Power balance risk index evaluation method, device, equipment and medium

Abstract

The present invention discloses a method for evaluating power balance risk indicators, which relates to the technical field of distribution network optimization and is used to solve the problem of single controllable resources in existing risk control models. The method includes: constructing a time-series characteristic power model based on historical source-load data and constructing multi-type adjustable resource models; evaluating the probability that the balance equation in the source-network-load-storage power model does not hold and using it as a risk indicator; constructing a risk indicator evaluation model with the goals of minimum risk and minimum adjustment cost; constructing a multi-objective distributionally robust model as the worst-case risk indicator to complete the construction of the power balance risk indicator evaluation model; and solving the model through truncated normal distribution to obtain the risk assessment result. The present invention also discloses a power balance risk indicator evaluation device, an electronic device, and a computer storage medium. Through multi-dimensional modeling and solving the model in combination with truncated normal distribution, the present invention realizes multi-dimensional risk assessment of the power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution network optimization, and in particular to a data-driven power system power balance risk indicator evaluation method, device, equipment and medium. Background Art

[0002] With the large-scale access of renewable energy sources to distribution networks, the intermittent and volatile output of renewable energy sources will put distribution networks in an uncertain state, and operating parameters will be more likely to exceed limits. In addition, the access of renewable energy distributed power sources has changed the topological structure of distribution networks, making power flow distribution more complex, and has had a significant impact on the voltage, active power and other state quantities of distribution networks, increasing the risk level of the power system, which also puts higher requirements on the risk control of source-grid-load interaction.

[0003] To ensure that risks are controllable, power balance risk constraints need to include two aspects: safety constraints and zero-carbon constraints. Safety constraints refer to the risk of power grid operation when the maximum grid supply capacity of the upper-level power grid is less than the capacity required for the operation of the power grid at this level, that is, the thermal stability constraint risk of the power grid; zero-carbon constraints refer to the risk of self-sufficiency or transmission of internal load energy within the power grid at this level, which cannot guarantee zero-carbon operation.

[0004] Risk control needs to reduce the possibility of high risk or the severity of consequences by implementing preventive control and emergency management measures that are consistent with the assessment results, based on risk assessment and early warning, and provide decision-making solutions for operation and dispatch personnel. The risk control model established by traditional risk control methods has a single controllable resource, making it difficult to coordinate the use of all resources on the demand side and the power supply side, making it difficult to accurately assess the risks of the power grid, and unable to provide a reliable reference for subsequent power grid dispatching. Summary of the invention

[0005] In order to overcome the shortcomings of the prior art, one of the purposes of the present invention is to provide a power balance risk index assessment method, which realizes the risk assessment of the power grid by considering the uncertainty of multiple types of adjustable resources and load probability distribution for modeling, and using truncated normal distribution to transform and solve the random measure in the objective function.

[0006] One of the purposes of the present invention is achieved by the following technical solution:

[0007] A power balance risk indicator evaluation method comprises the following steps:

[0008] Build a time-series characteristic power model based on historical source-load data, and build multiple types of adjustable resource models based on adjustable resources in the power system;

[0009] Assess the probability that the balance equation in the source-grid-load-storage power model is not valid and use it as a risk indicator;

[0010] Construct a risk index evaluation model with the goal of minimizing risk and adjustment cost;

[0011] Construct a multi-objective distributionally robust model as the worst-case risk index to complete the construction of the power balance risk index evaluation model;

[0012] Solve the model by truncated normal distribution to obtain the risk assessment result.

[0013] The second object of the present invention is achieved by the following technical solutions:

[0014] A power balance risk index evaluation device, which includes:

[0015] A construction module for constructing a time-series characteristic power model based on the historical data of source and load, and constructing multi-type adjustable resource models based on the adjustable resources in the power system;

[0016] An index construction module for evaluating the probability that the balance equation in the source-network-load-storage power model does not hold and using it as a risk index; constructing a risk index evaluation model with the goal of minimizing risk and adjustment cost; constructing a multi-objective distributionally robust model as the worst-case risk index to complete the construction of the power balance risk index evaluation model;

[0017] An analysis module for solving the model to obtain the risk assessment result.

[0018] The third object of the present invention is to provide an electronic device for implementing the first object of the invention, which includes a processor, a storage medium, and a computer program. The computer program is stored in the storage medium, and when the computer program is executed by the processor, the above-mentioned power balance risk index evaluation method is realized.

[0019] The fourth object of the present invention is to provide a computer-readable storage medium storing the first object of the invention, on which a computer program is stored, and when the computer program is executed by the processor, the above-mentioned power balance risk index evaluation method is realized.

