis for the formulation of power grid dispatching instructions, thereby realizing the coordinated cooperation of water and power dispatching and improving resource utilization;

[0164] 2) The present invention provides an interaction mode between a water-wind-solar complementary system and a power grid, that is, the system reports the power delivery range to the power grid dispatching department by evaluating the adjustable range of its delivered power. The power grid dispatching department formulates and issues a feasible power generation plan based on the evaluation results, thereby achieving effective coordination between power dispatching and water dispatching;

[0165] 3) The present invention proposes a probabilistic evaluation method for the power output of a hydropower, wind and solar power system. The method can fully exploit the complementary characteristics of the system and maximize the boundary range of the system power output with a high degree of confidence on the basis of satisfying the constraints of reservoir scheduling, safe operation of units, and abandonment of water and power, thereby ensuring the reliable power supply support capability of the system;

[0166] 4) The present invention proposes a small-step-size iterative model solving method, which transforms the solution of a highly nonlinear model into an iterative solution of a linearized model, thereby improving the model solving efficiency;

[0167] 5) The present invention can maximize the complementary flexibility of the water, wind and solar systems, and evaluate the upper and lower limits of the system's power delivery at each time section with a certain confidence level. The obtained power delivery boundary can be used by the system to report the adjustable power range to the power grid, and serve as a reference for the power grid to formulate day-ahead power generation scheduling instructions, so as to realize the interaction between the water, wind and solar systems and the power grid, so that the water, wind and solar systems can meet the execution of the power grid's power generation plan with a greater confidence level while considering the uncertainty of wind and solar power. BRIEF DESCRIPTION OF THE DRAWINGS

[0168] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0169] Figure 1 is a flow chart of a method for evaluating the probability of power delivered by a water-wind-solar system according to an embodiment of the present invention;

[0170] Figure 2 This is a topological diagram of the water-wind-solar complementary system according to an embodiment of the present invention;

[0171] Figure 3 Schematic diagram of a typical wind power output uncertainty scenario according to an embodiment of the present invention;

[0172] Figure 4 Schematic diagram of a typical photovoltaic output uncertainty scenario according to an embodiment of the present invention;

[0173] Figure 5 Schematic diagram of system power delivery in 10 scenarios of an embodiment of the present invention;

[0174] Figure 6 It is a schematic diagram of the adjustable range of the power delivered by the water-wind-solar system at 90% confidence level in an embodiment of the present invention;

[0175] Figure 7 It is a violin diagram of the probability distribution of power delivered by the wind-solar system at time 35 in an embodiment of the present invention. DETAILED DESCRIPTION

[0176] The present invention provides a probabilistic evaluation method for the power delivered by a water-wind-solar complementary system, comprising the following steps:

[0177] Step 1: Based on the power output of the hydro-wind-solar hybrid system under typical wind-solar uncertainty scenarios, the Cornish-Fisher series method is used to construct a mathematical expression for the output power quantile;

[0178] Step 2: Based on the constructed mathematical expression of the transmission power quantile, with the goal of maximizing the upper and lower quantile range of the system's transmission power at each moment under a certain confidence level, a probability assessment model for the transmission power of the hydro-wind-solar hybrid system is established;

[0179] Step 3: Aiming at the nonlinear Cornish-Fisher constraints in the probabilistic assessment model, a small-step iterative solution method based on chain-structured data flow is proposed to transform the highly nonlinear model into a linear model through multiple iterative solutions to obtain the probabilistic assessment results of the power delivered by the hydro-wind-solar hybrid system.

[0180] In step 1, the following sub-steps are used when constructing the explicit mathematical expression for the delivered power quantile:

[0181] Step 1-1: Discretize the uncertainty of wind and solar power output into N based on the scenario method s A typical scene with uncertain scenery;

[0182] Step 1-2: Get N at time t s A set of one-dimensional arrays composed of the transmission power of the water-wind-solar hybrid system in the scenario, specifically: Where P 0,t is the system output power at time t, They are scene 1, scene 2 and scene N at time t respectively. s The system output power under s is the total number of scenes;

[0183] Step 1-3: Use the Cornish-Fisher series method to establish the mathematical relationship between the transmitted power and its quantile.

[0184] In steps 1-3, the fifth-order Cornish-Fisher method of cumulant is used to establish the mathematical relationship between the transmission power and its quantile. The specific expression is as follows:

[0185]

[0186]

[0187]

[0188]

[0189]

[0190]

[0191]

[0192] Among them, P 0,s,t is the system output power under scenario s at time t; μ P0,t , σ P0,t is the mean and standard deviation of the system output power P0 at time t in all scenarios; μ P0,t，3 , μ P0,t,4 , μ P0,t,5 are the normalized 3rd, 4th and 5th order origin moments of the system output power P0 at time t; ξ q is the quantile of the standard Gaussian distribution corresponding to probability q; based on the given probability q, is the normalized quantile of the system output power P0 at time t, F p0,t,q is the denormalized quantile of the system output power P0 at time t with probability q.

[0193] In step 2, the established probability assessment model for the power delivered by the water-wind-solar hybrid system is as follows:

[0194] (1) Objective function

[0195] Based on the displayed mathematical expression of the system power delivery quantile constructed in step 1, the objective function of the probability assessment model of the power delivery of the water-wind-solar hybrid system is to maximize the upper and lower quantile range of the system power delivery at each moment under a certain confidence level:

[0196]

[0197] Among them, T is the total number of scheduling periods; is the probability q at time t max The denormalized quantile of the system output power under ; is the probability q in period t min The denormalized quantile of the system output power under ;

[0198] (2) Constraints

[0199] 1) Water balance constraints

[0200]

[0201] Where V s,i,t+1 , V s,i,t are the storage capacities of the i-th hydropower station at time t+1 and time t under scenario s; I s,i,t is the inflow flow of the i-th hydropower station at time t under scenario s; Q s,i,t is the outflow of the ith hydropower station at time t under scenario s; △t is the scheduling time interval; τi-1 is the water flow delay time from the i-1th hydropower station to the i-th hydropower station; is the i-1th hydropower station in time period t-τ after considering water flow lag under scenario s i-1 The flow rate of R s,i,t is the interval flow between power station i-1 and power station i at time t under scenario s; is the power generation flow of the i-th hydropower station at time t under scenario s; is the abandoned water flow of the i-th hydropower station in time period t under scenario s;

