Wind power-containing electric power system source load cooperative scheduling method considering smelting type high-energy load participation

Abstract

The invention discloses a wind power-containing electric power system source load collaborative scheduling method considering the participation of a smelting type high-energy load, and the method comprises the steps: analyzing the scheduling demands of a wind power-containing electric power system, and building a model considering the load of an electric arc furnace and the wind power output uncertainty based on the schedulable characteristics of the load of the electric arc furnace; a data driving method is adopted to generate a large-scale scene, the uncertainty superposition influence of the electric arc furnace load and the wind power output is analyzed, and the reserve capacity needed by the electric power system for coping with uncertainty fluctuation is determined; a smelting type high-energy load self-adaptive electricity price mechanism considering wind power output change and source load fluctuation is provided, and iron and steel enterprises are stimulated to participate in demand response; and establishing an optimal scheduling model giving consideration to interests of multiple parties, and solving an optimal scheduling scheme by adopting an artificial bee colony algorithm. According to the method, economic benefits and social responsibilities of smelting type high-energy load enterprises are considered, support is provided for the enterprises to participate in consumption of blocked wind power, and new energy consumption and economic and stable operation of a power system are facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid dispatching. Specifically, it relates to a source-load coordinated dispatching method for a wind power-integrated power system considering the participation of high-energy-consuming loads in smelting. Background Art

[0002] As the construction of China's new power system enters a critical period, the development of new energy has reached a record high, which is of great significance for achieving the "dual carbon" goal. However, its uncertainty also poses challenges to the stable operation of the power grid. At the same time, due to the reduction in the proportion of traditional units, the system's peak shaving capacity is affected, resulting in frequent occurrences of wind power curtailment. Therefore, exploring the adjustable potential on the load side and guiding the load-side adjustable resources to respond to power grid dispatching through incentive mechanisms such as electricity prices has become an important way to improve the wind power consumption capacity of the system and ensure stable operation. Currently, a large number of studies have been carried out on the coordinated dispatching of the electric arc furnace, a scheduling entity with strong adjustable potential in high-energy-consuming load enterprises under the guidance of the electricity price mechanism, and new energy. However, most of the studies do not conform to the actual situation of the electric arc furnace's gear shifting regulation in power modeling, or ignore the uncertainty it shares with new energy. The electricity prices formulated are difficult to actually guide the electric arc furnace to change its production plan, resulting in a large deviation between the formulated dispatching plan and the actual situation of the power system, seriously affecting the consumption of new energy and the economic, safe and stable operation of the power system.

[0003] Currently, there is no unified standard for the source-load coordinated dispatching method of a wind power-integrated power system considering the participation of high-energy-consuming loads in smelting. Existing technologies often insufficiently consider the actual dispatching requirements of the power grid and the uncertainty of the source and load. To solve the above problems, it is necessary to propose an effective source-load coordinated dispatching method for a wind power-integrated power system considering the participation of high-energy-consuming loads in smelting, so as to improve the consumption of new energy in the wind power-integrated power system and ensure the safe, stable and economic operation of the system. Summary of the Invention

[0004] The problem to be solved by the present invention is that, aiming at the problem that the power modeling of the electric arc furnace in existing research does not conform to the actual situation of its gear shifting regulation, or ignores the uncertainty it shares with new energy, and the formulated dispatching plan often has a large deviation from the actual situation of the power system, seriously affecting the consumption of new energy and the economic, safe and stable operation of the power system, a source-load coordinated dispatching method for a wind power-integrated power system considering the participation of high-energy-consuming loads in smelting is proposed, which is beneficial to providing strong theoretical support for the participation of high-energy-consuming loads in smelting in power grid dispatching and wind power consumption, ensuring the accuracy of power grid dispatching, and promoting the consumption of new energy and the safe operation of the power system.

[0005] To solve the above problems, the present invention provides a source-load coordinated dispatching method for a wind power-integrated power system considering the participation of high-energy-consuming loads in smelting, including:

[0006] Step S1: Analyze the dispatching requirements of the power system with wind power, and establish a model considering the uncertainties of arc furnace load and wind power output based on the dispatchable characteristics of the arc furnace load;

[0007] Step S2: Use a data-driven method to generate a large number of scenarios, analyze the superposition effect of the uncertainties of arc furnace load and wind power output, and determine the reserve capacity required for the power system to cope with the uncertainty fluctuations;

[0008] Step S3: Propose an adaptive electricity price mechanism for smelting-type high-energy-consuming loads that takes into account the changes in wind power output and source-load fluctuations, and encourage steel enterprises to participate in demand response;

[0009] Step S4: Establish an optimal dispatching model that takes into account the interests of multiple parties, and use the Artificial Bee Colony (ABC) algorithm to solve the optimal dispatching plan.

[0010] Furthermore, in Step S1, analyze the dispatching requirements of the power system with wind power, and establish a model considering the uncertainties of arc furnace load and wind power output based on the dispatchable characteristics of the arc furnace load; As large power users, high-energy-consuming load enterprises have great advantages in participating in wind power consumption with the arc furnace as the main load: First, the arc furnace can be adjusted quickly and can respond to wind power fluctuations in a short time; Second, the single-unit power of the arc furnace is large, and the load can be transferred to the peak period of wind power output to consume wind power;

[0011] Define the equivalent load:

[0012] P t equ = P t LD - P t W (1)

[0013] Where: P t equ , P t LD and P t W are the equivalent load, basic load and wind power output at time t, respectively;

[0014] When the arc furnace load does not participate in dispatching, if the equivalent load at time t is less than the minimum total output of thermal power units , the power system can only ensure the safe operation of the system by abandoning wind, and the amount of abandoned wind is:

[0015]

[0016] Where: T and Δt are the number and duration of scheduling periods respectively.