[0020] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0021] Based on the risk assessment theory of the power grid, the present invention makes full use of the massive historical data of the power sources and loads in the power grid system and adjustable resources to construct a time-series characteristic power model and an adjustable resource model. By evaluating the probability that the power balance equation of the power source, grid, load, and storage in the power system does not hold, it serves as a risk indicator. Considering the risk of reduced adjustment ability of the regional power grid, a multi-objective risk indicator evaluation model with the objectives of minimum risk and minimum adjustment cost is constructed. In addition, the present invention also considers the uncertainty of the probability distribution and proposes a multi-objective distributionally robust model to evaluate the risk indicators in the worst-case scenario. By using the truncated normal distribution to transform and solve the random measure in the objective function, and by utilizing all the resources on the demand side and the supply side of the power grid system and considering the assessment of various risks, a multi-dimensional risk assessment of the power grid is achieved, which can provide a reliable reference basis for the dispatching of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a flowchart of a power balance risk indicator evaluation method in Embodiment 1;

[0023] Figure 2 is a schematic diagram of the risk level of the power balance safety constraint of the power source, grid, and load in Embodiment 2;

[0024] Figure 3 is a schematic diagram of the risk level of the power balance safety constraint of the power source, grid, and load in Embodiment 2;

[0025] Figure 4 is a schematic diagram of the zero-carbon constraint risk of the power balance of the power source, grid, and load in Embodiment 2;

[0026] Figure 5 is a schematic diagram of the risk level of the zero-carbon constraint of the power balance of the power source, grid, and load in Embodiment 2;

[0027] Figure 6 is a Pareto frontier diagram of the multi-objective optimization in Embodiment 2;

[0028] Figure 7 is a schematic diagram of the risk level corresponding to the multi-objective optimization in Embodiment 2;

[0029] Figure 8 is a structural block diagram of the power balance risk indicator evaluation device in Embodiment 3;

[0030] Figure 9 is a structural block diagram of the electronic device in Embodiment 4. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] The present invention will be described in more detail below with reference to the accompanying drawings. It should be noted that the description of the present invention with reference to the accompanying drawings below is only illustrative and not restrictive. Various different embodiments can be combined with each other to form other embodiments not shown in the following description.

[0032] Embodiment 1

[0033] Embodiment 1 provides a power balance risk index evaluation method, which studies a data-driven power system power balance risk index evaluation method based on the risk evaluation theory of the power grid.

[0034] The power balance risks involved in this embodiment mainly include two aspects:

[0035] 1. Safety constraints: The maximum supply capacity of the upper-level power grid is less than the capacity required for the operation of the power grid at this level, which creates risks in the operation of the power grid, namely, the thermal stability constraint risk of the power grid.

[0036] 2. Zero-carbon constraint: The power grid at this level can be self-sufficient or supply electricity to ensure zero carbon. The risk refers to the risk of not being able to ensure zero-carbon operation, that is, using electricity from the upper-level power grid.

[0037] In order to deal with the above two power balance risks, this embodiment establishes multiple models through multiple dimensions and comprehensively considers various risks that may arise to evaluate the risk indicators of the power grid system.

[0038] Specifically, firstly, considering the uncertainty of the probability distribution of multiple types of distributed energy and loads, a time series characteristic power model in the form of moment information fuzzy sets is constructed based on massive historical data of sources and loads, and multiple types of adjustable resources in the system are modeled; secondly, based on the operation principle of random variables, the random measurement method is used to evaluate the probability that the power balance equation of source, grid, load and storage in the power system is not valid, as a risk indicator; further, considering the regional power grid regulation ability to reduce risks, a multi-objective risk indicator evaluation model with minimum risk and minimum regulation cost as the goals is proposed; on this basis, considering the uncertainty of the probability distribution of uncertain factors, a multi-objective distributed robust model is proposed to evaluate the risk indicators under the worst conditions; finally, the truncated normal distribution is used to transform and solve the random measure in the objective function, and the ε constraint method is used to solve the multi-objective optimization problem involved in the risk indicator evaluation considering regulation measures.

[0039] According to the above principles, please refer to Figure 1 As shown, a power balance risk indicator evaluation method includes the following steps:

[0040] S1. Build a time series characteristic power model based on source and load historical data, and build multiple types of adjustable resource models based on adjustable resources in the power system;

[0041] S1 considers the uncertainty of probability distribution of multiple types of distributed energy and load, builds a time series characteristic power model in the form of moment information fuzzy set based on massive historical data of sources and loads, and models multiple types of adjustable resources in the system.

[0042] Construct a time-series characteristic power model based on the historical data of power sources and loads. The construction of the power model includes:

[0043] Based on the normal distribution probability model and the time-series characteristic curves of the distributed energy output in the municipal power grid, for the photovoltaic and wind power system i, at each moment t, establish a time-series characteristic fuzzy set of distributed energy output based on data-driven moment information:

[0044]

[0045] where D DG1 and D DG2 are the fuzzy sets of the random variable ξ DG,i,t of the distributed power source and the mean sequence μ DG,i,t and the variance sequence σ DG,i,t of the distributed power source output curve, respectively. Ω ξDG is the distribution set of the random variable ξ DG,i,t . P(·) is the probability function, indicating that the probability that the distribution of the random variable ξ DG,i,t belongs to is 1. E(·) is the expectation function, and μ DG,i,t are the upper and lower limits of the mean, respectively, and are the upper and lower limits of the variance, respectively. μ DG,pre,i,t and σ DG,pre,i,t are the predicted values of the mean and variance, respectively. α and β are the allowable deviation coefficients for controlling the conservatism of the fuzzy set. N DG is the number of distributed power source configurations;

[0046] Construct a hydropower power model, including:

[0047] The output of the hydropower station HT is jointly determined by the generation efficiency η h , the specific variable h of the hydropower station, the water head H ht at time t, and the generation flow rate Q ht . The hydropower conversion function is:

[0048] where η h is the generation efficiency, H ht is the water head, Q ht is the generation flow rate, and g is the hydropower conversion coefficient, usually taken as 9.81. Model the actual value of the generation flow rate Q ht as a random variable subject to a normal distribution:

