Multi-degree-of-freedom quantitative evaluation method for flexibility of large-scale hydropower and new energy cluster

Abstract

The invention discloses a large-scale hydropower and new energy cluster flexibility multi-degree-of-freedom quantitative evaluation method, and the method comprises the steps: constructing a wind and light prediction model based on a BP neural network, inputting a training set into the wind and light prediction model, training the wind and light prediction model, and obtaining a trained wind and light prediction model; calculating a load prediction value based on normal distribution to obtain a load prediction curve; constructing an optimal economic dispatching model; and calculating the abandoned electricity quantity, the abandoned wind quantity, the abandoned light quantity and the load shedding quantity of the large-scale hydropower and new energy cluster by using the optimal economic dispatching model, setting a flexibility threshold value, calculating a flexibility index in the large-scale hydropower and new energy cluster, and evaluating the flexibility of the large-scale hydropower and new energy cluster. According to the method, a flexibility quantitative evaluation system is constructed from multiple degrees of freedom such as the adjustment speed, the adjustment range and the adjustment direction, quantitative evaluation of the flexibility of the new energy cluster is realized, and setting of the standby capacity of the power generation cluster is guided according to the evaluation result.

Description

Technical Field

[0001] The present invention relates to the field of renewable energy power generation technology, and in particular to a multi-freedom quantitative evaluation method for the flexibility of large-scale hydropower and renewable energy clusters. Background Art

[0002] Energy is the basic guarantee for the steady development of society and the foundation of human survival. With the impact of climate change and human activities, human society's demand for energy is increasing, and the problems of fossil energy depletion and environmental degradation are becoming more prominent. It has become a focus to tap the potential of new energy and accelerate the transformation of energy structure. Therefore, with the continuous advancement of the process of clean energy, large-scale hydropower and new energy clusters have been rapidly developed, and my country's wind power and photovoltaic installed capacity have increased exponentially. By the end of 2021, my country's total hydropower installed capacity was 390 million kW, accounting for 28.7% of the world; the total wind power installed capacity was 329 million kW, accounting for 39.9% of the world; the total photovoltaic installed capacity was 307 million kW, accounting for 36.1% of the world, ranking first in the world; at the same time, with the rapid development of new energy, my country's wind power and photovoltaic installed capacity are expected to grow to 480GW and 570GW in 2030, and to 1440GW and 2160GW in 2050. Wind and solar renewable energy have inherent properties such as uncertainty and randomness, and hydropower has the characteristics of abundance and drought. As the scale of new energy access to the power system gradually increases, the problems caused by the uncertainty and randomness of new energy are becoming more and more prominent. The abundance and drought characteristics of hydropower also have an impact on the quantification of cluster flexibility. The energy base relies on the rich water, wind and solar resources in the basin and the natural complementarity between the three resources. It makes full use of the hydropower stations in the basin to achieve the rapid response and flexible adjustment capabilities of the cluster, compensate for the uncertainty and randomness of wind and solar, smooth out the fluctuations of wind and solar, and realize the efficient development of the integration of "wind, light and water" in the basin; in summary, the study comprehensively considers the volatility and uncertainty of wind power and photovoltaic power and the impact of the abundance and drought characteristics of hydropower on the flexibility of the cluster, and accurately quantifies the evaluation of the flexible adjustment capabilities of the cluster. It is also of great significance to promote the transformation of the energy structure and accelerate the construction of energy bases.

[0003] At present, the quantitative evaluation of multi-freedom of large-scale hydropower and new energy cluster flexibility mainly has the following shortcomings:

[0004] (1) The research on flexibility assessment only considers the impact of a single power source or two resources, wind and solar. The impact of the combination of multiple resources, water, wind and solar, on cluster flexibility is not comprehensively considered.

[0005] (2) The quantitative evaluation indicators for the flexibility of large-scale hydropower and new energy clusters are relatively simple, and fail to achieve quantitative evaluation of cluster flexibility from a multi-dimensional perspective. Summary of the invention

[0006] In view of the above-mentioned deficiencies in the prior art, the present invention provides a multi-freedom quantitative evaluation method for the flexibility of large-scale hydropower and new energy clusters, which solves the problem of the single dimension of quantitative evaluation of the flexibility of large-scale hydropower and new energy clusters.

[0007] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0008] A multi-freedom quantitative evaluation method for the flexibility of large-scale hydropower and new energy clusters is provided, which includes the following steps:

[0009] S1: Collect historical wind and solar output data as a training set, build a wind and solar prediction model based on BP neural network, input the training set into the wind and solar prediction model, train the wind and solar prediction model, output the inter-layer relationship matrix of the wind and solar prediction model, and obtain a trained wind and solar prediction model;

[0010] S2: Input the data to be predicted into the trained wind-solar prediction model, output the wind-solar processing data, and calculate the load prediction value based on the normal distribution to obtain the load prediction curve;

[0011] S3: Obtain the amount of power abandonment and load shedding in large-scale hydropower and new energy clusters according to the load forecast curve, take the minimum loss cost of power abandonment and load shedding in large-scale hydropower and new energy clusters as the objective function, construct the mathematical model of the constraint conditions of large-scale hydropower and new energy clusters and the relationship between the unit power characteristic curve and the water head, and obtain the optimal economic dispatch model;

[0012] S4: Use the optimal economic dispatch model to calculate the amount of power abandonment, wind abandonment, solar abandonment and load shedding of large-scale hydropower and new energy clusters, set flexibility thresholds, calculate flexibility indicators within large-scale hydropower and new energy clusters, and evaluate the flexibility of large-scale hydropower and new energy clusters.

[0013] Furthermore, the inter-layer relationship model of the BP neural network in the wind and solar prediction model is:

[0014]

[0015] in, is the activation vector of the i-th neuron in the k-th layer of the BP neural network. The activation vector is composed of historical data of wind and solar power output. is the bias vector of the i-th neuron in the k-th layer of the BP neural network, the bias vector includes the bias data related to the historical data of wind and solar power output, j is the neuron in the k+1-th layer of the BP neural network, is the activation vector of the jth neuron in the k+1th layer of the BP neural network, δ(.) is the activation function, is the weight of the i-th neuron in the k-th layer;

[0016] Further, step S2 includes:

[0017] S21: input the data to be predicted into the trained wind-solar prediction model, output the wind-solar processing data, calculate the load prediction value based on the normal distribution, calculate the load prediction error according to the load prediction value and the load measured data, and obtain the load prediction error set;

[0018]

[0019] in, is the load forecast value, is the predicted correction value; ΔP load,t is the load forecast error;

[0020] S22: superimposing the load forecast errors in the load forecast error set onto the load curve of the large-scale hydropower and new energy cluster to obtain a load forecast curve.