[0202] 2) Reservoir water level constraints

[0203]

[0204] In the formula, is the water level in front of the dam of the i-th hydropower station at time t under scenario s; Z i,t , are the lower limit and upper limit of the water level of the ith hydropower station at time t. The lower limit of the water level is usually the dead water level, the lower limit of the dispatching line, etc., and the upper limit of the water level is the flood limit water level, the normal high water level, or the upper limit of the dispatching line, etc. This constraint limits the fluctuation range of the water level;

[0205] 3) Initial and final water level constraints

[0206]

[0207]

[0208] In the formula, They are the initial water level during the dispatch period and the controlled water level at the end of the period; are the scheduling decision water levels of the ith hydropower station at time 1 and time T under scenario s respectively; △Z is the allowable water level deviation at the end of the scheduling period to avoid affecting the water volume in the next scheduling period;

[0209] 4) Outbound flow constraints

[0210]

[0211] In the formula, Q s,i,t is the outflow of the i-th hydropower station in time period t under scenario s; Q i,t , are the lower limit and upper limit of the outflow flow of the ith hydropower station in time period t. The lower limit of the outflow flow is usually determined by the ecological water demand, and the upper limit of the outflow flow is determined by the reservoir's discharge capacity. This constraint ensures the normal range of the reservoir's discharge flow.

[0212] 5) Power station output constraints

[0213]

[0214] In the formula, is the output of the i-th hydropower station in time period t under scenario s; are the lower and upper limits of the output of the ith hydropower station in time period t, respectively;

[0215] 6) Water level-reservoir capacity constraints

[0216]

[0217] In the formula, f i ZV (·) is the nonlinear relationship curve function between water level and storage capacity of the reservoir where power station i is located;

[0218] 7) Tailwater level-discharge flow constraint

[0219]

[0220] In the formula, f i ZQ (·) is the nonlinear relationship curve function between the tailwater level and discharge flow of power station i; is the tailwater level of power station i in time period t under scenario s;

[0221] 8) Unit output constraints

[0222]

[0223] Where P s,i,n,t , P i,n is the output and its upper and lower limits of the nth unit of power station i in time period t under scenario s; u s,i,n,t is the start-up and shutdown state variable of the nth unit of power station i in time period t under scenario s. It takes 1 to represent the start-up state and 0 to represent the shutdown state. The output of power station i can be expressed as:

[0224]

[0225] Where N i represents the number of units in power station i;

[0226] 9) Unit power generation flow constraints

[0227]

[0228] In the formula, is the power generation flow and its upper and lower limits of the nth unit of power station i in time period t under scenario s; then the power generation flow of power station i can be expressed as:

[0229]

[0230] 10) Unit vibration zone constraints

[0231]

[0232] In the formula, are the upper and lower limits of the output of the kth vibration zone of the nth unit in power station i;

[0233] 11) Unit start-up and shutdown duration constraints

[0234]

[0235] In the formula, is the startup operation variable of the nth unit in power station i in time period t, and takes 1 to indicate startup operation; is the shutdown operation variable of the nth unit of power station i in time period t, and takes 1 to indicate shutdown operation; are the minimum startup and shutdown durations of the nth unit in power station i, respectively; is the maximum number of startup times of the nth unit of power station i in the scheduling cycle; u s,i,n,t-1 is the start-up and shutdown state variable of the nth unit of power station i in time period t-1 under scenario s, and λ is the time mark variable;

[0236] 12) Unit climbing constraints

[0237] -△P i,n ≤P s,i,n,t+1 -P s,i,n,t ≤△P i,n

[0238] In the formula, △P i,n is the climbing capability of the nth unit in power station i, P s,i,n,t+1 It refers to the output of the nth unit of power station i in time period t+1 under scenario s;

[0239] 13) Unit power generation head constraints

[0240]

[0241] In the formula, H s,i,n,t , is the generating head and head loss of the nth unit of power station i in time period t under scenario s, It refers to the water level in front of the dam of the i-th hydropower station at time t-1 under scenario s;

[0242] 14) Head loss function constraint

[0243]

[0244] In the formula, a i 、b iare the head loss coefficient and loss constant of power station i;

[0245] 15) Unit dynamic characteristics constraints

[0246]

[0247] In the formula, is the output-head-flow nonlinear relationship function of the nth unit of power station i;

[0248] 16) Power balance constraints

[0249]

[0250] Where P 0,s,t is the channel output power in time period t under scenario s; P PV,s,t , are the output and abandoned power of photovoltaic power in time period t under scenario s; P WT,s,t , are the wind power output and abandoned power in time period t under scenario s; P L,s,t is the load in the domain at time period t under scenario s, and I refers to the total number of hydropower stations;

[0251] 17) Restrictions on power and water abandonment

[0252]

[0253] Where Pr{·} is the confidence expression; η is the confidence level;

[0254] 18) Transmission channel constraints

[0255]

[0256] In the formula, is the upper limit of the power transmission channel;

[0257] 19) Nonlinear Cornish-Fisher Constraint

[0258] The specific expression of the Cornish-Fisher constraint is the mathematical relationship between the transmission power and its quantile proposed in steps 1-3. The Cornish-Fisher constraint establishes the mathematical relationship between the system transmission power and its quantile in each scenario.