[0049]

[0050] where μ ht is the generation flow rate The mean value of σ ht is the power generation flow The coefficient of variation of the standard deviation;

[0051] Considering the uncertainty of its mean value and standard deviation, a fuzzy set of hydropower output is constructed:

[0052]

[0053] Among them, D HT1 and D HT2 are the fuzzy set of the random variable and the mean value of hydropower output and variance of the uncertainty set; is the distribution set of the random variable , P(·) is the probability function, indicating that the distribution of the random variable belongs to has a probability of 1, and E(·) is the expectation function; and are the upper and lower limits of the mean value respectively, and are the upper and lower limits of the variance respectively; and are the predicted values of the mean value and variance respectively, α and β are the allowable deviation coefficients, and N HT is the number of hydropower configurations;

[0054] Among them, different types of hydropower stations have different head regulation characteristics. There are three common types of hydropower stations: run-of-river hydropower stations, adjustable hydropower stations, and cascade hydropower stations.

[0055] For a run-of-river hydropower station, it has no reservoir capacity regulation ability, that is, the head H of the hydropower station h at time t ht is a fixed value h h , satisfying H ht =h h ;

[0056] For an adjustable hydropower station, it has good regulation ability, and the power generation head is determined by the upstream and downstream water levels, that is, the head changes with the reservoir capacity. The power generation head of the adjustable power station is a linear function of the reservoir capacity of the hydropower station, that is:

[0057] H ht =h 0,h +α h V ht , where h 0,h and α h are constants respectively, and V ht represents the reservoir capacity of the hydropower station. The regulation range of the reservoir capacity of the hydropower station is: and They are the upper and lower bounds of reservoir capacity regulation respectively.

[0058] For cascade hydropower stations, which are a special type of adjustable hydropower stations, the upstream and downstream adjustable hydropower stations are closely linked and significantly affect each other. The incoming water volume of the current adjustable hydropower station includes the natural incoming water volume and the power generation flow of the upper-level hydropower station. In addition, there is a certain distance between the upstream and downstream adjustable hydropower stations, so the time-delay effect of the water flow needs to be considered. To sum up, the regulation of cascade hydropower stations satisfies:

[0059] Among them, τ h is the water flow time-delay of hydropower station h; is the power generation flow of the upper-level hydropower station h-1 at time t-τ h moment.

[0060] Construct a load power model, including: using the time-series characteristic curve prediction data of loads in a large number of prefecture-level power grids, considering the differences in the time-series characteristics of multiple types of loads and the uncertainty of their means and variances, and establishing a fuzzy set of the time-series characteristics of load output based on data-driven moment information. The loads considered in the load power model include residential loads, commercial loads, industrial loads, and administrative loads. For load i, it satisfies:

[0061]

[0062]

[0063] Among them, D L1 and D L2 are the fuzzy set of the random variable and the uncertain set of the mean and variance of the load output respectively; is the distribution set of the random variable , P(·) is the probability function, indicating that the distribution of the random variable belongs to with a probability of 1, and E(·) is the expectation function; and are the upper and lower limits of the mean respectively, and are the upper and lower limits of the variance respectively; and are the predicted values of the mean and variance respectively, α and β are the allowable deviation coefficients respectively, and N L is the number of loads.

[0064] In S1, construct multi-type adjustable resource models according to the adjustable resources in the power system, including:

[0065] Combined with the energy constraint of hydrogen energy storage, degradation cost, and scheduling period, a long-term hydrogen energy storage model is constructed. The hydrogen energy storage needs to meet the power constraint and energy constraint. The power constraint needs to meet the fuel cell power limit, electrolyzer power limit, and the constraint that charging and discharging cannot be carried out simultaneously. Specifically, the output power constraint of hydrogen energy storage is satisfied as follows:

[0066]

[0067] Among them, FC represents the fuel cell, and ED represents the electrolyzer. respectively represent the upper and lower bounds of the output power of the fuel cell and the electrolyzer. and are binary variables indicating whether the fuel cell and the electrolyzer are working at time t in the m-th month's scheduling period of hydrogen energy storage respectively. is the output power of hydrogen energy storage at time t in the m-th month's scheduling period, and η ED and η FC are the working efficiencies of the electrolyzer and the fuel cell respectively;

[0068] The energy constraint needs to meet the hydrogen energy storage capacity recurrence limit and the hydrogen energy storage capacity limit at each moment:

[0069]

[0070] Among them, is the capacity of hydrogen energy storage at time t in the m-th month's scheduling period, and are the upper and lower limits of the capacity.

[0071] The operating cost of hydrogen energy storage includes the sum of the degradation cost of the fuel cell and the degradation cost of the electrolyzer.

[0072] The degradation cost of the fuel cell is satisfied as follows:

[0073]

[0074] Among them, R0 is the initial value of the ohmic resistance of the FC, and C FC_cap is the investment cost of the fuel cell, is the utilization rate of the fuel cell, U ocv is the open circuit voltage of the FC, I rated is the rated current of the FC, θ t is the temperature of the FC, R t is the ohmic resistance at time t, S A is the area of the battery.

[0075] The degradation cost of the electrolyzer is satisfied as follows:

[0076] Among them, C ED_capis the investment cost of the electrolysis device, and Δt is the scheduling period. is the life cycle of the FC.