[0017] After high energy load participates in regulation, the load in other periods can be transferred to this period, and the abandoned air volume P t W,abd for:

[0018]

[0019] Where: N F is the number of electric arc furnaces; is the planned operating power of arc furnace i during period t.

[0020] From the above equations (2) and (3), it can be seen that the additional wind power consumption of the power system after high-energy load participates in the dispatch is P W，abd for:

[0021]

[0022] Therefore, the participation of high-energy load in regulation can effectively improve the power system's ability to absorb wind power;

[0023] When regulating the power of an arc furnace, the power change is not continuous, but rather a discrete change based on the position of the transformer tap;

[0024] The planned power of the electric arc furnace is:

[0025]

[0026] Where: is the start and stop status of arc furnace i in time period t; α i is the percentage of initial power of arc furnace i to rated power; is the transformer tap position of arc furnace i in time period t; is the total number of tap positions of transformer i of electric arc furnace; is the power of the transformer tap of arc furnace i when it is in the initial position;

[0027] Due to the instability of the arc inside the arc furnace and the collapse of materials during operation, the active power of the arc furnace is uncertain. The maximum active power fluctuation of some arc furnaces can reach more than 20% of the planned operating power, so its impact on the power system dispatch cannot be ignored.

[0028] The present invention describes the fluctuation of the electric arc furnace through normal distribution, providing a reference for selecting the system spare capacity; assuming that the fluctuation of the electric arc furnace To obey the expectation of 0 and the variance of The normal distribution is:

[0029]

[0030] To generate a distribution function that better conforms to the actual situation of the electric arc furnace, a standard normal distribution function with a 95% confidence interval and an interval width of 20% of the installed capacity of the equipment is intercepted as the probability function of the active power fluctuation of the electric arc furnace. The actual operating power of the electric arc furnace after adding the fluctuation is:

[0031]

[0032] Adjusting the load of the electric arc furnace needs to meet the following constraints:

[0033] The steel enterprise must complete the production task during the scheduling period, and its total energy consumption needs to meet the minimum production energy consumption constraint:

[0034]

[0035] In the formula: is the minimum production energy consumption of the steel enterprise;

[0036] The smelting time of one heat of the electric arc furnace cannot be too long or too short:

[0037]

[0038] In the formula: and are the longest and shortest smelting times of one heat respectively.

[0039] The tap changer of the electric arc furnace transformer cannot be adjusted across grades:

[0040]

[0041] In the formula: τ is the multiple of the time ε required to adjust the tap changer once. Since the operation time of one heat of the electric arc furnace is relatively long, while the adjustment process of the tap changer position of the transformer is relatively rapid, it is difficult to observe the step-by-step adjustment process of the tap changer from the measured data. The duration of a time period in the scheduling period satisfies:

[0042] ε = Vt (12)

[0043] Therefore, this constraint does not need to be considered subsequently.

[0044] Furthermore, in step S2, a data-driven method is used to generate large-scale scenarios, analyze the superposition effect of the uncertainties of the electric arc furnace load and wind power output, and determine the reserve capacity required for the power system to cope with the uncertain fluctuations; since the wind power output is affected by weather, its predicted data often deviates from the actual value, that is, there is uncertainty. If the predicted data is directly used to formulate the scheduling plan, it does not conform to the actual situation. The present invention generates multiple sets of scenario sets based on the predicted wind power output as the basic data set, analyzes the superposition of the uncertainties of the source and load, and selects typical scenarios to formulate an optimal scheduling plan;

[0045] The Auto-regressive and Moving Average (ARMA) model is a high-precision time series model suitable for simulating the uncertainty of wind power. Its expression is as follows:

[0046]

[0047] ΔP t W = P t W - P t W,pre (14)

[0048] In the formula: P t W is the wind power output at time t; is the autoregressive parameter; is the moving average parameter; ε t is the random disturbance at time t of the time series; a is the autoregressive order; b is the moving average order; ΔP t W is the wind power output fluctuation at time t;

[0049] To eliminate similar scenarios in the wind power scenario set to generate typical scenarios, the present invention comprehensively considers data validity and computational complexity, and uses a fast forward elimination method based on probability distance to reduce scenarios for the large-scale scenario set generated by the ARMA model. The process is as Figure 2 shown, including:

[0050] (1) Calculate the geometric distance between each pair of scenarios c and c' in the scenario set C;

[0051] (2) Select the scenario d with the smallest sum of probability distances from the remaining scenarios;

[0052] (3) Replace scenario d with the scenario e in C that is closest to scenario d in geometric distance;

[0053] (4) Add the probability of d to the probability of scenario e to eliminate d and form a new scenario set C';

[0054] (5) Check whether the number of remaining scenarios meets the requirements. If it meets the requirements, output the reduced typical scenario set; if not, return to step (2) to continue the process until the requirements are met.

[0055] Since both the arc furnace load and the wind power output have uncertainties, the interaction effects of the power fluctuations of the two need to be considered in the optimization model. Let the total number of sequences to be simulated be N, and the source-load power fluctuation sequences are obtained by Monte Carlo algorithm and ARMA model respectively:

[0056]

[0057] In the formula: and are the arc furnace and wind power fluctuation sequences under scenario n respectively; and are the arc furnace and wind power fluctuation values at time period t under scenario n respectively;

[0058] At time period t of scenario n, if and are both positive or both negative, the two have a positive superposition effect, and the arc furnace fluctuation will absorb the wind power output fluctuation; if and are one positive and one negative, the two have a reverse superposition effect, pushing up the equivalent load and having an adverse impact on the stable operation of the power grid;

[0059] Generally speaking, the reserve capacity is obtained based on the unit parameters and it is difficult to accurately measure the capacity actually required by the system to accommodate fluctuations. When the selected reserve capacity is too large, the unit output is wasted, and when it is too small, the wind abandonment will be pushed up; the maximum and minimum values of the source-load superposition effect are used to describe the system reserve capacity:

[0060]

[0061] In the formula: P t u,LD and P t d,LD are the upper and lower spinning reserves required to cope with load forecasting at time period t respectively; P t u,FW and are the upper and lower spinning reserves required to cope with the superposition effect of the arc furnace load and wind power output fluctuation at time period t respectively; N G is the number of thermal power units in the system; is the start-stop state of thermal power unit i at time period t; and are the maximum and minimum outputs of thermal power unit i respectively; is the actual output of thermal power unit i at time period t.