[0259] In step 3, a small-step iterative solution method based on a chain structure data stream is proposed for the nonlinear Cornish-Fisher constraint in the probability assessment model of the power delivered by the hydro-wind-solar hybrid system. The highly nonlinear model is converted into a linear model for multiple iterative solutions to obtain the probability assessment result of the power delivered by the hydro-wind-solar hybrid system. The method includes the following steps:

[0260] Step 3-1: Initialize the number of iterations k = 0;

[0261] Step 3-2: Obtain the feasible solution of the kth iteration of the power delivered by the hydro-wind-solar hybrid system in the dispatch period under the typical wind-solar uncertainty scenario, and use the Cornish-Fisher series method to calculate the quantile of the power delivered at the kth iteration. max With q min The quantile of the system output power is and

[0262] Step 3-3: Based on the chain structure data flow, derive the partial derivative function of the output power quantile and calculate the partial derivative coefficient of the kth iteration:

[0263]

[0264]

[0265]

[0266]

[0267]

[0268]

[0269]

[0270] in, are the denormalized quantile, normalized quantile, normalized fifth-order origin moment, normalized fourth-order origin moment, normalized third-order origin moment, mean and standard deviation increments of the system output power under probability q at time t at the kth iteration, respectively; is the increment of the system output power under scenario s at time t at the k-th iteration; Substitute the feasible solution of the k-th iteration of the system output power in step 3-2 to obtain the partial derivative coefficients;

[0271] Step 3-4: Set the small step change range of the transmit power. Based on the partial derivative function of the transmit power quantile derived in step 3-3 and the calculation results of the k-th iteration partial derivative coefficient, the following linearized Cornish-Fisher constraint is obtained:

[0272]

[0273]

[0274]

[0275]

[0276]

[0277]

[0278]

[0279]

[0280]

[0281] Among them, δ is the small step change range limit of the output power; are the denormalized quantiles of the power delivered by the wind-water hybrid system with probability q at time t at the k-th iteration and the k+1-th iteration respectively; are the normalized quantiles of the power delivered by the wind-water hybrid system with probability q at time t at the k-th iteration and the k+1-th iteration respectively; are the normalized fifth-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the normalized fourth-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the normalized third-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the standard deviations of the power delivered by the hydro-wind-solar hybrid system at time t in the k-th iteration and the k+1-th iteration respectively; are the mean values ​​of the power delivered by the water-wind-solar hybrid system at time t in the k-th iteration and the k+1-th iteration respectively; it should be noted that the other nonlinear constraints in the model, such as water level-storage capacity constraint, unit vibration zone constraint, unit dynamic characteristic constraint, etc., can be linearized using existing technologies, and the present invention will not make too many explanations;

[0282] Step 3-5: Process and solve the probability assessment model of the hydro-wind-solar hybrid system output power at the kth iteration, and determine whether the model converges;

[0283] Step 3-6: When the model iteration is completed, the obtained That is the probabilistic assessment result of the power delivered by the water, wind and solar systems.

[0284] In step 3-5, determine whether the model has converged. The specific steps are:

[0285] Step 3-5-1: Replace all nonlinear Cornish-Fisher constraints in the probability assessment model of the power output of the hydro-wind-solar hybrid system with the linearized Cornish-Fisher constraints obtained in step 3-4, and All are regarded as decision variables, then the probability evaluation model of the power delivered by the hydro-wind-solar hybrid system becomes a mixed integer linear programming model;

[0286] Step 3-5-2: To gradually increase the probability q during the iteration process max With q min The quantile range under the condition is as follows: Add the following constraints to the mixed integer linear programming model obtained in step 3-5-1:

[0287]

[0288]

[0289] in, They are respectively the probability q at time t at the kth iteration max With q min The quantile of the power delivered by the lower system is a known quantity and has been obtained in step 3-2; They represent the probability q at time t in the k+1th iteration. max With q min The quantile of the power delivered by the wind-solar hybrid system in the upper and lower water level. Both are variables to be determined in the model of the kth iteration;

[0290] Step 3-5-3: Use the Gurobi solver to solve the model obtained in step 3-5-2 to obtain the power delivered by the hydro-wind-solar hybrid system under scenario s at time t at the k+1th iteration. To avoid the error caused by linearization, we further use Update the actual quantile using the mathematical relationship between the transmitted power and its quantile proposed in steps 1-3

[0291] Step 3-5-4: Determine whether the model has converged iteratively. If any of the following two conditions is met, the model converges and the iterative solution ends; otherwise, set k = k + 1 and return to step 3-2;

[0292] Condition 1: If the probability q max With q min The absolute value deviation of the transmitted power quantile results obtained in two consecutive iterations is less than the allowable error ε, that is,

[0293] Condition 2: The number of iterations k reaches the maximum number of iterations K.

[0294] The present invention also includes a small step iterative solution method based on a chain structure data stream, which converts a highly nonlinear model into a linear model for multiple iterative solutions to obtain a probability evaluation result of the power delivered by the water-wind-solar complementary system; specifically, the following steps are included:

[0295] Step S1: Initialize the number of iterations k=0;

[0296] Step S2: Obtain the feasible solution of the kth iteration of the power delivered by the hydro-wind-solar hybrid system in the scheduling period under the typical wind-solar uncertainty scenario, and use the Cornish-Fisher series method to calculate the quantile of the power delivered at the kth iteration. max With q min The quantile of the system output power is and

[0297] Step S3: Based on the data flow of the chain structure, derive the partial derivative function of the output power quantile and calculate the partial derivative coefficient of the kth iteration:

[0298]

[0299]

[0300]

[0301]

[0302]

[0303]

[0304]

[0305] in, are the denormalized quantile, normalized quantile, normalized fifth-order origin moment, normalized fourth-order origin moment, normalized third-order origin moment, mean and standard deviation increments of the system output power under probability q at time t at the kth iteration, respectively; is the increment of the system power delivery under scenario s at time t at the k-th iteration; the k-th iteration feasible solution of the system power delivery in step S2 is substituted into the system power delivery, and the partial derivative coefficients can be obtained;

[0306] Step S4: Set the small step change range of the transmitted power. Based on the partial derivative function of the transmitted power quantile derived in step S3 and the calculation results of the k-th iteration partial derivative coefficient, the following linearized Cornish-Fisher constraint is obtained:

[0307]

[0308]

[0309]

[0310]

[0311]

[0312]

[0313]

[0314]

[0315]

[0316] Among them, δ is the small step change range limit of the output power; are the denormalized quantiles of the power delivered by the wind-water hybrid system with probability q at time t at the k-th iteration and the k+1-th iteration respectively; are the normalized quantiles of the power delivered by the wind-water hybrid system with probability q at time t in the k-th iteration and the k+1-th iteration respectively; are the normalized fifth-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the normalized fourth-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the normalized third-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the standard deviations of the power delivered by the hydro-wind-solar hybrid system at time t in the k-th iteration and the k+1-th iteration respectively; are the mean values ​​of the power delivered by the water-wind-solar hybrid system at time t in the k-th iteration and the k+1-th iteration respectively; it should be noted that the other nonlinear constraints in the model, such as water level-storage capacity constraint, unit vibration zone constraint, unit dynamic characteristic constraint, etc., can be linearized using existing technologies, and the present invention will not make too many explanations;

[0317] Step S5: Process and solve the probability assessment model of the power delivered by the water-wind-solar hybrid system at the kth iteration, and determine whether the model converges;

[0318] Step S6: When the model iteration solution is completed, the obtained That is the probabilistic assessment result of the power delivered by the water, wind and solar systems.