[0077] The total cost of hydrogen storage satisfies:

[0078] where and are the degradation costs of the fuel cell and the electrolysis device, respectively.

[0079] In addition, for the power balance regulation on a longer time scale with a smaller time level, the monthly regulation effect of hydrogen storage is mainly considered. The method is as follows: According to the predicted difference between the monthly grid power generation and power consumption (energy gap or surplus), with the goal of minimizing the total adjusted difference, optimize the charging / discharging amount of the hydrogen storage system for each month, and then evenly distribute it to each day.

[0080] For the daily scheduling period, the hydrogen energy storage needs to satisfy the following constraints:

[0081] where the number of daily scheduling time periods is T 1， is the energy at the end of the m-th monthly scheduling period of the hydrogen energy storage.

[0082] The optimization of the monthly hydrogen storage energy change satisfies:

[0083]

[0084] where E net represents the change in the energy of the monthly hydrogen storage, represents the load output, represents the wind power output, represents the hydropower output, represents the photovoltaic output, represents the output of the hydrogen energy storage.

[0085] Construct a short-time primary electricity energy storage model that satisfies:

[0086]

[0087] SOC min ≤ SOC t ≤ SOC max

[0088]

[0089] where SOC t is the state of charge of the electricity energy storage, SOC max and SOC min are the upper and lower limit constraints of the state of charge, and Et is the energy of the electrical energy storage at time t, and is the charge-discharge power of the electrical energy storage at time t, is the upper limit of the charge-discharge power of the electrical energy storage;

[0090] The cost of the electrical energy storage includes construction cost, operation and maintenance cost, and degradation cost:

[0091] C ESS = α ESS M ESS + β ESS M ESS

[0092] M ESS = c2P ESS.max + c3E r

[0093]

[0094] C ESS is the cost of the electrical energy storage, M ESS is the one-time investment cost of the energy storage device, P ESS.max is the maximum charge-discharge power of the energy storage, E r is the capacity of the energy storage, r is the discount rate, n is the operation years of the energy storage, α ESS is the equivalent annual value coefficient of the electrical energy storage, β ESS is the operation cost coefficient, c2 and c3 are constants.

[0095] Build a regulation means model for adjustable hydropower, wind and solar abandonment, and load demand response. The reservoir capacity and flow satisfy:

[0096]

[0097] The reduction amount satisfies:

[0098]

[0099]

[0100] Among them, are the reduction amounts of wind and solar respectively, are the maximum reduction degrees of wind and solar respectively; and build the wind and solar reduction cost and total incentive cost functions.

[0101] Among them, the cost of wind and solar reduction is as follows:

[0102] Among them, λ CUR is the cost per unit power of wind and solar reduction, is the power of wind and solar reduction, and ΔT is the scheduling period.

[0103] The load demand response satisfies:

[0104]

[0105] Among them, is the reduced load, represents the maximum reduction degree.

[0106] The incentive cost of the i-th load is The total incentive cost satisfies:

[0107] Among them, the total incentive cost of the load, is the incentive cost per unit power of the load demand response, is the i-th load demand response volume, N Y is the number of loads for demand response.

[0108] S2. Evaluate the probability that the power balance equation in the source-network-load-storage power model does not hold, and use it as a risk index;

[0109] In S2, the risk index evaluation idea is as follows:

[0110] Step 1: Obtain the probabilistic power models of multiple types of sources (wind, light, water) and loads in the region considering the time series characteristics;

[0111] Step 2: Construct the multi-time-section power balance equations for various types of source-load-storage resources in the region;

[0112] Step 3: Based on the operation principle of random variables, use the random measure method to evaluate the probability that the balance equation does not hold, and calculate the risk index of the regional power balance.

[0113] According to the above principle, evaluate the probability that the power balance equation in the source-network-load-storage power model does not hold, and use it as a risk index, satisfying:

[0114]

[0115] Among them, represents the net load after regulation; is the load output of node i in substation j; is the demand response output; is the wind power output; is the wind power reduction; is the hydropower output after regulation; is the photovoltaic output; is the photovoltaic reduction; is the output of hydrogen storage; is the output of electrical energy storage; K sub,jis the substation capacity-load ratio; is the capacity of the j-th substation; is the power factor of the substation.

[0116] S3. Construct a risk index evaluation model with the goal of minimizing risk and regulation cost;

[0117] S3. Considering the risk of reducing the regulation ability of the regional power grid, a multi-objective risk index evaluation model with the goal of minimizing risk and regulation cost is proposed.

[0118] Among them, a power balance risk index is constructed with the goal of minimizing risk, satisfying:

[0119]

[0120] Among them, represents the net load after regulation; is the probability; is the demand response output; is the wind power curtailment; V ht is the hydropower reservoir capacity; is the photovoltaic curtailment; is the output of the hydrogen storage; is the output of the electrical energy storage; K sub,j is the substation capacity-load ratio; S sub,j is the capacity of the j-th substation; is the power factor of the substation;

[0121] An energy storage risk index evaluation model is constructed with the goal of minimizing regulation cost, satisfying:

[0122] min[C ESS +C HHBES_deg +C DR +C CUR . Among them, C ESS is the cost of electrical energy storage,

[0123] C HHBES_deg is the cost of hydrogen energy storage, C DR is the demand response cost, C CUR is the penalty cost for wind and solar curtailment.