[0062] Furthermore, in step S3, based on the generated scenario set, an adaptive electricity price mechanism for smelting-type high-energy-consuming loads considering wind power output changes and source-load fluctuations is proposed to encourage steel enterprises to participate in demand response;

[0063] At present, the penetration rate of new energy sources such as wind power in the power system is increasing continuously. Its uncertainty poses a serious threat to the safe and stable operation of the power system, and also makes the existing fixed time-of-use electricity price division method often not in line with the actual situation of the power system, unable to ensure effective incentives for users when there are dispatching requirements such as peak shaving or wind power consumption in the power system to guide them to respond to dispatching; The present invention proposes an adaptive electricity price mechanism for smelting-type high-energy-consuming loads oriented to wind power consumption. On the basis of the original time-of-use electricity price, the mechanism introduces the influence of the equivalent load containing wind power and the uncertainty fluctuation of source and load in each time period, and the expression is as follows:

[0064]

[0065] In the formula: is the adaptive electricity price of the smelting-type high-energy-consuming load in the t-th time period; and are respectively the conversion coefficient of the standard time-of-use electricity price for industrial users and the standard time-of-use electricity price for industrial users in the t-th time period; and are respectively the equivalent load influence coefficient and the equivalent load compensation electricity price in the t-th time period; and are respectively the fluctuation compensation coefficient and the fluctuation compensation electricity price in the t-th time period; and are respectively the average values of the generated arc furnace load fluctuation scenario set and the wind power output fluctuation scenario set in the t-th time period; T max is for all scenario sets the time period with the largest; and are the average values of the large-scale wind power output fluctuation and the arc furnace load fluctuation scenario set at T max below.

[0066] Furthermore, in step S4, an optimal dispatching model that takes into account the interests of multiple parties is established, and the artificial bee colony algorithm is used to solve the optimal dispatching scheme; To ensure the economic and stable operation of the power system containing wind power, the following objective function and constraint conditions are established:

[0067] Objective function 1: The lowest system operation cost;

[0068] The system operation cost mainly comes from thermal power units and wind farms, and the expression is as follows:

[0069]

[0070] In the formula: a i 、b i and c i are the unit operation cost coefficients; N W is the number of wind farms; μ co,W is the wind farm operation cost coefficient; The output of wind power of wind farm j in period t;

[0071] Objective function 2: The lowest curtailment penalty;

[0072] The curtailment penalty reflects the system's curtailment volume, and the expression is:

[0073]

[0074] In the formula: μ abd,W is the curtailment penalty coefficient of the wind farm;

[0075] Objective function 3: The lowest electricity purchase cost of iron and steel enterprises;

[0076] The electricity purchase cost reflects the enthusiasm of high energy-consuming loads such as smelting for response dispatching, and the expression is:

[0077]

[0078] System power balance constraint:

[0079]

[0080] Upper and lower limits constraint of thermal power unit output power:

[0081]

[0082] In the formula: and are the upper and lower limits of the output power of thermal power unit i in period t;

[0083] Wind power output constraint:

[0084]

[0085] In the formula: is the predicted value of the output power of wind farm i in period t;

[0086] Thermal power unit ramp rate constraint:

[0087]

[0088] In the formula: and are the upper and lower limits of the ramp rate power of thermal power unit i in period t;

[0089] Minimum on / off time constraint of thermal power unit:

[0090]

[0091] In the formula: T i,on,min and T i,off,min are the minimum on / off time of thermal power unit i;

[0092] Since the dimensions of the objective functions are the same, normalization is not required. The multi-objective optimization problem can be directly converted into a single-objective optimization problem by adding the three objective functions. The total objective function F after conversion is to minimize the total system cost, and the expression is:

[0093] min F = F1 + F2 + F3 (33)

[0094] Since the model of the present invention contains a non-linear part of the multiplication of 0-1 variables and integer variables, the artificial bee colony algorithm (ABC) is used to solve the optimization model. This algorithm is a meta-heuristic algorithm based on bee colonies, which can solve non-linear programming problems, avoid converting non-linear problems into linear problems, and has good solving performance and convergence ability, suitable for solving the non-linear optimization model proposed in the present invention. The flow chart of using the artificial bee colony algorithm to solve the source-load coordinated scheduling model of a wind power integrated power system considering high-energy-consuming loads in smelting is as Figure 3 shown, including:

[0095] (1) Initialize the parameters in the optimization model to generate an initial population containing the solutions of the source-load coordinated scheduling model of a wind power integrated power system considering high-energy-consuming loads in smelting.

[0096] (2) Calculate the fitness values in the initial population of the source-load coordinated scheduling model of a wind power integrated power system considering high-energy-consuming loads in smelting, providing a basis for subsequent sorting and selection.

[0097] (3) The leading bee randomly generates a new solution of the source-load coordinated scheduling model of a wind power integrated power system considering high-energy-consuming loads in smelting according to the original position and makes a greedy selection: calculate the fitness value of the new solution and evaluate it. If the fitness value of the new solution is better than the old solution, the leading bee updates the old solution to the new solution; otherwise, the old solution is retained.