[0021] Further, step S3 includes:

[0022] S31: obtaining the amount of abandoned electricity and load shedding in large-scale hydropower and new energy clusters according to the load forecast curve, and taking the minimum loss cost of the amount of abandoned electricity and load shedding in large-scale hydropower and new energy clusters as the objective function;

[0023] The objective function of the minimum loss cost of the amount of abandoned power and load shedding is:

[0024]

[0025] Among them, c E 、c L are the penalty amounts for unit power abandonment and unit load shedding, ΔT is the set time step, ΔP L,t is the load shedding amount at time t, ΔP E,t is the amount of power abandoned at time t, T is the dispatch period, E is the variable of power abandoned, and L is the variable of load shedding;

[0026] The amount of abandoned electricity includes abandoned wind and abandoned solar power. The objective function of the minimum loss cost of abandoned wind, abandoned solar and load shedding is:

[0027]

[0028] Among them, ΔP W,t is the wind power abandonment at time t, W is the wind power variable, ΔP S,t is the amount of abandoned light at time t, S is the photovoltaic variable, c W 、c S The penalty amounts for unit solar curtailment and unit wind curtailment respectively;

[0029] S32: Constraints for building large-scale hydropower and new energy clusters, including:

[0030] Power balance constraints:

[0031] P d,t =P allL,t ·(E d / E all );

[0032] Among them, P d,t is the total output of large-scale hydropower and new energy clusters at time t, d is the large-scale hydropower and new energy clusters in the study area, P allL,t is the total load of large-scale hydropower and new energy clusters at time t, allL is all clusters in the study area, L is the load of the cluster, E d is the corresponding power generation of large-scale hydropower and new energy clusters at time t, E all is the total load demand;

[0033] The upper and lower limit constraints of wind / solar abandonment:

[0034]

[0035] in, They are the upper limits of wind power and photovoltaic power forecast output respectively;

[0036] The upper and lower limits of the amount of abandoned electricity are as follows:

[0037]

[0038] Among them, ΔP D,t is the power curtailment variable at time t, and D is the total amount of wind and solar power curtailment; is the total output value of new energy at time t;

[0039] The upper and lower limits of load shedding are as follows:

[0040]

[0041] The constraints on hydropower stations in large-scale hydropower and renewable energy clusters include;

[0042] Hydraulic constraints of upstream and downstream reservoirs: I i,t =Q i-1,t-τ +R i,t ;in, is the inflow of the i-th reservoir at time t, is the power generation flow of the upstream i-1 hydropower station in the (t-τ) period, is the interval flow between hydropower station i-1 and hydropower station i, τ is the water flow lag time between hydropower station i-1 and hydropower station i;

[0043] Water balance constraint: V i,t =V i,t-1 +3600(I i,t -Q i,t )Δt; where V i,t is the storage capacity of the reservoir where hydropower station i is located at time t, V i,t-1 is the storage capacity of the reservoir where hydropower station i is located at time t-1, Q i,t is the outflow of the reservoir where hydropower station i is located at time t;

[0044] Water level constraints: in, Z i , Z i,t , The water level in front of the dam and the upper and lower limits of the water level of the reservoir where the hydropower station i is located at time t;

[0045] Initial and final water level constraints: Among them, Z i,begin ,Z i,end are the target control water levels of the reservoir where the hydropower station i is located at the initial and final stages of dispatch, Z i,0 is the water level of the reservoir where the hydropower station i is located at the initial stage of dispatch, Z i,T Water level at the end of the dispatch period;

[0046] Reservoir outflow constraints: in, Q i , are the upper and lower limits of the outflow of the i-th hydropower station, Q i,t is the actual outflow value of the i-th hydropower station at time t;

[0047] The constraints of the generating head of the hydropower station unit are: H i,t =(Z i,t-1 +Z i,t ) / 2-Z di,t ; Among them, H i,t is the water head of the i-th hydropower station at time t, Z di,t is the tailwater level of the i-th hydropower station at time t, Z i,t-1 is the water level of the i-th hydropower station at time t-1, Z i,t is the water level of the hydropower station at the i-th hydropower station at time t;

[0048] Water level-storage capacity constraint: V i,t =g i,zν (Z i,t ), where g i,zν (.) is the water level-storage capacity relationship function of the reservoir where the i-th hydropower station is located, Vi,t is the storage capacity of the reservoir where the i-th hydropower station is located at time t;

[0049] Tailwater level-discharge constraint condition: z di,t =g i,zu (Q i,t ), where g i,zu (.) is the tailwater level-discharge relationship function of the reservoir where the i-th hydropower station is located, Q i,t is the discharge of the hydropower station at the i-th hydropower station at time t;

[0050] Hydropower output - power generation head constraint: P i,g,t =g i,g,pqh (q i,g,t ,H i,g,t ), where g i,g,pqh (.) is the binary relationship function between the output of unit g of the i-th hydropower station and the power generation flow and power generation head, P i,g,t is the power output of unit g of the i-th hydropower station at time t, q i,g,t is the discharge of the hydropower station of unit g at time t, H i,g,t is the water head of unit g of the i-th hydropower station at time t;

[0051] Unit power characteristic constraints: P i,t =g i,pqh (Q i,t ,H i,g,t ), where g i,pqh (·) is the binary relationship function between the output of the ith hydropower station and the power generation flow and power generation head, P t i The power output of the ith hydropower station at time t;

[0052] S33: According to the maximum and minimum head interval of the i-th hydropower station The interval Split into three continuous water head sub-intervals and in, The power generation head of hydropower station i at any time satisfies the following constraints:

[0053]

[0054]

[0055] in, is the indicator variable of the kth water head sub-interval of unit g of hydropower station i at time t, k is the number of three consecutive water head sub-intervals, Indicates that the generating head of unit g is within the kth head sub-interval, is the head height of unit g of hydropower station i in the k-1 head sub-interval; The head height of unit g of hydropower station i in the kth head sub-interval;