[0124] S4. Construct a multi-objective distributionally robust model as the worst-case risk index to complete the construction of the power balance risk index evaluation model;

[0125] S4 Specifically, construct a multi-objective distributionally robust model as the worst-case risk index, and the goals include the uncertainty of the probability model and minimizing the cost. The goal of the probability model uncertainty satisfies:

[0126]

[0127] The minimized cost includes the costs of two types of energy storage, demand response cost, wind power, and PV curtailment cost, satisfying: min[C ESS +C HHBES_deg +C DR +C CUR . Wherein, is the wind power output; is the PV output; is the hydropower output; is the load output; represents the net load after regulation; Pr{·} is the probability; is the demand response output; is the wind power curtailment; V ht is the hydropower reservoir capacity; is the PV curtailment; is the output of hydrogen energy storage; is the output of electrical energy storage; K sub,j is the substation capacity-load ratio; S sub,j is the capacity of the jth substation; is the power factor of the substation. C ESS is the electrical energy storage cost, C HHBES_deg is the hydrogen energy storage cost, C DR is the demand response cost, C CUR is the penalty cost for wind and PV curtailment.

[0128] S5. Solve the model through truncated normal distribution to obtain the risk assessment result.

[0129] S5 specifically includes transforming the function in the model through truncated normal distribution, and the absolute value of the net load of the function satisfies:

[0130]

[0131] μ net,j =μ L -μ WT -μ HT -μ PV -μ Hy σ net,j 2 =σ L 2 +σ WT 2 +σ HT 2 +σ PV 2 ,

[0132] wherein, Φ is the cumulative distribution function of P net,j ; μ net,jis the mean of P net,j , and σ net,j 2 is the variance of P net,j ; μ L is the mean of the load output; μ WT is the mean of the wind power output; μ HT is the mean of the hydropower output; μ PV is the mean of the photovoltaic output; μ Hy is the mean of the energy storage output; σ L 2 is the variance of the load output; σ WT 2 is the variance of the wind power output; σ HT 2 is the variance of the hydropower output; σ PV 2 is the variance of the photovoltaic output;

[0133] Similarly for the solution of , similar to the above method, it can be transformed. The two parts in the objective function can be respectively transformed into:

[0134] Obj1: Pr(x > K sub,j S sub,j ) = 1 - Pr(x ≤ K sub,j S sub,j )

[0135] = 1 - ξ(K sub,j S sub,j )

[0136]

[0137] The above solution can be solved using a non - linear solver. For the power balance of the source - grid - load considering the regulation ability, it can be directly solved based on the worst - case scenario obtained from the source - load - grid balance.

[0138] The transformed function is solved by the ε - constraint method. Specifically:

[0139] Assume that the multi - objective optimization includes objective functions F1 and F2. The specific solution steps are as follows:

[0140] Step 1: Optimize and solve each objective function separately. At this time, other objective functions are in an unconstrained state, and we get (taking F1 as the optimization objective, the optimal value obtained and the value of the other objective function at this time F′2) and (taking F2 as the optimization objective, the optimal value obtained and the value of the other objective function at this time F′1);

[0141] Step 2: The value range of F2 non-inferior solution is According to the number of non-inferior solutions n that you want to obtain, select e from the interval with equal spacing. k (k=1,…,n), where the constant value e k is the maximum value obtained from another objective function under unconstrained conditions. Value and minimum ( e k ) values ​​are obtained with equal spacing between them.

[0142] Step 3: Set F2 = e k As a condition, it is put into the optimization model with F1 as the optimization target, and the optimization result is F 1,k , thus obtaining the kth non-inferior solution F1=F 1,k ,F2=e k .

[0143] Step 4: After all n non-inferior solutions are solved, the Pareto frontier data of the multi-objective optimization problem can be constructed.

[0144] From the above, the optimal solution of the distribution network resource configuration model is obtained, and the Pareto frontier data is obtained.

[0145] Step 5: Adjust resource configuration parameters in the distributed distribution network based on the Pareto frontier data.

[0146] Specifically, each non-inferior solution data on the Pareto front is a feasible optimal solution, and the corresponding resource allocation plans are all optimized plans, but the combinations of objective function values ​​are different and can be selected according to actual needs. For example, a preference for a lower configuration budget cost can achieve the goal of minimizing risk and minimizing configuration budget cost.

[0147] In summary, through the model building process and the model solving process described in this embodiment, not only can the risk be predicted, but also the optimal solution for minimizing the risk can be obtained, providing a reference for power grid dispatching.

[0148] Embodiment 2

[0149] Embodiment 2 is a specific test description of Embodiment 1.

[0150] In order to illustrate the effectiveness of the model and the described method in Example 1, a risk indicator assessment is performed on a power supply area of ​​a power station. Specifically, 24 points are taken on a typical day. The maximum capacity after considering the load-to-capacity ratio is 60MVA, the maximum photovoltaic power is 74MW, the maximum wind power is 18MW, the maximum hydropower power is 30MW, and the maximum regional load is 145MW.

[0151] Please refer to Example 1 for the specific calculation and risk assessment process. The risk assessment results are as follows:

[0152] For safety constraint risks, please refer to Figure 2 , Figure 3 As shown, the violations of safety constraints are more obvious at 7 o'clock and 8 o'clock. This is because the wind power is relatively large at this time, the PV power has started to ramp up, but the load power is relatively small, making it easy to violate safety constraints.