[0098] (4) The follower bee performs roulette selection according to the fitness of the solution, calculates the selection probability of each solution of the source-load coordinated scheduling model of a wind power integrated power system considering high-energy-consuming loads in smelting, and determines whether to update the optimal solution variable to the new solution according to the probability.

[0099] (5) The scout bee determines whether there is an update within the exploration limit. If there is no update, it is determined that the algorithm has fallen into a local optimal solution, then the original solution is discarded and the leading bee corresponding to this solution is changed into a scout bee, and a new solution is generated to replace the original solution, that is, the original leading bee.

[0100] (6) Determine whether the end condition is reached. If so, output the optimal solution of the source-load coordinated scheduling model of a wind power integrated power system considering high-energy-consuming loads in smelting; otherwise, return to (3) to continue the loop. Description of the Drawings

[0101] Figure 1 Flow chart of source-load coordinated dispatching method for a wind power-integrated power system considering high energy-consuming loads in smelting

[0102] Figure 2 Flow chart of scenario reduction using the fast forward elimination method based on probabilistic distance

[0103] Figure 3 Flow chart of solving the source-load coordinated dispatching model for a wind power-integrated power system considering high energy-consuming loads in smelting by the ABC algorithm Specific implementation manners

[0104] To make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to embodiments and drawings. The illustrative embodiments and descriptions thereof of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0105] For the current situation, a method for evaluating the control performance of flexible controllable resources for maximizing the consumption of new energy has a process as Figure 1 shown, including the following steps:

[0106] Step S1: Analyze the dispatching requirements of the wind power-integrated power system, and based on the dispatchable characteristics of the arc furnace load, establish a model considering the uncertainties of the arc furnace load and wind power output.

[0107] Step S2: Use a data-driven method to generate a large number of scenarios, analyze the superimposed influence of the uncertainties of the arc furnace load and wind power output, and determine the reserve capacity required for the power system to cope with uncertainty fluctuations.

[0108] Step S3: Based on the generated scenario set, propose an adaptive electricity price mechanism for high energy-consuming loads in smelting considering the changes in wind power output and source-load fluctuations, and encourage steel enterprises to participate in demand response.

[0109] Step S4: Establish an optimal dispatching model considering the interests of multiple parties, and use the Artificial Bee Colony (ABC) algorithm to solve the optimal dispatching scheme.

[0110] In step S1, analyze the dispatching requirements of the wind power-integrated power system, and based on the dispatchable characteristics of the arc furnace load, establish a model considering the uncertainties of the arc furnace load and wind power output. As large power users, high energy-consuming load enterprises, among which the arc furnace as the main load has great advantages in participating in the consumption of wind power: First, the arc furnace can be adjusted quickly and can respond to wind power fluctuations in a short time; second, the single arc furnace has a large power, and the load can be transferred to the peak period of wind power output to consume wind power. Define the equivalent load:

[0111] P tequ = P t LD -P t W (1)

[0112] Where: P t equ , P t LD and P t W are the equivalent load, base load and wind power output in period t, respectively.

[0113] When the arc furnace load does not participate in the scheduling, if the equivalent load in period t is less than the minimum total output of thermal power units at this time, the power system can only ensure the safe operation of the system by curtailing wind power, and the curtailed wind power is:

[0114]

[0115] Where: T and Δt are the number and duration of the scheduling periods, respectively.

[0116] After the high-energy-consuming load participates in the regulation, the load in other periods can be transferred to this period. At this time, the curtailed wind power P t W,abd is:

[0117]

[0118] Where: N F is the number of arc furnaces; is the planned operating power of arc furnace i in period t.

[0119] From the above two equations, it can be seen that the newly added wind power consumption capacity of the power system after the high-energy-consuming load participates in the scheduling is:

[0120]

[0121] Therefore, the participation of high-energy-consuming loads in the regulation can effectively improve the wind power consumption capacity of the power system.

[0122] When adjusting the power of the arc furnace, the change in power is not continuous, but discrete based on the tap position of the transformer. The planned power of the arc furnace is:

[0123]

[0124] Where: is the start-stop state of arc furnace i in period t; α i is the percentage of the initial power of arc furnace i to the rated power; is the gear position of the transformer tap of electric arc furnace i at time period t; is the total number of gear positions of the transformer tap of electric arc furnace i; is the power when the transformer tap of electric arc furnace i is at the initial gear position.

[0125] Due to the instability of the arc inside the electric arc furnace and the existence of material collapse during operation, there is uncertainty in its active power. The maximum active power fluctuation during the operation of some electric arc furnaces can reach more than 20% of the planned operating power. Therefore, its impact on the power system dispatching cannot be ignored. The present invention describes the fluctuation situation of the electric arc furnace through normal distribution, providing a reference for the selection of system reserve capacity. Let the fluctuation amount of the electric arc furnace obeys a normal distribution with an expectation of 0 and a variance of , that is:

[0126]

[0127] In order to generate a distribution function that better conforms to the actual situation of the electric arc furnace, the standard normal distribution function with a confidence interval of 95% and an interval width of 20% of the equipment installed capacity is intercepted as the probability function of the active power fluctuation amount of the electric arc furnace. The actual operating power of the electric arc furnace after adding the fluctuation amount is:

[0128]

[0129] Adjusting the load of the electric arc furnace needs to meet the following constraints:

[0130] The steel enterprise must complete the production task during the dispatching period, and its total energy consumption needs to meet the minimum production energy consumption constraint:

[0131]

[0132] In the formula: is the minimum production energy consumption of the steel enterprise.