[0056] S34: Get the water head height interval of the unit of hydropower station i When the water head height of hydropower station i When , the unit power characteristic curve under the average water head is selected as the representative curve, the representative curve is linearized, and the mathematical model of the relationship between the unit power characteristic curve and water head of hydropower station i is constructed:

[0057]

[0058]

[0059]

[0060]

[0061] in, is the functional relationship between the output and power generation flow of hydropower station i in the kth head sub-interval, is the relationship coefficient of the unit power characteristic curve of hydropower station i at the kth head section, the relationship coefficient Only one is equal to 1, f1 i,pq|h (.), They are the unit power characteristic curve and water head correlation function of the 1st, 2nd and 3rd sections respectively;

[0062] S35: Based on the objective function, the constraints of large-scale hydropower and new energy clusters, and the mathematical model of the relationship between the unit power characteristic curve and the water head, the optimal economic dispatch model is obtained.

[0063] Further, step S4 includes:

[0064] S41: Calculate the amount of power curtailment, wind curtailment, solar curtailment and load shedding of large-scale hydropower and new energy clusters using the optimal economic dispatch model, set flexibility thresholds, and calculate the probability P of insufficient upward / downward flexibility of large-scale hydropower and new energy clusters;

[0065]

[0066] Among them, x is the number of times the amount of power abandonment, wind abandonment, solar abandonment and load shedding is not 0 in the scheduling period T;

[0067] S42: Set the initial value μ of the intermediate quantity for the upward flexibility calculation, downward flexibility calculation, wind power flexibility calculation, and photovoltaic flexibility calculation of large-scale hydropower and new energy clusters up , =0, the abandoned wind volume ΔP solved by the optimal economic dispatch model W,t , Amount of abandoned light ΔP S,t and load shedding capacity ΔP L,t Calculate the expected indicators of upward flexibility and lack of downward flexibility;

[0068]

[0069]

[0070] Among them, η up is the median value of the calculation results of the insufficient upward flexibility index between / within clusters, up is the upward flexibility; η down-out It is the middle value of the calculation results of the inadequate downward flexibility index between clusters, and down-out is the downward flexibility between clusters; It is the middle value of the calculation result of the wind power index of the cluster with insufficient downward flexibility, down is the downward flexibility, and wind is wind power; is the middle value of the calculation result of the photovoltaic index of the cluster with insufficient downward flexibility, solar is photovoltaic, k is the total number of cycles, E UFNS,k It is the expected indicator result of insufficient upward flexibility between / within clusters after k cycles, and UFNS is insufficient upward flexibility; They are the expected index results of wind power and photovoltaic power after the cluster has undergone k cycles, and the DFNS is the insufficient flexibility for downward adjustment; E DFNS-out,k is the expected index result of insufficient downward flexibility between clusters after k cycles, DFNS-out is insufficient downward flexibility between clusters, E DFNS-in,k It is the expected indicator result of insufficient downward flexibility within the cluster after k cycles, and DFNS-in is insufficient downward flexibility within the cluster;

[0071] S43: Calculate the expected standard deviation coefficient V according to the insufficient flexibility index of large-scale hydropower and new energy clusters for upward or downward adjustment obtained in step S41 and step S42 σ ;

[0072]

[0073]

[0074] Among them, [X k ] is an array, including the upward / downward flexibility index value and the probability of insufficient upward / downward flexibility under the conditions of multiple cycles of large-scale hydropower and new energy clusters; For the array [X k], which represents the average value of the upward / downward adjustment flexibility index value and the probability of insufficient upward / downward adjustment flexibility. n is the array [X k ]The amount of data in is the standard deviation coefficient of the probability index of insufficient flexibility between clusters, P UFNS-out The probability of insufficient flexibility is adjusted upwards between clusters; is the standard deviation coefficient of the expected index of insufficient flexibility between clusters, E UFNS-out The flexibility to adjust upwards between clusters is insufficient to meet expectations; is the standard deviation coefficient of the probability index of insufficient flexibility within the i-th cluster, P UFNS-in Adjust the probability of insufficient flexibility downward within the cluster; is the standard deviation coefficient of the expected index of downward flexibility within the i-th cluster, E UFNS-in is the expectation of insufficient flexibility within the cluster; ε is the set cluster flexibility judgment threshold;

[0075] S44: If It is judged that the flexibility of large-scale hydropower and new energy clusters is good; otherwise, the flexibility is poor.

[0076] The beneficial effects of the present invention are:

[0077] 1. The large-scale hydropower and new energy cluster of the present invention includes three new energy forms: hydropower, photovoltaics, and wind power. At the same time, it comprehensively considers the volatility and uncertainty of wind power and photovoltaics as well as the impact of the wet and dry characteristics of hydropower on the flexibility of the cluster. By building a wind and solar prediction model based on the BP neural network, it simulates a variety of uncertainty scenarios and ultimately realizes the prediction of wind and solar output.

[0078] 2. The present invention provides a multi-freedom quantitative evaluation method for the flexibility of large-scale hydropower and new energy clusters, with the minimum cost of power abandonment and load shedding as the objective function, the minimum cost of wind and solar abandonment and load shedding within the cluster as the objective function, and the optimal economic dispatch model is solved based on the power balance constraints and the hydropower, wind and solar new energy power generation constraints to obtain the amount of new energy abandoned and the amount of load shedding in the cluster, and use these quantities to calculate the flexibility index, ultimately realizing the evaluation of the cluster flexibility.

[0079] 3. The research object of this invention is the new energy cluster, and it mainly studies the new power system based on new energy power generation such as hydropower, wind power, photovoltaic power, etc. It constructs a flexibility quantitative evaluation system from multiple degrees of freedom such as adjustment speed, adjustment range, and adjustment direction, and realizes the quantitative evaluation of the flexibility of the new energy cluster. At the same time, it guides the setting of the backup capacity of the power generation cluster according to the evaluation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 Schematic diagram of the process of the present invention

[0081] Figure 2 It is a schematic diagram of the analysis of the wind and solar power prediction curve and the measured curve of the present invention.

[0082] Figure 3 This is a schematic diagram of hydropower regulation during different water inflow periods of the present invention.