[0153] For zero-carbon constraint risks, please refer to Figure 4 , Figure 5 As shown, the violations of zero-carbon constraints are more obvious at 17 o'clock and 18 o'clock. This is because the load power is relatively large during the evening peak, but the PV output is already relatively small, making it easy to violate zero-carbon constraints.

[0154] Please refer to Figure 6 , Figure 7 The Pareto front of the power balance risk index shown, and the risk index evaluation situation corresponding to the Pareto point (maximum risk probability, all-day cost) is at the circle.

[0155] In summary, it can be seen that by using the method described in Embodiment 1 and combining the implementation of adjustment measures, the power balance risk in the power grid has been significantly reduced, the safety of the power grid system has been improved, and the low-carbon operation of the system has been ensured.

[0156] Embodiment 3

[0157] Embodiment 3 discloses a device corresponding to the power balance risk index evaluation method corresponding to the above embodiment. For the virtual device structure of the above embodiment, please refer to Figure 8 As shown, it includes:

[0158] A construction module 210, configured to construct a time-series characteristic power model according to the source-load historical data, and construct multiple types of adjustable resource models according to the adjustable resources in the power system;

[0159] An index construction module 220, configured to evaluate the probability that the balance equation in the source-network-load-storage power model does not hold, and use it as a risk index; construct a risk index evaluation model with the minimum risk and minimum adjustment cost as the goal; construct a multi-objective distributionally robust model as the worst-case risk index, and complete the construction of the power balance risk index evaluation model;

[0160] An analysis module 230, configured to solve the model to obtain a risk assessment result.

[0161] Embodiment 4

[0162] Figure 9 The structural schematic diagram of an electronic device provided in Embodiment 4 of the present invention is as Figure 9 shown. The electronic device includes a processor 310, a memory 320, an input device 330, and an output device 340; the number of processors 310 in the computer device can be one or more.Figure 9 Taking a processor 310 in the middle as an example; the processor 310, memory 320, input device 330, and output device 340 in the electronic device can be connected by a bus or other means. Figure 9 Taking the connection by bus as an example in the middle.

[0163] The memory 320, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the power balance risk index evaluation method in the embodiments of the present invention. The processor 310 executes various functional applications and data processing of the electronic device by running the software programs, instructions, and modules stored in the memory 320, that is, implements the power balance risk index evaluation method in the above-mentioned Embodiment 1 to Embodiment 2.

[0164] The memory 320 may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the terminal, etc. In addition, the memory 320 may include high-speed random access memory and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage devices. In some instances, the memory 320 may further include a memory remotely set relative to the processor 310, and these remote memories can be connected to the electronic device through a network. Examples of the above network include but are not limited to the Internet, enterprise intranet, local area network, mobile communication network, and their combinations.

[0165] The input device 330 can be used to receive input user identity information, grid data, etc. The output device 340 may include a display device such as a display screen.

[0166] Embodiment 5

[0167] Embodiment 5 of the present invention further provides a storage medium containing computer-executable instructions, and this storage medium can be used by a computer to execute the power balance risk index evaluation method, and this method includes:

[0168] Construct a time-series characteristic power model according to the source-load historical data, and construct multi-type adjustable resource models according to the adjustable resources in the power system;

[0169] Evaluate the probability that the balance equation in the source-network-load-storage power model does not hold, and use it as a risk index;

[0170] Construct a risk index evaluation model with the goal of minimizing risk and minimizing adjustment cost;

[0171] Construct a multi-objective distributionally robust model as the worst-case risk index to complete the construction of the power balance risk index evaluation model;

[0172] Solve the model through a truncated normal distribution to obtain a risk assessment result.

[0173] Certainly, for a storage medium containing computer-executable instructions provided by an embodiment of the present invention, the computer-executable instructions are not limited to the method operations described above, and can also execute relevant operations in the power balance risk index assessment method provided by any embodiment of the present invention.

[0174] From the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software and necessary general hardware. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk, or optical disc of a computer, and includes several instructions for causing an electronic device (which can be a mobile phone, personal computer, server, or network device, etc.) to execute the methods described in various embodiments of the present invention.

[0175] It should be noted that in the embodiments of the above-mentioned device for the power balance risk index assessment method, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0176] For those skilled in the art, according to the technical solutions and concepts described above, various corresponding changes and deformations can be made, and all such changes and deformations should fall within the protection scope of the claims of the present invention.