[0133] The smelting time of one heat of the electric arc furnace cannot be too long or too short:

[0134]

[0135] In the formula: and are the longest and shortest smelting times of one heat respectively.

[0136] The transformer tap of the electric arc furnace cannot be adjusted across gears:

[0137]

[0138] Where: τ is the multiple of the time ε required to adjust the primary tap. Since the operation time of one heat in arc furnace smelting is relatively long, while the adjustment process of the transformer tap position is relatively rapid, it is difficult to observe the stepwise adjustment process of the tap from the measured data. The duration of one period of the scheduling cycle satisfies:

[0139] ε = Vt (12)

[0140] Therefore, this constraint does not need to be considered subsequently.

[0141] In step S2, a data-driven method is used to generate large-scale scenarios, analyze the superposition effect of the uncertainties of the arc furnace load and wind power output, and determine the reserve capacity required for the power system to cope with the uncertain fluctuations. Since the wind power output is affected by weather, there are often deviations from the actual values in its prediction data, that is, uncertainties. If the prediction data is directly used to formulate the scheduling plan, it does not conform to the actual situation. The present invention generates multiple sets of scenario sets based on the wind power output prediction as the basic data set, analyzes the superposition of uncertainties of the source and load, and selects typical scenarios to formulate an optimal scheduling plan.

[0142] The autoregressive moving average model (ARMA) is a high-precision time series model, which is suitable for simulating the uncertainty of wind power. The expression is:

[0143]

[0144] ΔP t W = P t W - P t W,pre (14)

[0145] Where: P t W is the wind power output at time t; is the autoregressive parameter; is the moving average parameter; ε t is the random disturbance at time t of the time series; a is the autoregressive order; b is the moving average order; ΔP t W is the wind power output fluctuation at time t.

[0146] To eliminate the similar scenarios in the wind power scenario set to generate typical scenarios, the present invention comprehensively considers the data validity and the amount of calculation, and uses the fast forward elimination method based on probability distance to reduce the scenarios for the large-scale scenario set generated by the ARMA model. The process is as Figure 2 shown, including:

[0147] (1) Calculate the geometric distance between each pair of scenarios c and c' in the set of scenarios C;

[0148] (2) Select the scenario d with the minimum sum of probability distances from the remaining scenarios;

[0149] (3) Replace the scenario d with the scenario e in C that has the closest geometric distance to the scenario d;

[0150] (4) Add the probability of d to the probability of scenario e to eliminate d and form a new set of scenarios C';

[0151] (5) Check whether the number of remaining scenarios meets the requirements. If it meets the requirements, output the reduced set of typical scenarios; if not, go back to step (2) to continue the process until the requirements are met.

[0152] Due to the uncertainties in both the arc furnace load and the wind power output, the interaction effects of the power fluctuations of the two need to be considered in the optimization model. Let the total number of sequences to be simulated be N, and the power fluctuation sequences of the source and load are respectively obtained through the Monte Carlo algorithm and the ARMA model:

[0153]

[0154] In the formula: and are respectively the power fluctuation sequences of the arc furnace and the wind power under scenario n; and are respectively the power fluctuation values of the arc furnace and the wind power at time t under scenario n. At time t of scenario n, if and (regarded as negative load) are both positive or both negative, the two have a positive superposition effect, and the arc furnace fluctuation will absorb the wind power output fluctuation; if and (regarded as negative load) are one positive and one negative, the two have a positive superposition effect, pushing up the equivalent load and having an adverse impact on the stable operation of the power grid.

[0155] Generally speaking, the reserve capacity is obtained based on the unit parameters and it is difficult to accurately measure the capacity actually required by the system to accommodate fluctuations. When the selected reserve capacity is too large, the unit output is wasted, and when it is too small, the wind curtailment will be pushed up. The maximum and minimum values of the superposition effect of the source and load are used to describe the system reserve capacity:

[0156]

[0157] In the formula: P t u,LD and P t d,LD are respectively the upward and downward spinning reserves required to cope with the load forecast at time t; P t u,FW and They are the up and down spinning reserves required to cope with the superposition of the arc furnace load and wind power output fluctuations during period t; N G is the number of thermal power units in the system; is the start-stop state of thermal power unit i during period t; and are the maximum and minimum outputs of thermal power unit i, respectively; is the actual output of thermal power unit i during period t.

[0158] In step S3, based on the generated scenario set, an adaptive electricity price mechanism for smelting-type high-energy-consuming loads considering wind power output changes and source-load fluctuations is proposed to encourage steel enterprises to participate in demand response.

[0159] At present, the penetration rate of new energy sources such as wind power in the power system is increasing continuously. Its uncertainty poses a serious threat to the safe and stable operation of the power system, and also makes the existing fixed time-of-use electricity price division method often not in line with the actual situation of the power system, and cannot ensure effective incentives for users to respond to dispatching when there are dispatching requirements such as peak shaving or wind power accommodation in the power system. The present invention proposes an adaptive electricity price mechanism for smelting-type high-energy-consuming loads oriented to wind power accommodation. On the basis of the original time-of-use electricity price, the influence of the equivalent load containing wind power and the uncertainty fluctuations of source-load in each period is introduced, and the expression is as follows:

[0160]

[0161] In the formula: is the adaptive electricity price of smelting-type high-energy-consuming loads during period t; and are the conversion coefficient of the standard time-of-use electricity price for industrial users and the standard time-of-use electricity price for industrial users during period t, respectively; and are the equivalent load influence coefficient and the equivalent load compensation electricity price during period t, respectively; and are the fluctuation compensation coefficient and the fluctuation compensation electricity price during period t, respectively; and are the average values of the generated arc furnace load fluctuation scenario set and wind power output fluctuation scenario set during period t; T max is for all scenario sets the maximum period; and are the average values of the large-scale wind power output fluctuation and the arc furnace load fluctuation scenario set at T max respectively.