[0083] Figure 4 It is a schematic diagram of the flexibility assessment of large-scale hydropower and new energy clusters of the present invention.

[0084] Figure 5 This is a data set diagram of the historical wind power and photovoltaic output data of the four clusters in the embodiment.

[0085] Figure 6 This is a comparison chart between the load prediction curve and the measured curve.

[0086] Figure 7 It is the output curve diagram of 4 clusters in the embodiment.

[0087] Figure 8 It is a trend chart of the expected insufficiency index of the four clusters in the embodiment with the number of simulations. DETAILED DESCRIPTION

[0088] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0089] like Figure 1 As shown, a multi-freedom quantitative evaluation method for the flexibility of large-scale hydropower and new energy clusters includes the following steps:

[0090] S1: Collect historical data of wind and solar power output as a training set, build a wind and solar power prediction model based on BP neural network, input the training set into the wind and solar power prediction model, train the wind and solar power prediction model, output the inter-layer relationship matrix of the wind and solar power prediction model, and obtain a trained wind and solar power prediction model.

[0091] The inter-layer relationship model of the BP neural network in the wind and solar prediction model is:

[0092]

[0093] in, is the activation vector of the i-th neuron in the k-th layer of the BP neural network. The activation vector is composed of historical data of wind and solar power output. is the bias vector of the i-th neuron in the k-th layer of the BP neural network, the bias vector includes the bias data related to the historical data of wind and solar power output, j is the neuron in the k+1-th layer of the BP neural network, is the activation vector of the jth neuron in the k+1th layer of the BP neural network, δ(.) is the activation function, is the weight of the i-th neuron in the k-th layer;

[0094] When constructing a wind and solar prediction model, historical wind and solar output data is imported into the input end of the BP neural network, and then the weights and bias values ​​are counted in the output layer to generate an inter-layer relationship matrix, and finally the prediction is made in combination with the input data to be predicted.

[0095] S2: input the data to be predicted into the trained wind-solar prediction model, output the wind-solar processing data, and calculate the load prediction value based on the normal distribution to obtain the load prediction curve; step S2 specifically includes:

[0096] S21: input the data to be predicted into the trained wind-solar prediction model, output the wind-solar processing data, calculate the load prediction value based on the normal distribution, calculate the load prediction error according to the load prediction value and the load measured data, and obtain the load prediction error set;

[0097]

[0098] in, is the load forecast value, is the predicted correction value; ΔP load,t is the load forecast error;

[0099] S22: superimposing the load forecast errors in the load forecast error set onto the load curve of the large-scale hydropower and new energy cluster to obtain a load forecast curve.

[0100] like Figure 2 As shown in the figure, before point A, the predicted data is higher than the measured data. At this time, the system needs to reduce the output of hydropower in order to connect more wind and solar resources. If the system is not flexible enough to increase the output, it will cause load shedding events. Between points AB, the predicted data is lower than the measured data. In order to make up for the shortage of wind and solar power, the system will increase the output of hydropower. At this time, if the system does not reduce the resources enough, it will cause the abandonment of wind and solar power events. Near points A and B, the curve has a steep climb. If the system's climbing ability is insufficient, it will cause load shedding / abandonment of wind and solar power events. Therefore, analyzing the period of insufficient flexibility of large-scale hydropower and new energy clusters is crucial to quantitatively evaluate the flexibility of the system.

[0101] S3: Obtain the amount of abandoned power and load shedding in large-scale hydropower and new energy clusters according to the load forecast curve, take the minimum loss cost of abandoned power and load shedding in large-scale hydropower and new energy clusters as the objective function, construct the mathematical model of the constraint conditions of large-scale hydropower and new energy clusters and the relationship between the unit power characteristic curve and the water head, and obtain the optimal economic dispatch model.

[0102] Step S3 specifically includes:

[0103] S31: obtaining the amount of abandoned electricity and load shedding in large-scale hydropower and new energy clusters according to the load forecast curve, and taking the minimum loss cost of the amount of abandoned electricity and load shedding in large-scale hydropower and new energy clusters as the objective function;

[0104] The objective function of the minimum loss cost of the amount of abandoned power and load shedding is:

[0105]

[0106] Among them, c E 、c L are the penalty amounts for unit power abandonment and unit load shedding, ΔT is the set time step, ΔP L,t is the load shedding amount at time t, ΔP E,t is the amount of power abandoned at time t, T is the dispatch period, E is the variable of power abandoned, and L is the variable of load shedding;

[0107] The amount of abandoned electricity includes abandoned wind and abandoned solar power. The objective function of the minimum loss cost of abandoned wind, abandoned solar and load shedding is:

[0108]

[0109] Among them, ΔP W,t is the wind power abandonment at time t, W is the wind power variable, ΔP S,t is the amount of abandoned light at time t, S is the photovoltaic variable, c W 、c S The penalty amounts for unit solar curtailment and unit wind curtailment respectively;

[0110] S32: Constraints for building large-scale hydropower and new energy clusters, including:

[0111] Power balance constraints:

[0112] P d,t =P allL,t ·(E d / E all );

[0113] Among them, P d,tis the total output of large-scale hydropower and new energy clusters at time t, d is the large-scale hydropower and new energy clusters in the study area, P allL,t is the total load of large-scale hydropower and new energy clusters at time t, allL is all clusters in the study area, L is the load of the cluster, E d is the corresponding power generation of large-scale hydropower and new energy clusters at time t, E all is the total load demand;

[0114] The upper and lower limit constraints of wind / solar abandonment:

[0115]

[0116] in, They are the upper limits of wind power and photovoltaic power forecast output respectively;

[0117] The upper and lower limits of the amount of abandoned electricity are as follows:

[0118]

[0119] Among them, ΔP D,t is the power curtailment variable at time t, and D is the total amount of wind and solar power curtailment; is the total output value of new energy at time t;

[0120] The upper and lower limits of load shedding are as follows:

[0121]

[0122] The constraints on hydropower stations in large-scale hydropower and renewable energy clusters include;

[0123] Hydraulic constraints of upstream and downstream reservoirs: I i,t =Q i-1,t-τ +R i,t ; Among them, I t i is the inflow of the i-th reservoir at time t, is the power generation flow of the upstream i-1 hydropower station in the (t-τ) period, is the interval flow between hydropower station i-1 and hydropower station i, τ is the water flow lag time between hydropower station i-1 and hydropower station i;