[0319] In step S5, it is determined whether the model has converged. The specific steps are as follows:

[0320] Step S5-1: Replace all nonlinear Cornish-Fisher constraints in the probability assessment model of the power delivered by the hydro-wind-solar hybrid system with the linearized Cornish-Fisher constraints obtained in step S4, and All are regarded as decision variables, then the probability evaluation model of the power delivered by the hydro-wind-solar hybrid system becomes a mixed integer linear programming model;

[0321] Step S5-2: To gradually increase the probability q in the iterative process max With q min The quantile range under the condition is as follows:

[0322]

[0323]

[0324] in, They are respectively the probability q at time t at the kth iteration max With q min The quantile of the power delivered by the lower system is a known quantity and has been obtained in step S2; They represent the probability q at time t in the k+1th iteration. max With q min The quantile of the power delivered by the wind-solar hybrid system in the upper and lower water level. Both are variables to be determined in the model of the kth iteration;

[0325] Step S5-3: Use the Gurobi solver to solve the model obtained in step S5-2 to obtain the power delivered by the water-wind-solar hybrid system under the scenario s at time t at the k+1th iteration: To avoid the error caused by linearization, we further use Update the actual quantile by the mathematical relationship between the delivered power and its quantile

[0326] Step S5-4: Determine whether the model has converged iteratively. If any of the following two conditions is met, the model has converged and the iterative solution ends; otherwise, set k=k+1 and return to step S2;

[0327] Condition 1: If the probability q max With q min The absolute value deviation of the transmitted power quantile results obtained in two consecutive iterations is less than the allowable error ε, that is,

[0328] Condition 2: The number of iterations k reaches the maximum number of iterations K.

[0329] An electronic device includes: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus. The processor can call the logic instructions in the memory to execute a probability evaluation method for the power delivered by a water-wind-solar complementary system and a small step iterative solution method based on a chain structure data stream.

[0330] In addition, the logic instructions in the memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the present invention. The storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0331] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the probabilistic evaluation method of the power transmitted by a water-wind-solar complementary system and a small-step iterative solution method based on a chain-structured data stream.

[0332] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute a probabilistic evaluation method for the power delivered by a water-wind-solar complementary system and a small-step iterative solution method based on a chain-structured data stream.

[0333] Example:

[0334] A method for evaluating the probability of day-ahead power transmission of a hydro-wind-solar hybrid system considering the uncertainty of wind and solar power. The overall process is as follows: Figure 1 As shown; the system topology diagram corresponding to the embodiment is as shown Figure 2 As shown in Figure 1, the system includes 15 units of 4 hydropower stations, 1 wind farm and 1 photovoltaic power station, with 1 day as the dispatching cycle and 15 minutes as the dispatching period. The characteristic parameters of each cascade hydropower station are shown in Table 1. Other model parameters are set as follows: the capacity of the wind farm and photovoltaic power station are 550MW and 500MW respectively; the capacity of each transmission section is 1500MW, 1500MW, 1000MW and 4000MW.

[0335] Table 1. Hydropower station characteristic parameters

[0336]

[0337] A method for evaluating the probability of the day-ahead power delivered by a hydro-wind-solar hybrid system considering the uncertainty of wind and solar power is implemented for the system. The method includes the following steps:

[0338] 1) Based on the transmitted power of the hydro-wind-solar hybrid system under typical wind-solar uncertainty scenarios, the Cornish-Fisher series method is used to construct an explicit mathematical expression of the transmitted power quantile.

[0339] Due to the prediction error of the wind and solar power day-ahead power, the uncertainty of wind and solar power output is discretized into N based on the scenario method. s Typical scene uncertainty scenarios, such as Figure 3 The following are 10 typical wind power output uncertainty scenarios: Figure 4 The following are 10 typical scenarios of photovoltaic power output uncertainty. At time t, the power output of the hydro-wind-solar hybrid system under different scenarios can form a one-dimensional array, namely Where P 0,t is the system output power at time t, P 0,1,t , P 0,2,t , They are scene 1, scene 2 and scene N respectively. s The system output power under s The total number of scenarios. The fifth-order Cornish-Fisher method of cumulant is used to establish the mathematical relationship between the transmission power and its quantile, as shown below:

[0340]

[0341]

[0342]

[0343]

[0344]

[0345]

[0346]

[0347] Among them, P 0,s,t The power delivered to the system in scenario s; μ P0,t , σ P0,t is the mean and standard deviation of the system output power in period t under all scenarios; μ P0,t，3, μ P0,t,4 , μ P0,t,5 are the normalized 3rd, 4th and 5th order origin moments respectively; ξ q is the quantile of the standard Gaussian distribution corresponding to probability q; based on the given probability q, is the normalized quantile, F p0,t,q is the denormalized quantile.

[0348] Taking time 15 as an example, the system output power in 10 scenarios is as follows: Figure 5 As shown, q is taken as 0.05 and 0.95 respectively, and the Cornish-Fisher calculation results are shown in Table 2:

[0349] Table 2 Cornish-Fisher calculation results at time 15

[0350]

[0351] 2) Establish a probability assessment model for the day-ahead power delivery of the hydro-wind-solar hybrid system, and maximize the upper and lower boundaries of the system's power delivery at each time section with a certain degree of confidence as follows:

[0352] (1) Objective function

[0353] Taking advantage of the complementary characteristics of wind power, photovoltaic power and hydropower output, a probabilistic assessment model for the transmission capacity of the hydro-wind-solar complementary system is established. The assessment model aims to maximize the joint transmission range of each time section of the system under a certain confidence interval:

[0354]

[0355] Among them, T is the total number of scheduling periods; is the probability q in period t max The quantile of the system output power under ; is the probability q in period t min In this embodiment, q max is 0.95, q min is 0.05.