[0162] In step S4, an optimal scheduling model considering the interests of multiple parties is established, and the artificial bee colony algorithm is used to solve the optimal scheduling plan. To ensure the economic and stable operation of the power system containing wind power, the following objective function and constraint conditions are established:

[0163] Objective function 1: The lowest system operation cost. The system operation cost mainly comes from thermal power units and wind farms, and the expression is as follows:

[0164]

[0165] In the formula: a i , b i and c i are the unit operation cost coefficients; N W is the number of wind farms; μ co,W is the wind farm operation cost coefficient; is the wind power output of wind farm j at time t.

[0166] Objective function 2: The lowest curtailment penalty. The curtailment penalty reflects the system curtailment volume, and the expression is:

[0167]

[0168] In the formula: μ abd,W is the wind farm curtailment penalty coefficient.

[0169] Objective function 3: The lowest electricity purchase cost of iron and steel enterprises. The electricity purchase cost reflects the enthusiasm of high-energy-consuming loads such as smelting for response dispatching, and the expression is:

[0170]

[0171] System power balance constraint:

[0172]

[0173] Thermal power unit output power upper and lower limit constraints:

[0174]

[0175] In the formula: and are the upper and lower limits of the output power of thermal power unit i at time t.

[0176] Wind power output constraint:

[0177]

[0178] In the formula: is the predicted value of the output power of wind farm i at time t.

[0179] Thermal power unit ramping constraint:

[0180]

[0181] In the formula: and are the upper and lower limits of the ramp power of thermal power unit \(i\) during period \(t\).

[0182] Minimum start-up and shut-down time constraint of thermal power unit:

[0183]

[0184] where: \(T\) i,on,min and \(T\) i,off,min are the minimum start-up and shut-down times of thermal power unit \(i\).

[0185] Since the dimensions of each objective function are the same, there is no need for normalization. The multi-objective optimization problem can be directly converted into a single-objective optimization problem by adding the three objective functions. The converted total objective function \(F\) is to minimize the total cost of the system, and the expression is:

[0186] \(\min F = F_1 + F_2 + F_3\ (33)\)

[0187] Since the model of the present invention contains a non-linear part of the multiplication of 0-1 variables and integer variables, the artificial bee colony algorithm (ABC) is used to solve the optimization model. This algorithm is a meta-heuristic algorithm based on bee colonies, which can solve non-linear programming problems, avoid converting non-linear problems into linear problems, and has good solving performance and convergence ability, and is suitable for solving the non-linear optimization model proposed by the present invention. The flow chart of using the ABC algorithm to solve the source-load coordinated dispatch model of a wind power integrated power system considering high-energy-consuming loads in smelting is as Figure 3 shown, including:

[0188] (1) Initialize the parameters in the optimization model to generate an initial population containing the solutions of the source-load coordinated dispatch model of a wind power integrated power system considering high-energy-consuming loads in smelting.

[0189] (2) Calculate the fitness values in the initial population of the source-load coordinated dispatch model of a wind power integrated power system considering high-energy-consuming loads in smelting, providing a basis for subsequent sorting and selection.

[0190] (3) The leading bee randomly generates a new solution (i.e., a new food source) of the source-load coordinated dispatch model of a wind power integrated power system considering high-energy-consuming loads in smelting according to the original position, and makes a greedy selection: calculate the fitness value of the new solution and evaluate it. If the fitness value of the new solution is better than the old solution, the leading bee updates the old solution to the new solution. Otherwise, the old solution is retained.

[0191] (4) The follower bee performs roulette according to the fitness of the solution, calculates the selection probability of each solution of the source-load coordinated dispatch model of a wind power integrated power system considering high-energy-consuming loads in smelting, and judges whether to update the optimal solution variable to the new solution according to the probability.

[0192] (5) The scout bee determines whether an update has occurred within the exploration limit times. If there is no update, it is determined that the algorithm has fallen into a local optimal solution. Then, the original solution is discarded, and the leading bee corresponding to this solution is changed into a scout bee, and a new solution is generated to replace the original solution, that is, the original leading bee.

[0193] (6) Determine whether the end condition is reached. If so, output the optimal solution of the source-load coordinated scheduling model of the power system with wind power considering the participation of high-energy-consuming loads in the smelting category. Otherwise, return to (3) to continue the loop.

Claims

1. A method for coordinated dispatching of wind power systems with high energy loads in consideration of smelting, characterized in that: include: Step S1, analyzing the dispatching demand of the wind power power system, and establishing a model that considers the uncertainty of the arc furnace load and wind power output based on the dispatchable characteristics of the arc furnace load; Step S2, using a data-driven approach to generate large-scale scenarios, analyzing the combined effects of uncertainty of arc furnace load and wind power output, and determining the reserve capacity required by the power system to cope with uncertain fluctuations; Step S3, proposing an adaptive electricity price mechanism for high-energy-load smelting that takes into account wind power output changes and source-load fluctuations, to encourage steel companies to participate in demand response; Step S4, establishing an optimization scheduling model that takes into account the interests of multiple parties, and using an artificial bee colony algorithm to solve the optimal scheduling solution.