[0124] Water balance constraint: V i,t =V i,t-1 +3600(I i,t -Q i,t )Δt; where V i,t is the storage capacity of the reservoir where hydropower station i is located at time t, V i,t-1 is the storage capacity of the reservoir where hydropower station i is located at time t-1, Q i,tis the outflow of the reservoir where hydropower station i is located at time t;

[0125] Water level constraints: in, Z i , Z i,t , The water level in front of the dam and the upper and lower limits of the water level of the reservoir where the hydropower station i is located at time t;

[0126] Initial and final water level constraints: Among them, Z i,begin ,Z i,end are the target control water levels of the reservoir where the hydropower station i is located at the initial and final stages of dispatch, Z i,0 is the water level of the reservoir where the hydropower station i is located at the initial stage of dispatch, Z i,T Water level at the end of the dispatch period;

[0127] Reservoir outflow constraints: in, Q i , are the upper and lower limits of the outflow of the i-th hydropower station, Q i,t is the actual outflow value of the i-th hydropower station at time t;

[0128] The constraints of the generating head of the hydropower station unit are: H i,t =(Z i,t-1 +Z i,t ) / 2-Z di,t ; Among them, H i,t is the water head of the i-th hydropower station at time t, Z di,t is the tailwater level of the i-th hydropower station at time t, Z i,t-1 is the water level of the i-th hydropower station at time t-1, Z i,t is the water level of the hydropower station at the i-th hydropower station at time t;

[0129] Water level-storage capacity constraint: V i,t =g i,zν (Z i,t ), where g i,zν (.) is the water level-storage capacity relationship function of the reservoir where the i-th hydropower station is located, V i,t is the storage capacity of the reservoir where the i-th hydropower station is located at time t;

[0130] Tailwater level-discharge constraint condition: z di,t =g i,zu (Q i,t ), where g i,zu (.) is the tailwater level-discharge relationship function of the reservoir where the i-th hydropower station is located, Q i,tis the discharge of the hydropower station at the i-th hydropower station at time t;

[0131] Hydropower output - power generation head constraint: P i,g,t =g i,g,pqh (q i,g,t ,H i,g,t ), where g i,g,pqh (.) is the binary relationship function between the output of unit g of the i-th hydropower station and the power generation flow and power generation head, P i,g,t is the power output of unit g of the i-th hydropower station at time t, q i,g,t is the discharge of the hydropower station of unit g at time t, H i,g,t is the water head of unit g of the i-th hydropower station at time t;

[0132] Unit power characteristic constraints: P i,t =g i,pqh (Q i,t ,H i,g,t ), where g i,pqh (·) is the binary relationship function between the output of the ith hydropower station and the power generation flow and power generation head, P t i The power output of the ith hydropower station at time t;

[0133] S33: According to the maximum and minimum head interval of the i-th hydropower station The interval Split into three continuous water head sub-intervals and in, The power generation head of hydropower station i at any time satisfies the following constraints:

[0134]

[0135]

[0136] in, is the indicator variable of the kth water head sub-interval of unit g of hydropower station i at time t, k is the number of three consecutive water head sub-intervals, Indicates that the generating head of unit g is within the kth head sub-interval, is the head height of unit g of hydropower station i in the k-1 head sub-interval; The head height of unit g of hydropower station i in the kth head sub-interval;

[0137] S34: Get the water head height interval of the unit of hydropower station i When the water head height of hydropower station i When , the unit power characteristic curve under the average water head is selected as the representative curve, the representative curve is linearized, and the mathematical model of the relationship between the unit power characteristic curve and water head of hydropower station i is constructed:

[0138]

[0139]

[0140]

[0141]

[0142] in, is the functional relationship between the output and power generation flow of hydropower station i in the kth head sub-interval, is the relationship coefficient of the unit power characteristic curve of hydropower station i at the kth head section, the relationship coefficient Only one is equal to 1, f1 i,pq|h (.), They are the unit power characteristic curve and water head correlation function of the 1st, 2nd and 3rd sections respectively;

[0143] S35: Based on the objective function, the constraints of large-scale hydropower and new energy clusters, and the mathematical model of the relationship between the unit power characteristic curve and the water head, the optimal economic dispatch model is obtained.

[0144] Hydropower regulation in different water inflow periods Figure 3 As shown in the figure, the amount of abandoned electricity is solved by the optimal economic dispatch model. The optimization target is to minimize the cost of abandoned electricity and load shedding. The optimization variables are the abandoned electricity of the wind and solar power stations in the cluster, the output and abandoned electricity of the hydropower station, the cluster load shedding amount and the relevant operating parameters of the hydropower station. The model is solved.

[0145] S4: Use the optimal economic dispatch model to calculate the amount of power abandonment, wind abandonment, solar abandonment and load shedding of large-scale hydropower and new energy clusters, set flexibility thresholds, calculate flexibility indicators within large-scale hydropower and new energy clusters, and evaluate the flexibility of large-scale hydropower and new energy clusters.

[0146] Step S4 specifically includes:

[0147] S41: Calculate the amount of power curtailment, wind curtailment, solar curtailment and load shedding of large-scale hydropower and new energy clusters using the optimal economic dispatch model, set flexibility thresholds, and calculate the probability P of insufficient upward / downward flexibility of large-scale hydropower and new energy clusters;

[0148]

[0149] Among them, x is the number of times the amount of power abandonment, wind abandonment, solar abandonment and load shedding is not 0 in the scheduling period T;

[0150] S42: Set the initial value μ of the intermediate quantity for the upward flexibility calculation, downward flexibility calculation, wind power flexibility calculation, and photovoltaic flexibility calculation of large-scale hydropower and new energy clusters up , =0, the abandoned wind volume ΔP solved by the optimal economic dispatch model W,t , Amount of abandoned light ΔP S,t and load shedding capacity ΔP L,t , and then calculate the expected index of upward flexibility and insufficient downward flexibility;

[0151]

[0152]