[0356] (2) Constraints

[0357] 1) Water balance constraints

[0358]

[0359] Where V s,i,t+1 , V s,i,t are the storage capacities of the i-th hydropower station at time t+1 and time t under scenario s; I s,i,t is the inflow of the i-th hydropower station in time period t under scenario s; Q s,i,tis the outflow of the ith hydropower station in time period t under scenario s; △t is the scheduling time interval; τ i-1 is the water flow delay time from the i-1th hydropower station to the i-th hydropower station; is the i-1th hydropower station in time period t-τ after considering water flow lag under scenario s i-1 The flow rate of R s,i,t is the interval flow between power station i-1 and power station i in time period t under scenario s; is the power generation flow of the i-th hydropower station in time period t under scenario s; is the abandoned water flow of the i-th hydropower station in time period t under scenario s.

[0360] 2) Reservoir water level constraints

[0361]

[0362] In the formula, is the water level in front of the dam of the i-th hydropower station at time t under scenario s; Z i,t , are the lower and upper limits of the water level of the ith hydropower station at time t. The lower limit of the water level is often the dead water level, the lower limit of the dispatching line, etc., and the upper limit of the water level is the flood limit water level, the normal high water level, or the upper limit of the dispatching line, etc. This constraint limits the fluctuation range of the water level.

[0363] 3) Initial and final water level constraints

[0364]

[0365]

[0366] In the formula, are the initial water level and the controlled water level at the end of the scheduling period; is the allowable water level deviation at the end of the scheduling period to avoid affecting the water volume in the next scheduling period.

[0367] 4) Outbound flow constraints

[0368]

[0369] In the formula, Q s,i,t is the outflow of the i-th hydropower station in time period t under scenario s; Q i,t , are the lower limit and upper limit of the outflow flow of the ith hydropower station in time period t. The lower limit of the outflow flow is usually determined by the ecological water demand, while the upper limit of the outflow flow is determined by the reservoir's discharge capacity. This constraint ensures the normal range of the reservoir's discharge flow.

[0370] 5) Power station output constraints

[0371]

[0372] In the formula, is the output of the i-th hydropower station in time period t under scenario s; are the lower and upper limits of the output of the i-th hydropower station in time period t respectively.

[0373] 6) Water level-reservoir capacity constraints

[0374]

[0375] In the formula, f i ZV (·) is the nonlinear relationship curve function between water level and storage capacity of the reservoir where power station i is located.

[0376] 7) Tailwater level-discharge flow constraint

[0377]

[0378] In the formula, f i ZQ (·) is the nonlinear relationship curve function between the tailwater level and discharge flow of power station i; is the tailwater level of power station i in time period t under scenario s.

[0379] 8) Unit output constraints

[0380]

[0381] Where P s,i,n,t , P i,n is the output and its upper and lower limits of the nth unit of power station i in time period t under scenario s; u s,i,n,t is the start-up and shutdown state variable of the nth unit of power station i in time period t under scenario s. It takes 1 to represent the start-up state and 0 to represent the shutdown state. The output of power station i can be expressed as:

[0382]

[0383] Where N i Represents the number of units in power station i.

[0384] 9) Unit power generation flow constraints

[0385]

[0386] In the formula, is the power generation flow and its upper and lower limits of the nth unit of power station i in time period t under scenario s; then the power generation flow of power station i can be expressed as:

[0387]

[0388] 10) Unit vibration zone constraints

[0389]

[0390] In the formula, It is the upper and lower limits of the output of the kth vibration zone of the nth unit in power station i.

[0391] 11) Unit start-up and shutdown duration constraints

[0392]

[0393] In the formula, is the startup operation variable of the nth unit in power station i in time period t, and takes 1 to indicate startup operation; is the shutdown operation variable of the nth unit of power station i in time period t, and takes 1 to indicate shutdown operation; are the minimum startup and shutdown durations of the nth unit in power station i, respectively; is the maximum number of startup times of the nth unit of power station i within the scheduling cycle.

[0394] 12) Unit climbing constraints

[0395]

[0396] In the formula, △P i,n is the climbing capacity of the nth unit in power station i.

[0397] 13) Unit power generation head constraints

[0398]

[0399] In the formula, H s,i,n,t , is the generating head and head loss of the nth unit of power station i in time period t under scenario s.

[0400] 14) Head loss function constraint

[0401]

[0402] In the formula, a i 、b i are the head loss coefficient and loss constant of power station i.

[0403] 15) Unit dynamic characteristics constraints

[0404]

[0405] In the formula, It is the nonlinear relationship function between output, head and flow rate of the nth unit of power station i.

[0406] 16) Power balance constraints

[0407]

[0408] Where P 0,s,t is the channel output power in time period t under scenario s; P PV,s,t , are the output and abandoned power of photovoltaic power in time period t under scenario s; P WT,s,t , are the wind power output and abandoned power in time period t under scenario s; P L,s,t is the load in the domain at time period t under scenario s.

[0409] 17) Restrictions on power and water abandonment

[0410]

[0411] Where Pr{·} is the confidence expression; η is the confidence level.

[0412] 18) Transmission channel constraints

[0413]

[0414] In the formula, is the upper limit of the transmission channel power.

[0415] 19) Cornish-Fisher constraint

[0416] 3) A small-step-size iterative model solving method is proposed to transform the highly nonlinear model into a linear model through multiple iterative solutions. The specific steps are as follows:

[0417] 3.1) Initialize the number of iterations k = 0;

[0418] 3.2) Obtain the feasible solution of the kth iteration of the power delivered by the hydro-wind-solar hybrid system within the scheduling period under the typical wind-solar uncertainty scenario, and use the Cornish-Fisher series method to calculate the quantile of the kth iteration. The initial quantile of the system power delivered at time t with probability q is

[0419] 3.3) Based on the chain structure data flow, derive the partial derivative function of the output power quantile and calculate the partial derivative coefficient of the kth iteration:

[0420]

[0421]

[0422]

[0423]

[0424]

[0425]

[0426]

[0427] in, are the increments of the denormalized quantile, normalized quantile, normalized fifth-order origin moment, normalized fourth-order origin moment, normalized third-order origin moment, mean and standard deviation of the system output power at the kth iteration respectively; is the increment of the system power delivery under scenario s at time t at the k-th iteration; substituting the k-th iteration feasible solution of the system power delivery in S4.2 into it, the partial derivative coefficients can be obtained.