2. The source-load coordinated dispatching method for a wind power system taking into account the participation of high-energy loads of smelting according to claim 1 is characterized in that: In step S1, the dispatching demand of the wind power power system is analyzed, and based on the dispatchable characteristics of the arc furnace load, a model that considers the uncertainty of the arc furnace load and wind power output is established; high-energy load enterprises are large-scale power users, among which the arc furnace as the main load has great advantages in participating in wind power consumption: first, the arc furnace has fast regulation and can respond to wind power fluctuations in a short time; second, the arc furnace has large power units, which can transfer the load to the peak period of wind power output to consume wind power; Define the equivalent load: P t equ =P t LD -P t W (1) Where: P t equ , P t LD and P t W are the equivalent load, base load and wind power output in period t respectively; When the arc furnace load does not participate in the dispatch, if the equivalent load in period t is less than the total minimum output of the thermal power units When the power system can only guarantee safe operation by abandoning wind, the abandoned wind volume for: Where: T and Δt are the number and duration of scheduling periods respectively. After high energy load participates in regulation, the load in other periods can be transferred to this period, and the abandoned air volume P t W,abd for: Where: N F is the number of electric arc furnaces; is the planned operating power of arc furnace i during period t. From the above equations (2) and (3), it can be seen that the additional wind power consumption of the power system after high-energy load participates in the dispatch is P W，abd for: Therefore, the participation of high-energy load in regulation can effectively improve the power system's ability to absorb wind power; When regulating the power of an arc furnace, the power change is not continuous, but rather a discrete change based on the position of the transformer tap; The planned power of the electric arc furnace is: Where: is the start and stop status of arc furnace i in time period t; α i is the percentage of initial power of arc furnace i to rated power; is the transformer tap position of arc furnace i in time period t; is the total number of tap positions of transformer i of electric arc furnace; is the power of the transformer tap of arc furnace i when it is in the initial position; Due to the instability of the arc inside the arc furnace and the collapse of materials during operation, the active power of the arc furnace is uncertain. The maximum active power fluctuation of some arc furnaces can reach more than 20% of the planned operating power, so its impact on the power system dispatch cannot be ignored. The present invention describes the fluctuation of the electric arc furnace through normal distribution, providing a reference for selecting the system spare capacity; assuming that the fluctuation of the electric arc furnace To obey the expectation of 0 and the variance of The normal distribution is: In order to generate a distribution function that is more in line with the actual situation of the electric arc furnace, the standard normal distribution function with a confidence interval of 95% and an interval width of 20% of the installed capacity of the equipment is taken as the probability function of the fluctuation of the active power of the electric arc furnace. The actual operating power of the electric arc furnace after adding the fluctuation is: The following constraints need to be met to adjust the arc furnace load: Steel enterprises must complete production tasks during the scheduling cycle, and their total energy consumption must meet the minimum production energy consumption constraint: Where: Minimum production energy consumption for steel enterprises; The smelting time of one arc furnace cannot be too long or too short: Where: and They are respectively the longest and shortest smelting time for one furnace. The tap of the arc furnace transformer cannot be adjusted across levels: Where: τ is a multiple of the time ε required to adjust the tap once. Since the operation time of an arc furnace is relatively long, and the adjustment process of the transformer tap position is relatively fast, it is difficult to observe the hierarchical adjustment process of the tap from the measured data. The duration of a period in the scheduling cycle is consistent with: ε=Vt (12) Therefore, this constraint does not need to be considered in the following.

3. The source-load coordinated dispatching method for a wind power system taking into account the participation of high-energy loads of smelting according to claim 1 is characterized in that: In step S2, a data-driven method is used to generate large-scale scenarios, analyze the superposition of uncertainty between arc furnace load and wind power output, and determine the reserve capacity required by the power system to cope with uncertain fluctuations; since wind power output is affected by weather, its predicted data often deviates from the actual value, i.e., uncertainty. If the predicted data is directly used to formulate a scheduling plan, it does not conform to the actual situation. The present invention generates multiple sets of scenario sets based on wind power output prediction as the basic data set, analyzes the source-load superposition uncertainty, and selects typical scenarios to formulate an optimized scheduling plan; The regression moving average model is a high-precision time series model suitable for simulating wind power uncertainty; the expression is: Where: P t W is the wind power output during period t; is the autoregressive parameter; is the sliding average parameter; ε t is the random disturbance of the time series at time t; a is the autoregressive order; b is the sliding average order; ΔP t W is the fluctuation of wind power output during period t; In order to eliminate similar scenes in the wind power scene set to generate typical scenes, the present invention comprehensively considers data validity and computational complexity, and adopts a fast predecessor elimination method based on probability distance to reduce scenes in the large-scale scene set generated by the ARMA model, including: (1) Calculate the geometric distance between each pair of scenes c and c' in the scene set C; (2) Select the scene d with the smallest sum of probability distances to the remaining scenes; (3) Replace scene d with scene e in C that has the closest geometric distance to scene d; (4) Add the probability of d to the probability of scene e to eliminate d and form a new scene set C'; (5) Whether the number of remaining scenes meets the requirement, if it does, the reduced set of typical scenes is output; if not, the process returns to step (2) and continues until the requirement is met. Since there are uncertainties in both arc furnace load and wind power output, the interaction between the power fluctuations of the two needs to be considered in the optimization model. Assuming that the total number of sequences to be simulated is N, the source-load power fluctuation sequence is simulated by the Monte Carlo algorithm and the ARMA model respectively: Where: and They are the arc furnace and wind power fluctuation sequences under scenario n respectively; and are the fluctuation values ​​of arc furnace and wind power in period t under scenario n, respectively; In the t period of scenario n, if and When both are positive or negative, the two have a positive superposition effect, and the fluctuation of the arc furnace will absorb the fluctuation of wind power output; if and When one is positive and the other is negative, the two have an inverse superposition effect, pushing up the equivalent load and having an adverse effect on the stable operation of the power grid; Generally speaking, the reserve capacity is obtained based on the unit parameters, which makes it difficult to accurately measure the capacity actually required by the system to accommodate fluctuations. When the selected reserve capacity is too large, the unit output is wasted, and when it is too small, the wind curtailment is increased. The maximum and minimum values ​​of the source-load superposition effect are used to describe the system reserve capacity: Where: P t u,LD and P t d,LD are the upper and lower spinning reserves required to cope with load forecast in period t; P t u,FW and They are the upper and lower rotating reserves required to cope with the combined effects of arc furnace load and wind power output fluctuations during period t; N G is the number of thermal power units in the system; is the start and stop status of thermal power unit i in period t; and are the maximum and minimum outputs of thermal power unit i respectively; is the actual output of thermal power unit i in period t.