[0153] Among them, η up is the median value of the calculation results of the insufficient upward flexibility index between / within clusters, up is the upward flexibility; η down-out It is the middle value of the calculation results of the inadequate downward flexibility index between clusters, and down-out is the downward flexibility between clusters; It is the middle value of the calculation result of the wind power index of the cluster with insufficient downward flexibility, down is the downward flexibility, and wind is wind power; is the middle value of the calculation result of the photovoltaic index of the cluster with insufficient downward flexibility, solar is photovoltaic, k is the total number of cycles, E UFNS,k It is the expected indicator result of insufficient upward flexibility between / within clusters after k cycles, and UFNS is insufficient upward flexibility; They are the expected index results of wind power and photovoltaic power after the cluster has undergone k cycles, and the DFNS is the insufficient flexibility for downward adjustment; E DFNS-out,k is the expected index result of insufficient downward flexibility between clusters after k cycles, DFNS-out is insufficient downward flexibility between clusters, E DFNS-in,k It is the expected indicator result of insufficient downward flexibility within the cluster after k cycles, and DFNS-in is insufficient downward flexibility within the cluster;

[0154] S43: Calculate the expected standard deviation coefficient V according to the insufficient flexibility index of large-scale hydropower and new energy clusters for upward or downward adjustment obtained in step S41 and step S42 σ ;

[0155]

[0156]

[0157] Among them, [X k ] is an array, including the upward / downward flexibility index value and the probability of insufficient upward / downward flexibility under the conditions of multiple cycles of large-scale hydropower and new energy clusters; For the array [X k ], which represents the average value of the upward / downward adjustment flexibility index value and the probability of insufficient upward / downward adjustment flexibility. n is the array [X k ]The amount of data in is the standard deviation coefficient of the probability index of insufficient flexibility between clusters, P UFNS-out The probability of insufficient flexibility is adjusted upwards between clusters; is the standard deviation coefficient of the expected index of insufficient flexibility between clusters, E UFNS-out The flexibility to adjust upwards between clusters is insufficient to meet expectations; is the standard deviation coefficient of the probability index of insufficient flexibility within the i-th cluster, P UFNS-in Adjust the probability of insufficient flexibility downward within the cluster; is the standard deviation coefficient of the expected index of downward flexibility within the i-th cluster, E UFNS-in is the expectation of insufficient flexibility within the cluster; ε is the set cluster flexibility judgment threshold;

[0158] S44: If It is judged that the flexibility of large-scale hydropower and new energy clusters is good; otherwise, the flexibility is poor.

[0159] like Figure 4 As shown in Figure 1, the flexibility diagram of large-scale hydropower and new energy clusters. Flexibility is mainly analyzed from three degrees of freedom, namely, P min -P max The adjustment range represented by it, the unit output adjustment rate, that is, the adjustment speed, and the adjustment direction represented by up and down adjustment.

[0160] The large-scale hydropower and new energy cluster of the present invention includes three new energy forms: hydropower, photovoltaics, and wind power. At the same time, it comprehensively considers the volatility and uncertainty of wind power and photovoltaics as well as the impact of the wet and dry characteristics of hydropower on the flexibility of the cluster. By building a wind and solar prediction model based on the BP neural network, it simulates various uncertainty scenarios and ultimately realizes the prediction of wind and solar output.

[0161] The present invention provides a multi-freedom quantitative evaluation method for the flexibility of large-scale hydropower and new energy clusters, taking the minimum cost of power abandonment and load shedding as the objective function, taking the minimum cost of wind and solar abandonment and load shedding within the cluster as the objective function, and solving the optimal economic dispatch model based on the power balance constraints and the hydropower, wind and solar new energy power generation constraints, to obtain the amount of new energy abandoned and the amount of load shedding of the cluster, and use these quantities to calculate the flexibility index, ultimately realizing the evaluation of the cluster flexibility.

[0162] The research object of this invention is the new energy cluster. It mainly studies the new power system based on new energy power generation such as hydropower, wind power, photovoltaic power, etc., and constructs a flexibility quantitative evaluation system from multiple degrees of freedom such as adjustment speed, adjustment range, and adjustment direction. It realizes the quantitative evaluation of the flexibility of the new energy cluster, and guides the setting of the backup capacity of the power generation cluster according to the evaluation results.

[0163] This example takes four large-scale hydropower and new energy clusters in a certain area in the southwest as the research object. The area contains a total of 12 hydropower stations. The historical wind and solar output data of the four clusters in a year are input to train the wind and solar prediction model based on the BP neural network. The training set is as follows: Figure 5 The load forecast curves corresponding to the four clusters are shown in Figure 6 As shown in the figure, the output of the four clusters is as follows: Figure 7 shown.

[0164] A quantitative assessment of the multi-degree-of-freedom of cluster flexibility in a certain southwestern region can yield the values ​​of the inflexibility deficiency index within and between clusters as shown in Table 1.

[0165] Table 1 Indicators of inadequate flexibility between and within clusters in a certain region in Southwest China (probability, expectation)

[0166]

[0167] Based on this quantitative evaluation scheme, the variation trend of the inflexibility index within and between clusters with the number of simulations is obtained as follows: Figure 8 As shown. It can be seen that with the increase in the number of simulations, the inflexibility deficiency index tends to converge, which means that the technical solution of the present invention can fully cope with various power generation scenarios and can quantitatively evaluate the flexibility of the power generation system. It can be seen from Table 1 that the inflexibility deficiency and the expected inflexibility deficiency are both 0, indicating that the power generation system based on clusters has sufficient flexibility to increase, and there will be no load shedding events; then analyzing the internal flexibility of the four clusters, the inflexibility deficiency indicators for clusters 3 and 4 are both 0, which means that clusters 3 and 4 have sufficient flexible adjustment capabilities, and there will be no wind abandonment, solar abandonment or load shedding events; among the four clusters, cluster 2 has the worst flexibility.

[0168] The wind and solar power abandonment rates of each cluster are analyzed, and the wind and solar power abandonment rates of the four clusters are shown in Table 2.

[0169] Table 2 Wind and solar power abandonment rates in four clusters in a certain area in southwest China

[0170]

[0171] From Table 2, we can see that the wind and solar power abandonment rates of the four clusters are all maintained at a relatively low level. Cluster 2 has a slightly higher wind and solar power abandonment rate than clusters 1, 3, and 4. This is because cluster 2 has a poor downward adjustment and flexibility, resulting in more wind and solar power abandonment. At the same time, cluster 1 only abandoned solar power but not wind power. This is because hydropower cooperates more with wind power for regulation, so it is better at smoothing wind power fluctuations and the clusters have less wind abandonment; clusters 3 and 4 have better overall flexible adjustment capabilities, so there will be no power abandonment.