[0428] 3.4) Set the small step change range of the output power and linearize the Cornish-Fisher constraint based on the obtained partial derivative coefficients:

[0429]

[0430]

[0431]

[0432]

[0433]

[0434]

[0435]

[0436]

[0437]

[0438] Among them, δ is the range limit of the small step change of the output power. It should be noted that the other nonlinear constraints in the model, such as water level-storage capacity constraints, unit vibration zone constraints, unit dynamic characteristics constraints, etc., can be linearized by existing technologies, and the present invention will not make too many explanations.

[0439] 3.5) Replace the Cornish-Fisher constraint in the probability assessment model of the transmission capacity of the hydro-wind-solar hybrid system with the S4.3 constraint, and use the Gurobi solver to solve the model to obtain Further update is calculated through step 1) like If it holds true or the number of iterations k reaches the maximum number of iterations K, the solution ends; otherwise, set k=k+1 and return to step 3.2).

[0440] Under the 90% confidence interval, the model solution is as follows Figure 6 As shown in , the adjustable range of the system output power can be obtained. The system can report this range to the grid dispatching department, and the grid dispatching department will formulate a power generation dispatching plan that meets the grid operation needs within this range, thereby realizing the interaction between the water, wind and solar systems and the grid. Taking time 35 as an example, the probability distribution of the system output power changes with the number of iterations as shown in Figure 7 As shown in the figure, it can be seen that with the increase of the number of iterations, the distribution range of the system output power expands, which verifies the effectiveness of the proposed model.

Claims

1. A probabilistic evaluation method for the power output of a hydro-wind-solar hybrid system. It is characterized in that The following steps are involved: Step 1: Based on the power output of the hydro-wind-solar hybrid system under typical wind-solar uncertainty scenarios, the Cornish-Fisher series method is used to construct a mathematical expression for the output power quantile; Step 2: Based on the constructed mathematical expression of the transmission power quantile, with the goal of maximizing the upper and lower quantile range of the system's transmission power at each moment under a certain confidence level, a probability assessment model for the transmission power of the hydro-wind-solar hybrid system is established; Step 3: Aiming at the nonlinear Cornish-Fisher constraints in the probabilistic assessment model, a small-step iterative solution method based on chain-structured data flow is proposed to transform the highly nonlinear model into a linear model through multiple iterative solutions to obtain the probabilistic assessment results of the power delivered by the hydro-wind-solar hybrid system.

2. The method according to claim 1, It is characterized in that In step 1, the following sub-steps are used when constructing the explicit mathematical expression for the delivered power quantile: Step 1-1: Discretize the uncertainty of wind and solar power output into N based on the scenario method s A typical scene with uncertain scenery; Step 1-2: Get N at time t s A set of one-dimensional arrays composed of the transmission power of the water-wind-solar hybrid system in the scenario, specifically: Where P 0,t is the system output power at time t, P 0,1,t , P 0,2,t , They are scene 1, scene 2 and scene N at time t respectively. s The system output power under s is the total number of scenes; Step 1-3: Use the Cornish-Fisher series method to establish the mathematical relationship between the transmitted power and its quantile.

3. The method according to claim 2, It is characterized in that In steps 1-3, the fifth-order Cornish-Fisher method of cumulant is used to establish the mathematical relationship between the transmission power and its quantile. The specific expression is as follows: Among them, P 0,s,t is the system output power under scenario s at time t; μ P0,t , σ P0,t The system output power P at time t in all scenarios 0 The mean and standard deviation of P0,t，3 , μ P0,t,4 , μ P0,t,5 The system output power P at time t is 0 Normalized 3rd, 4th, and 5th-order origin moments of ξ; q is the quantile of the standard Gaussian distribution corresponding to probability q; based on the given probability q, P is the power delivered by the system at time t 0 The normalized quantile of p0,t,q is the power P delivered by the system at time t with probability q 0 The denormalized quantile of .