4. The source-load coordinated dispatching method for a wind power system taking into account the participation of high-energy loads of smelting according to claim 1 is characterized in that: In step S3, based on the generated scenario set, an adaptive electricity price mechanism for high-energy loads of smelting is proposed that takes into account the change of wind power output and source-load fluctuations, so as to encourage steel enterprises to participate in demand response; At present, the penetration rate of new energy sources such as wind power in the power system is increasing. Its uncertainty poses a serious threat to the safe and stable operation of the power system. It also makes the existing fixed time-of-use electricity price division method often not in line with the actual situation of the power system, and cannot guarantee that when the power system has peak load regulation or wind power consumption and other dispatching needs, it will effectively motivate users to guide them to respond to dispatch; the present invention proposes an adaptive electricity price mechanism for smelting high-energy loads facing wind power consumption. On the basis of the original time-of-use electricity price, this mechanism introduces the influence of the fluctuation of wind power equivalent load and source load uncertainty in each period, and the expression is as follows: Where: It is the adaptive electricity price for high energy load of smelting in period t; and are the conversion coefficient of the standard time-of-use electricity price for industrial users in period t and the standard time-of-use electricity price for industrial users respectively; and are the equivalent load impact coefficient and equivalent load compensation electricity price during period t respectively; and are the fluctuation compensation coefficient and fluctuation compensation electricity price in period t respectively; and are the average values ​​of the generated arc furnace load fluctuation scenario set and wind power output fluctuation scenario set in period t; T max Set for all scenes The largest time period; and The large-scale wind power output fluctuation and arc furnace load fluctuation scenarios are set in T max The average value below.

5. The source-load coordinated dispatching method for a wind power system taking into account the participation of high-energy loads of smelting according to claim 1 is characterized in that: In step S4, an optimization dispatching model that takes into account the interests of multiple parties is established, and an artificial bee colony algorithm is used to solve the optimal dispatching scheme; in order to ensure the economic and stable operation of the wind power power system, the following objective function and constraints are established: Objective function 1: Minimum system operation cost; The system operation cost mainly comes from thermal power units and wind farms, and the expression is as follows: Where: a i 、b i and c i is the unit operation cost coefficient; N W is the number of wind farms; μ co,W is the wind farm operation cost coefficient; is the wind power output of wind farm j in period t; Objective function 2: Minimum penalty for wind curtailment; The wind curtailment penalty reflects the amount of wind curtailment in the system, and the expression is: Where: μ abd,W is the wind farm abandonment penalty coefficient; Objective function 3: The electricity purchase cost of steel enterprises is the lowest; The electricity purchase cost reflects the enthusiasm of smelting high-energy load response dispatch, and the expression is: System power balance constraints: The upper and lower limits of thermal power unit output power are as follows: Where: and are the upper and lower limits of the output power of thermal power unit i during period t; Wind power output constraints: Where: is the predicted output power of wind farm i in period t; Thermal power unit climbing constraints: Where: and are the upper and lower limits of the ramp power of thermal power unit i during period t; Minimum start and stop time constraints for thermal power units: Where: T i,on,min and T i,off,min is the minimum on / off time of thermal power unit i; Since the dimensions of each objective function are consistent, there is no need to normalize them. Simply adding the three objective functions can convert the multi-objective optimization problem into a single-objective optimization problem. The converted sum objective function F is the lowest total cost of the system, and the expression is: minF=F1+F2+F3(33) Since the model of the present invention contains a nonlinear part of multiplication of 0-1 variables and integer variables, an artificial bee colony algorithm is used to solve the optimization model; the process of using the artificial bee colony algorithm to solve the source-load coordinated dispatch model of the wind power power system with high energy loads participating in smelting includes: (1) Initializing the parameters in the optimization model to generate an initial population of solutions for the source-load coordinated dispatch model of the wind power system that takes into account the participation of high-energy loads such as smelting. (2) Calculate the fitness value of the initial population in the source-load coordinated dispatch model of the power system including wind power and the participation of high-energy loads such as smelting, and provide a basis for subsequent sorting and selection; (3) Based on the original position, the leading bee randomly generates a new solution for the source-load coordinated dispatch model of the power system containing wind power and taking into account the participation of high-energy loads such as smelting, and performs greedy selection: the fitness value of the new solution is calculated and evaluated. If the fitness value of the new solution is better than the old solution, the leading bee updates the old solution to the new solution; otherwise, the old solution is retained. (4) Follower bees play roulette according to the fitness of the solution, calculate the selection probability of each solution of the source-load coordinated dispatch model of the wind power power system taking into account the participation of high-energy loads such as smelting, and determine whether to update the optimal solution variable to a new solution based on the probability; (5) The scout bee determines whether it has been updated within the exploration limit. If not, the algorithm is considered to have fallen into a local optimal solution. The original solution is discarded and the leader bee corresponding to the solution is changed to a scout bee, generating a new solution to replace the original solution, i.e., the original leader bee. (6) Determine whether the termination condition is met. If so, output the optimal solution of the source-load coordinated dispatch model of the power system including wind power and the participation of high-energy loads such as smelting. Otherwise, return to (3) and continue the cycle.

Images