Claims

1. A multi-freedom quantitative evaluation method for the flexibility of large-scale hydropower and new energy clusters, characterized in that: The following steps are involved: S1: Collect historical wind and solar output data as a training set, build a wind and solar prediction model based on BP neural network, input the training set into the wind and solar prediction model, train the wind and solar prediction model, output the inter-layer relationship matrix of the wind and solar prediction model, and obtain a trained wind and solar prediction model; S2: Input the data to be predicted into the trained wind-solar prediction model, output the wind-solar processing data, and calculate the load prediction value based on the normal distribution to obtain the load prediction curve; S3: Obtain the amount of power abandonment and load shedding in large-scale hydropower and new energy clusters according to the load forecast curve, take the minimum loss cost of power abandonment and load shedding in large-scale hydropower and new energy clusters as the objective function, construct a mathematical model of the constraint conditions of large-scale hydropower and new energy clusters and the relationship between the unit power characteristic curve and the water head, and obtain the optimal economic dispatch model; S4: Use the optimal economic dispatch model to calculate the amount of power abandonment, wind abandonment, solar abandonment and load shedding of large-scale hydropower and new energy clusters, set flexibility thresholds, calculate flexibility indicators within large-scale hydropower and new energy clusters, and evaluate the flexibility of large-scale hydropower and new energy clusters.

2. The multi-freedom quantitative evaluation method for the flexibility of large-scale hydropower and new energy clusters according to claim 1 is characterized in that: The inter-layer relationship model of the BP neural network in the wind and solar prediction model is: in, is the activation vector of the i-th neuron in the k-th layer of the BP neural network. The activation vector is composed of historical data of wind and solar power output. is the bias vector of the i-th neuron in the k-th layer of the BP neural network, the bias vector includes the bias data related to the historical data of wind and solar power output, j is the neuron in the k+1-th layer of the BP neural network, is the activation vector of the jth neuron in the k+1th layer of the BP neural network, δ(.) is the activation function, is the weight of the i-th neuron in the k-th layer.

3. The multi-freedom quantitative evaluation method for the flexibility of large-scale hydropower and new energy clusters according to claim 2 is characterized in that: The step S2 comprises: S21: input the data to be predicted into the trained wind-solar prediction model, output the wind-solar processing data, calculate the load prediction value based on the normal distribution, calculate the load prediction error according to the load prediction value and the load measured data, and obtain the load prediction error set; in, is the load forecast value, is the predicted correction value; ΔP load,t is the load forecast error; S22: superimposing the load forecast errors in the load forecast error set onto the load curve of the large-scale hydropower and new energy cluster to obtain a load forecast curve.

4. The multi-freedom quantitative evaluation method for the flexibility of large-scale hydropower and new energy clusters according to claim 3 is characterized in that: The step S3 comprises: S31: obtaining the amount of abandoned electricity and load shedding in large-scale hydropower and new energy clusters according to the load forecast curve, and taking the minimum loss cost of the amount of abandoned electricity and load shedding in large-scale hydropower and new energy clusters as the objective function; The objective function of the minimum loss cost of the amount of abandoned power and load shedding is: Among them, c E 、c L are the penalty amounts for unit power abandonment and unit load shedding, ΔT is the set time step, ΔP L,t is the load shedding amount at time t, ΔP E,t is the amount of power abandoned at time t, T is the dispatch period, E is the variable of power abandoned, and L is the variable of load shedding; The amount of abandoned electricity includes abandoned wind and abandoned solar power. The objective function of the minimum loss cost of abandoned wind, abandoned solar and load shedding is: Among them, ΔP W,t is the wind power abandonment at time t, W is the wind power variable, ΔP S,t is the amount of abandoned light at time t, S is the photovoltaic variable, c W 、c S The penalty amounts for unit solar curtailment and unit wind curtailment respectively; S32: Constraints for building large-scale hydropower and new energy clusters, including: Power balance constraints: P d,t =P allL,t ·(AND d / AND all ); Among them, P d,t is the total output of large-scale hydropower and new energy clusters at time t, d is the large-scale hydropower and new energy clusters in the study area, P allL,t is the total load of large-scale hydropower and new energy clusters at time t, allL are all clusters in the study area, L is the load of the cluster, E d is the corresponding power generation of large-scale hydropower and new energy clusters at time t, E all is the total load demand; The upper and lower limit constraints of wind / solar abandonment: in, They are the upper limits of wind power and photovoltaic power forecast output respectively; The upper and lower limits of the amount of abandoned electricity are as follows: Among them, ΔP D,t is the power curtailment variable at time t, and D is the total amount of wind and solar power curtailment; is the total output value of new energy at time t; The upper and lower limits of load shedding are as follows: The constraints on hydropower stations in large-scale hydropower and renewable energy clusters include; Hydraulic constraints of upstream and downstream reservoirs: I i,t =Q i-1,t-τ +R i,t ;in, is the inflow of the i-th reservoir at time t, is the power generation flow of the upstream i-1 hydropower station in the (t-τ) period, is the interval flow between hydropower station i-1 and hydropower station i, τ is the water flow lag time between hydropower station i-1 and hydropower station i; Water balance constraint: V i,t =V i,t-1 +3600(I i,t -Q i,t )Δt; where V i,t is the storage capacity of the reservoir where hydropower station i is located at time t, V i,t-1 is the storage capacity of the reservoir where hydropower station i is located at time t-1, Q i,t is the outflow of the reservoir where hydropower station i is located at time t; Water level constraints: in, The water level in front of the dam and the upper and lower limits of the water level of the reservoir where the hydropower station i is located at time t; Initial and final water level constraints: Among them, Z i,begin ,Z i,end are the target control water levels of the reservoir where the hydropower station i is located at the initial and final stages of dispatch, Z i,0 is the water level of the reservoir where the hydropower station i is located at the initial stage of dispatch, Z i,T Water level at the end of the dispatch period; Reservoir outflow constraints: in, are the upper and lower limits of the outflow of the i-th hydropower station, Q i,t is the actual outflow value of the i-th hydropower station at time t; The constraints of the generating head of the hydropower station unit are: H i,t =(Z i,t-1 +Z i,t ) / 2-Z di,t ; Among them, H i,t is the water head of the i-th hydropower station at time t, Z di,t is the tailwater level of the i-th hydropower station at time t, Z i,t-1 is the water level of the i-th hydropower station at time t-1, Z i,t is the water level of the hydropower station at the i-th hydropower station at time t; Water level-storage capacity constraint: V i,t =g i,zν (Z i,t ), where g i,zν (.) is the water level-storage capacity relationship function of the reservoir where the i-th hydropower station is located, V i,t is the storage capacity of the reservoir where the i-th hydropower station is located at time t; Tailwater level-discharge constraint condition: z di,t =g i,zu (Q i,t ), where g i,zu (.) is the tailwater level-discharge relationship function of the reservoir where the i-th hydropower station is located, Q i,t is the discharge of the hydropower station at the i-th hydropower station at time t; Hydropower output - power generation head constraint: P i,g,t =g i,g,pqh (q i,g,t ,H i,g,t ), where g i,g,pqh (.) is the binary relationship function between the output of unit g of the i-th hydropower station and the power generation flow and power generation head, P i,g,t is the power output of unit g of the i-th hydropower station at time t, q i,g,t is the discharge of the hydropower station of unit g at time t, H i,g,t is the water head of unit g of the i-th hydropower station at time t; Unit dynamic characteristics constraints: P i,t =g i,pqh (Q i,t ,H i,g,t ), where g i,pqh (·) is the binary relationship function between the output of the i-th hydropower station and the power generation flow and power generation head, The power output of the ith hydropower station at time t; S33: According to the maximum and minimum head interval of the i-th hydropower station The interval Split into three continuous water head sub-intervals and in, The power generation head of hydropower station i at any time satisfies the following constraints: in, is the indicator variable of the kth water head sub-interval of unit g of hydropower station i at time t, k is the number of three consecutive water head sub-intervals, Indicates that the generating head of unit g is within the kth head sub-interval, is the head height of unit g of hydropower station i in the k-1 head sub-interval; The head height of unit g of hydropower station i in the kth head sub-interval; S34: Get the water head height interval of the unit of hydropower station i When the water head height of hydropower station i When , the unit power characteristic curve under the average water head is selected as the representative curve, the representative curve is linearized, and the mathematical model of the relationship between the unit power characteristic curve and water head of hydropower station i is constructed: in, is the functional relationship between the output and power generation flow of hydropower station i in the kth head sub-interval, is the relationship coefficient of the unit power characteristic curve of hydropower station i at the kth head section, the relationship coefficient Only one is equal to 1, They are the unit power characteristic curve and water head correlation function of the 1st, 2nd and 3rd sections respectively; S35: Based on the objective function, the constraints of large-scale hydropower and new energy clusters, and the mathematical model of the relationship between the unit power characteristic curve and the water head, the optimal economic dispatch model is obtained.