4. The method according to claim 1, It is characterized in that In step 2, the established probability assessment model for the power delivered by the water-wind-solar hybrid system is as follows: (1) Objective function Based on the displayed mathematical expression of the system power delivery quantile constructed in step 1, the objective function of the probability assessment model of the power delivery of the water-wind-solar hybrid system is to maximize the upper and lower quantile range of the system power delivery at each moment under a certain confidence level: Among them, T is the total number of scheduling periods; is the probability q at time t max The denormalized quantile of the system output power under ; is the probability q in period t min The denormalized quantile of the system output power under ; (2) Constraints 1) Water balance constraints Where V s,i,t+1 , V s,i,t are the storage capacities of the i-th hydropower station at time t+1 and time t under scenario s; I s,i,t is the inflow flow of the i-th hydropower station at time t under scenario s; Q s,i,t is the outflow of the ith hydropower station at time t under scenario s; Δt is the scheduling time interval; τ i-1 is the water flow delay time from the i-1th hydropower station to the i-th hydropower station; is the i-1th hydropower station in time period t-τ after considering water flow lag under scenario s i-1 The flow rate of R s,i,t is the interval flow between power station i-1 and power station i at time t under scenario s; is the power generation flow of the i-th hydropower station at time t under scenario s; is the abandoned water flow of the i-th hydropower station in time period t under scenario s; 2) Reservoir water level constraints In the formula, is the water level in front of the dam of the i-th hydropower station at time t under scenario s; Z i,t , are the lower limit and upper limit of the water level of the ith hydropower station at time t. The lower limit of the water level is usually the dead water level, the lower limit of the dispatching line, etc., and the upper limit of the water level is the flood limit water level, the normal high water level, or the upper limit of the dispatching line, etc. This constraint limits the fluctuation range of the water level; 3) Initial and final water level constraints In the formula, They are the initial water level during the dispatch period and the controlled water level at the end of the period; are the scheduling decision water levels of the i-th hydropower station at time 1 and time T under scenario s respectively; ΔZ is the allowable water level deviation at the end of the scheduling period to avoid affecting the water volume in the next scheduling period; 4) Outbound flow constraints In the formula, Q s,i,t is the outflow of the i-th hydropower station in time period t under scenario s; Q i,t , are the lower limit and upper limit of the outflow flow of the ith hydropower station in time period t. The lower limit of the outflow flow is usually determined by the ecological water demand, and the upper limit of the outflow flow is determined by the reservoir's discharge capacity. This constraint ensures the normal range of the reservoir's discharge flow. 5) Power station output constraints In the formula, is the output of the i-th hydropower station in time period t under scenario s; are the lower and upper limits of the output of the ith hydropower station in time period t, respectively; 6) Water level-reservoir capacity constraints In the formula, f i ZV (·) is the nonlinear relationship curve function between water level and storage capacity of the reservoir where power station i is located; 7) Tailwater level-discharge flow constraint In the formula, f i ZQ (·) is the nonlinear relationship curve function between the tailwater level and discharge flow of power station i; is the tailwater level of power station i in time period t under scenario s; 8) Unit output constraints Where P s,i,n,t , P i,n is the output and its upper and lower limits of the nth unit of power station i in time period t under scenario s; u s,i,n,t is the start-up and shutdown state variable of the nth unit of power station i in time period t under scenario s. It takes 1 to represent the start-up state and 0 to represent the shutdown state. The output of power station i can be expressed as: Where N i represents the number of units in power station i; 9) Unit power generation flow constraints In the formula, is the power generation flow and its upper and lower limits of the nth unit of power station i in time period t under scenario s; then the power generation flow of power station i can be expressed as: 10) Unit vibration zone constraints In the formula, are the upper and lower limits of the output of the kth vibration zone of the nth unit in power station i; 11) Unit start-up and shutdown duration constraints In the formula, is the startup operation variable of the nth unit in power station i in time period t, and takes 1 to indicate startup operation; is the shutdown operation variable of the nth unit of power station i in time period t, and takes 1 to indicate shutdown operation; are the minimum startup and shutdown durations of the nth unit in power station i, respectively; is the maximum number of startup times of the nth unit of power station i in the scheduling cycle; u s,i,n,t-1 is the start-up and shutdown state variable of the nth unit of power station i in time period t-1 under scenario s, and λ is the time mark variable; 12) Unit climbing constraints -ΔP i,n ≤P s,i,n,t+1 -P s,i,n,t ≤ΔP i,n Where ΔP i,n is the climbing capability of the nth unit in power station i, P s,i,n,t+1 It refers to the output of the nth unit of power station i in time period t+1 under scenario s; 13) Unit power generation head constraints In the formula, H s,i,n,t , is the generating head and head loss of the nth unit of power station i in time period t under scenario s, It refers to the water level in front of the dam of the i-th hydropower station at time t-1 under scenario s; 14) Head loss function constraint In the formula, a i 、b i are the head loss coefficient and loss constant of power station i; 15) Unit dynamic characteristics constraints In the formula, is the output-head-flow nonlinear relationship function of the nth unit of power station i; 16) Power balance constraints Where P 0,s,t is the channel output power in time period t under scenario s; P PV,s,t , are the output and abandoned power of photovoltaic power in time period t under scenario s; P WT,s,t , are the wind power output and abandoned power in time period t under scenario s; P L,s,t is the load in the domain at time period t under scenario s, and I refers to the total number of hydropower stations; 17) Restrictions on power and water abandonment Where Pr{·} is the confidence expression; η is the confidence level; 18) Transmission channel constraints In the formula, is the upper limit of the power transmission channel; 19) Nonlinear Cornish-Fisher Constraint The specific expression of the Cornish-Fisher constraint is the mathematical relationship between the transmission power and its quantile proposed in steps 1-3. The Cornish-Fisher constraint establishes the mathematical relationship between the system transmission power and its quantile in each scenario.

5. The method according to claim 1, It is characterized in that In step 3, a small-step iterative solution method based on a chain structure data stream is proposed for the nonlinear Cornish-Fisher constraint in the probability assessment model of the power delivered by the hydro-wind-solar hybrid system. The highly nonlinear model is converted into a linear model for multiple iterative solutions to obtain the probability assessment result of the power delivered by the hydro-wind-solar hybrid system. The method includes the following steps: Step 3-1: Initialize the number of iterations k = 0; Step 3-2: Obtain the feasible solution of the kth iteration of the power delivered by the hydro-wind-solar hybrid system in the dispatch period under the typical wind-solar uncertainty scenario, and use the Cornish-Fisher series method to calculate the quantile of the power delivered at the kth iteration. max With q min The quantile of the system output power is and Step 3-3: Based on the chain structure data flow, derive the partial derivative function of the output power quantile and calculate the partial derivative coefficient of the kth iteration: in, are the denormalized quantile, normalized quantile, normalized fifth-order origin moment, normalized fourth-order origin moment, normalized third-order origin moment, mean and standard deviation increments of the system output power under probability q at time t at the kth iteration, respectively; is the increment of the system output power under scenario s at time t at the k-th iteration; Substitute the feasible solution of the k-th iteration of the system output power in step 3-2 to obtain the partial derivative coefficients; Step 3-4: Set the small step change range of the transmit power. Based on the partial derivative function of the transmit power quantile derived in step 3-3 and the calculation results of the k-th iteration partial derivative coefficient, the following linearized Cornish-Fisher constraint is obtained: Among them, δ is the small step change range limit of the output power; are the denormalized quantiles of the power delivered by the wind-water hybrid system with probability q at time t at the k-th iteration and the k+1-th iteration respectively; are the normalized quantiles of the power delivered by the wind-water hybrid system with probability q at time t at the k-th iteration and the k+1-th iteration respectively; are the normalized fifth-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the normalized fourth-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the normalized third-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the standard deviations of the power delivered by the hydro-wind-solar hybrid system at time t in the k-th iteration and the k+1-th iteration respectively; are the mean values ​​of the power delivered by the hydro-wind-solar hybrid system at time t in the k-th iteration and the k+1-th iteration respectively; Step 3-5: Process and solve the probability assessment model of the hydro-wind-solar hybrid system output power at the kth iteration, and determine whether the model converges; Step 3-6: When the model iteration is completed, the obtained That is the probabilistic assessment result of the power delivered by the water, wind and solar systems.