Claims

1. A method for evaluating a power balance risk index, characterized in that, Including the following steps: Construct a time-series characteristic power model based on source-load historical data, and construct multi-type adjustable resource models according to the adjustable resources in the power system; Evaluate the probability that the balance equation in the source-network-load-storage power model does not hold, and use it as a risk index; Construct a risk index evaluation model with the goals of minimum risk and minimum adjustment cost; Construct a multi-objective distributionally robust model as the worst-case risk index to complete the construction of the power balance risk index evaluation model. Among them, constructing a multi-objective distributionally robust model as the worst-case risk index, the goals include probability model uncertainty and cost minimization, and the probability model uncertainty goal satisfies: The minimum cost objective satisfies: min[C ESS +C HHBES_deg +C DR +C CUR , Among them, is the wind power output; is the photovoltaic power output; is the hydropower output; is the load output; represents the net load after regulation; Pr{·} is the probability; is the demand response output; is the wind power curtailment; V ht is the hydropower reservoir capacity; is the photovoltaic power curtailment; is the output of the m-th monthly hydrogen storage scheduling cycle at time t; is the output of the electrical energy storage; K sub,j is the substation capacity ratio; S sub,j is the capacity of the j-th substation; is the substation power factor, C ESS is the cost of electrical energy storage, C HHBES_deg is the cost of hydrogen energy storage, C DR is the cost of demand response, C CUR is the penalty cost for wind and solar curtailment; Solve the model through truncated normal distribution to obtain the risk assessment result; Solving the model includes: Transform the function in the model through truncated normal distribution, and the absolute value of the net load of the function satisfies: μ net,j = μ L - μ WT - μ HT - μ PV - μ Hy ,σ net,j 2 = σ L 2 + σ WT 2 + σ HT 2 + σ PV 2 , Among them, P net,j is the net load, and Φ is the cumulative distribution function of P net,j ; μ net,j is the mean of P net,j , and σ net,j 2 is the variance of P net,j ; μ L is the mean of the load output; μ WT is the mean of the wind power output; μ HT is the mean of the hydropower output; μ PV is the mean of the photovoltaic output; μ Hy is the mean of the energy storage output; σ L 2 is the variance of the load output; σ WT 2 is the variance of the wind power output; σ HT 2 is the variance of the hydropower output; σ PV 2 is the variance of the photovoltaic output, and x is a random variable that follows the distribution ξ(x); Solve the transformed function by the ε-constraint method to obtain the risk assessment result.

2. The power balance risk index evaluation method according to claim 1, wherein Construct a time-series characteristic power model based on source-load historical data. The construction of the power model includes: Based on the normal distribution probability model and the time-series characteristic curves of distributed energy output in the prefecture-level power grid, for the photovoltaic and wind power systems i, for each moment t, establish a time-series characteristic fuzzy set of distributed energy output based on data-driven moment information: Among them, D DG1 and D DG2 are respectively the fuzzy set of the random variable ξ DG,i,t of the distributed power source and the mean sequence μ DG,i,t and the variance sequence σ DG,i,t of the output curve of the distributed power source, is the distribution set of the random variable ξ DG,i,t ; P(·) is the probability function, indicating that the distribution of the random variable ξ DG,i,t belongs to with a probability of 1; E(·) is the expectation function, and μ DG，i，t are respectively the upper and lower limits of the mean, and are respectively the upper and lower limits of the variance, μ DG,pre,i,t and σ DG,pre,i,t are respectively the predicted values of the mean and variance, α and β are respectively the allowable deviation coefficients used to control the conservatism of the fuzzy set, and N DG is the number of distributed power source configurations; Construct a hydropower power model, including: Establish a hydroelectric conversion function where η h is the power generation efficiency, H ht is the water head, is the power generation flow rate, and g is the hydroelectric conversion coefficient; Model the actual value of the power generation flow rate as a random variable that follows a normal distribution: where μ ht is the mean value of the power generation flow rate and σ ht is the coefficient of standard deviation of the power generation flow rate ; Construct a fuzzy set of hydropower output; Among them, and are the fuzzy set of the random variable and the mean value of the hydropower output and the variance of the uncertainty set; is the distribution set of the random variable , P(·) is the probability function, indicating that the distribution of the random variable belongs to with a probability of 1, and E(·) is the expectation function; and are the upper and lower limits of the mean value respectively, and are the upper and lower limits of the variance respectively; and are the predicted values of the mean value and the variance respectively, α and β are the allowable deviation coefficients respectively, and N HT is the number of hydropower configurations; Construct a load power model, including: Establish a time-series characteristic fuzzy set of load output based on data-driven moment information; Among them, D L1 and D L2 are the fuzzy set of the random variable and the mean value μ L,i,t and the variance of the load output, respectively, which are uncertain sets; is the distribution set of the random variable , P(·) is the probability function, indicating that the distribution of the random variable belongs to with a probability of 1, and E(·) is the expectation function; and are the upper and lower limits of the mean value, respectively, and are the upper and lower limits of the variance, respectively; and are the predicted values of the mean value and the variance, respectively, α and β are the allowable deviation coefficients, and N L is the number of loads.