5. The multi-freedom quantitative evaluation method for the flexibility of large-scale hydropower and new energy clusters according to claim 3 is characterized in that: The step S4 comprises: S41: Calculate the amount of power curtailment, wind curtailment, solar curtailment and load shedding of large-scale hydropower and new energy clusters using the optimal economic dispatch model, set flexibility thresholds, and calculate the probability P of insufficient upward / downward flexibility of large-scale hydropower and new energy clusters; Among them, x is the number of times the amount of power abandonment, wind abandonment, solar abandonment and load shedding is not 0 in the scheduling period T; S42: Set the initial value μ of the intermediate quantity for the upward flexibility calculation, downward flexibility calculation, wind power flexibility calculation, and photovoltaic flexibility calculation of large-scale hydropower and new energy clusters up , =0, the abandoned wind volume ΔP solved by the optimal economic dispatch model W,t , Amount of abandoned light ΔP S,t and load shedding capacity ΔP L,t Calculate the expected indicators of upward flexibility and lack of downward flexibility; Among them, η up is the median value of the calculation results of the insufficient upward flexibility index between / within clusters, up is the upward flexibility; η down-out It is the middle value of the calculation results of the inadequate downward flexibility index between clusters, and down-out is the downward flexibility between clusters; It is the middle value of the calculation result of the wind power index of the cluster with insufficient downward flexibility, down is the downward flexibility, and wind is wind power; is the middle value of the calculation result of the photovoltaic index of the cluster with insufficient downward flexibility, solar is photovoltaic, k is the total number of cycles, E UFNS,k It is the expected indicator result of insufficient upward flexibility between / within clusters after k cycles, and UFNS is insufficient upward flexibility; They are the expected index results of wind power and photovoltaic power after the cluster has undergone k cycles, and the DFNS is the insufficient flexibility for downward adjustment; E DFNS-out,k is the expected index result of insufficient downward flexibility between clusters after k cycles, DFNS-out is insufficient downward flexibility between clusters, E DFNS-in,k It is the expected indicator result of insufficient downward flexibility within the cluster after k cycles, and DFNS-in is insufficient downward flexibility within the cluster; S43: Calculate the expected standard deviation coefficient V according to the insufficient flexibility index of large-scale hydropower and new energy clusters for upward or downward adjustment obtained in step S41 and step S42 σ ; Among them, [X k ] is an array, including the upward / downward flexibility index value and the probability of insufficient upward / downward flexibility under the conditions of multiple cycles of large-scale hydropower and new energy clusters; For the array [X k ], which represents the average value of the upward / downward adjustment flexibility index value and the probability of insufficient upward / downward adjustment flexibility. n is the array [X k ]The amount of data in is the standard deviation coefficient of the probability index of insufficient flexibility between clusters, P UFNS-out The probability of insufficient flexibility is adjusted upwards between clusters; is the standard deviation coefficient of the expected index of insufficient flexibility between clusters, E UFNS-out The flexibility to adjust upwards between clusters is insufficient to meet expectations; is the standard deviation coefficient of the probability index of insufficient flexibility within the i-th cluster, P UFNS-in Adjust the probability of insufficient flexibility downward within the cluster; is the standard deviation coefficient of the expected index of downward flexibility within the i-th cluster, E UFNS-in is the expectation of insufficient flexibility within the cluster; ε is the set cluster flexibility judgment threshold; S44: If It is judged that the flexibility of large-scale hydropower and new energy clusters is good; otherwise, the flexibility is poor.

Images