6. The method according to claim 5, It is characterized in that In step 3-5, determine whether the model has converged. The specific steps are: Step 3-5-1: Replace all nonlinear Cornish-Fisher constraints in the probability assessment model of the power output of the hydro-wind-solar hybrid system with the linearized Cornish-Fisher constraints obtained in step 3-4, and All are regarded as decision variables, then the probability evaluation model of the power delivered by the hydro-wind-solar hybrid system becomes a mixed integer linear programming model; Step 3-5-2: To gradually increase the probability q during the iteration process max With q min The quantile range under the condition is as follows: Add the following constraints to the mixed integer linear programming model obtained in step 3-5-1: in, They are respectively the probability q at time t at the kth iteration max With q min The quantile of the power delivered by the lower system is a known quantity and has been obtained in step 3-2; They represent the probability q at time t in the k+1th iteration. max With q min The quantile of the power delivered by the wind-solar hybrid system in the upper and lower water level. Both are variables to be determined in the model of the kth iteration; Step 3-5-3: Use the Gurobi solver to solve the model obtained in step 3-5-2 to obtain the power delivered by the hydro-wind-solar hybrid system under scenario s at time t at the k+1th iteration. To avoid the error caused by linearization, we further use Update the actual quantile using the mathematical relationship between the transmitted power and its quantile proposed in steps 1-3 Step 3-5-4: Determine whether the model has converged iteratively. If any of the following two conditions is met, the model converges and the iterative solution ends; otherwise, set k = k + 1 and return to step 3-2; Condition 1: If the probability q max With q min The absolute value deviation of the transmitted power quantile results obtained in two consecutive iterations is less than the allowable error ε, that is, Condition 2: The number of iterations k reaches the maximum number of iterations K.

7. A small step iterative solution method based on chain structure data flow, It is characterized in that The method converts a highly nonlinear model into a linear model by multiple iterations to obtain a probability evaluation result of the power delivered by the water-wind-solar hybrid system; specifically, the method includes the following steps: Step S1: Initialize the number of iterations k=0; Step S2: Obtain the feasible solution of the kth iteration of the power delivered by the hydro-wind-solar hybrid system in the scheduling period under the typical wind-solar uncertainty scenario, and use the Cornish-Fisher series method to calculate the quantile of the power delivered at the kth iteration. max With q min The quantile of the system output power is and Step S3: Based on the data flow of the chain structure, derive the partial derivative function of the output power quantile and calculate the partial derivative coefficient of the kth iteration: in, are the denormalized quantile, normalized quantile, normalized fifth-order origin moment, normalized fourth-order origin moment, normalized third-order origin moment, mean and standard deviation increments of the system output power under probability q at time t at the kth iteration, respectively; is the increment of the system power delivery under scenario s at time t at the k-th iteration; the k-th iteration feasible solution of the system power delivery in step S2 is substituted into the system power delivery, and the partial derivative coefficients can be obtained; Step S4: Set the small step change range of the transmitted power. Based on the partial derivative function of the transmitted power quantile derived in step S3 and the calculation results of the k-th iteration partial derivative coefficient, the following linearized Cornish-Fisher constraint is obtained: Among them, δ is the small step change range limit of the output power; are the denormalized quantiles of the power delivered by the wind-water hybrid system with probability q at time t at the k-th iteration and the k+1-th iteration respectively; are the normalized quantiles of the power delivered by the wind-water hybrid system with probability q at time t at the k-th iteration and the k+1-th iteration respectively; are the normalized fifth-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the normalized fourth-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the normalized third-order origin moment of the power delivered by the wind-water hybrid system at time t with probability q at the k-th iteration and the k+1-th iteration respectively; are the standard deviations of the power delivered by the hydro-wind-solar hybrid system at time t in the k-th iteration and the k+1-th iteration respectively; are the mean values ​​of the power delivered by the hydro-wind-solar hybrid system at time t in the k-th iteration and the k+1-th iteration respectively; Step S5: Process and solve the probability assessment model of the power delivered by the water-wind-solar hybrid system at the kth iteration, and determine whether the model converges; Step S6: When the model iteration solution is completed, the obtained That is the probabilistic assessment result of the power delivered by the water, wind and solar systems.

8. The method according to claim 7, It is characterized in that In step S5, it is determined whether the model has converged. The specific steps are as follows: Step S5-1: Replace all nonlinear Cornish-Fisher constraints in the probability assessment model of the power delivered by the hydro-wind-solar hybrid system with the linearized Cornish-Fisher constraints obtained in step S4, and All are regarded as decision variables, then the probability evaluation model of the power delivered by the hydro-wind-solar hybrid system becomes a mixed integer linear programming model; Step S5-2: To gradually increase the probability q in the iterative process max With q min The quantile range under the condition is as follows: in, They are respectively the probability q at time t at the kth iteration max With q min The quantile of the power delivered by the lower system is a known quantity and has been obtained in step S2; They represent the probability q at time t in the k+1th iteration. max With q min The quantile of the power delivered by the wind-solar hybrid system in the upper and lower water level. Both are variables to be determined in the model of the kth iteration; Step S5-3: Use the Gurobi solver to solve the model obtained in step S5-2 to obtain the power delivered by the water-wind-solar hybrid system under the scenario s at time t at the k+1th iteration: To avoid the error caused by linearization, we further use Update the actual quantile by the mathematical relationship between the delivered power and its quantile Step S5-4: Determine whether the model has converged iteratively. If any of the following two conditions is met, the model has converged and the iterative solution ends; otherwise, set k=k+1 and return to step S2; Condition 1: If the probability q max With q min The absolute value deviation of the transmitted power quantile results obtained in two consecutive iterations is less than the allowable error ε, that is, Condition 2: The number of iterations k reaches the maximum number of iterations K.

9. An electronic device, It is characterized in that The method comprises a memory and a processor, wherein the processor and the memory are connected via a communication bus, and the memory is used to store a computer program, and when the computer program is executed by the processor, the method according to any one of claims 1 to 8 is implemented.

Images