3. The power balance risk index evaluation method according to claim 1, characterized in that Construct multi-type adjustable resource models according to the adjustable resources in the power system, including: Combined with the energy constraint, degradation cost and scheduling period of hydrogen energy storage, construct a long-time-scale hydrogen energy storage model, and the output power constraint satisfies: Among them, FC represents a fuel cell, and ED represents an electrolysis device. respectively represent the upper and lower bounds of the output power of the fuel cell and the electrolysis device. and are binary variables indicating whether the fuel cell and the electrolysis device are working at time t in the m-th monthly scheduling period of the hydrogen storage, respectively. is the output power of the hydrogen storage at time t in the m-th monthly scheduling period, and η ED and η FC are the working efficiencies of the electrolysis device and the fuel cell, respectively; the energy constraint satisfies the hydrogen storage capacity recurrence limit and the hydrogen storage capacity limit at each moment. Among them, the energy constraint needs to satisfy the hydrogen storage capacity recurrence limit and the hydrogen storage capacity limit at each moment as follows: Among them, is the capacity at time t of the m-th monthly scheduling period for hydrogen storage, and are the upper and lower limits of the capacity; The operating cost of hydrogen energy storage includes the sum of the degradation costs of fuel cells and electrolysis devices; The optimization of the monthly change in hydrogen energy storage energy satisfies: Among them, represents the energy at the end of the m-th monthly scheduling period of hydrogen energy storage, E net represents the change in the energy of hydrogen energy storage per month, represents the load output of substation j, represents the wind power output of substation j, represents the hydropower output of substation j, represents the photovoltaic output of substation j, represents the output of hydrogen energy storage of substation j, N L represents the number of loads, N WT represents the number of wind power, N HT represents the number of hydropower, N PV represents the number of photovoltaic, N Hy represents the number of hydrogen energy storage; Construct a short-time primary electricity energy storage model, satisfying: SOC min ≤ SOC t ≤ SOC max Among them, SOC t is the state of charge of the electrical energy storage, and SOC max and SOC min are the upper and lower limit constraints of the state of charge. E t is the energy of the electrical energy storage at time t, and is the charging and discharging power of the electrical energy storage at time t. is the upper limit of the charging and discharging power of the electrical energy storage. E r is the energy storage capacity, and η is the charging and discharging efficiency of the energy storage; Construct adjustment means models for adjustable hydropower, wind and light abandonment, and load demand response. For any hydropower station i, the reservoir capacity and flow satisfy the constraints: Among them, are respectively the upper and lower limits of the reservoir capacity and flow rate of the hydropower station; The reduction amount satisfies: Among them, are the curtailment amounts of wind and light respectively, are the maximum curtailment degrees of wind and light respectively; are the day-ahead predicted outputs of wind power and photovoltaic power respectively; and a function of the total cost of wind and light curtailment and incentives is constructed.

4. The power balance risk index evaluation method according to claim 1, characterized in that Evaluate the probability that the balance equation in the source-network-load-storage power model does not hold, and use it as a risk index, satisfying: Among them, represents the adjusted net load; is the load output in substation j; is the demand response output in substation j; is the wind power output in substation j; is the wind power curtailment in substation j; is the hydropower output after adjustment in substation j; is the photovoltaic output in substation j; is the photovoltaic curtailment in substation j; is the output of hydrogen storage in substation j; is the output of electrical energy storage in substation j; K sub,j is the capacity-load ratio of the substation; is the capacity of the j-th substation; is the power factor of the substation, N L represents the number of loads, N WT represents the number of wind power, N HT represents the number of hydropower, N PV represents the number of photovoltaics, N Hy represents the number of hydrogen storage, N ESS represents the number of electrical energy storage.

5. A power balance risk index evaluation device, characterized in that, It includes: A construction module for constructing a time-series characteristic power model based on source-load historical data and constructing multi-type adjustable resource models according to the adjustable resources in the power system; An index construction module for evaluating the probability that the balance equation in the source-network-load-storage power model does not hold and using it as a risk index; Constructing a risk index evaluation model with the goals of minimum risk and minimum adjustment cost; Constructing a multi-objective distributionally robust model as the worst-case risk index to complete the construction of the power balance risk index evaluation model. Among them, constructing a multi-objective distributionally robust model as the worst-case risk index, the goals include probability model uncertainty and cost minimization, and the probability model uncertainty goal satisfies: The minimized cost objective satisfies: min[C ESS +C HHBES_deg +C DR +C CUR , Among them, is the wind power output; is the photovoltaic power output; is the hydropower output; is the load output; represents the net load after regulation; Pr{·} is the probability; is the demand response output; is the wind power curtailment; V ht is the hydropower reservoir capacity; is the photovoltaic power curtailment; is the output of the m-th month's scheduling cycle t of the hydrogen storage; is the output of the electrical energy storage; K sub,j is the substation capacity ratio; S sub,j is the capacity of the j-th substation; is the substation power factor, C ESS is the electrical energy storage cost, C HHBES_deg is the hydrogen energy storage cost, C DR is the demand response cost, C CUR is the penalty cost for wind and light curtailment; An analysis module for solving the model to obtain the risk assessment result; Solving the model includes: Transform the function in the model through truncated normal distribution, and the absolute value of the net load of the function satisfies: μ net,j = μ L - μ WT - μ HT - μ PV - μ Hy ,σ net,j 2 = σ L 2 + σ WT 2 + σ HT 2 + σ PV 2 , Where, P net,j is the net load, and Φ is the cumulative distribution function of P net,j ; μ net,j is the mean value of P net,j , and σ net,j 2 is the variance of P net,j ; μ L is the mean value of the load output; μ WT is the mean value of the wind power output; μ HT is the mean value of the hydropower output; μ PV is the mean value of the photovoltaic output; μ Hy is the mean value of the energy storage output; σ L 2 is the variance of the load output; σ WT 2 is the variance of the wind power output; σ HT 2 is the variance of the hydropower output; σ PV 2 is the variance of the photovoltaic output, and x is a random variable that follows the distribution ξ(x); The converted function is solved by the ε-constraint method to obtain the risk assessment result.

6. An electronic device, comprising a processor, a storage medium, and a computer program stored in the storage medium, characterized in that, When the computer program is executed by a processor, it implements the power balance risk index assessment method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the power balance risk index assessment method according to any one of claims 1 to 4.

Images