A steel-power dispatching collaborative optimization method based on an improved resource task network model

Abstract

The application discloses a steel-electricity dispatching collaborative optimization method based on an improved resource task network model, and belongs to the field of industrial production dispatching and optimal operation of a power system. The method sets a production decision variable of a furnace period in the resource task network model, and reconstructs a task execution and product completion relationship, so that the production furnace period, task time sequence, equipment power and steel plant power consumption are determined simultaneously. Then, the steel plant power consumption is mapped to an electricity system dispatching period, and an optimal dispatching model containing thermal power, wind power, conventional load and a power transmission line is established. Finally, a joint solution is obtained by taking the sum of the steel plant steelmaking benefit and the power system benefit as a target, so that a collaborative plan of production tasks and power source output is formed. The application can avoid simplifying the steel plant into an aggregated load which is separated from a production process, and can play a regulating role of the steel load on wind power consumption while ensuring the multi-stage production executability.

Description

Technical Field

[0001] This invention relates to the field of industrial production scheduling and power system optimization operation technology, and in particular to a steel-power scheduling collaborative optimization method based on an improved resource task network model. Background Technology

[0002] As the proportion of renewable energy sources such as wind power integrated into the power system continues to increase, their volatility and uncertainty place higher demands on power balance, reserve configuration, and wind curtailment control. Relying solely on conventional generating units for regulation not only increases system operating costs but also makes it difficult to fully realize the regulatory value of controllable demand-side loads. Steel enterprises, with their large power consumption and time-transferable production tasks, are important industrial loads that can participate in the optimized dispatch of the power system.

[0003] Short-process steel production typically includes stages such as electric arc furnace melting, argon-oxygen decarburization, ladle refining, continuous casting, and rolling. These stages are coupled with factors such as equipment occupancy, material handling, task priority, transfer time limits, and product delivery. Changes in the power of any equipment or the start time of a task can have a ripple effect on subsequent processes. Resource-Task-Network (RTN) models can describe these discrete production processes using resource states and task execution. However, the scale of conventional RTN models is usually consistent with the preset number of production furnaces, meaning the number of furnaces is already determined before scheduling optimization. Therefore, they primarily optimize energy costs under a given production volume, making it difficult to simultaneously reflect steel sales revenue, underproduction costs, and production volume decisions.

[0004] Existing methods for involving steel loads in power system dispatch often treat electric arc furnaces or steel plants as aggregated loads of a given capacity, adjusting their power consumption curves only within upper and lower power limits. This approach makes it difficult to ensure that the resulting load curves are consistent with the steel plant's multi-stage production tasks, resource utilization, and final delivery requirements. Furthermore, it fails to determine the number of furnace runs to be produced, the associated tasks for each furnace run, and the power generation output within the same optimization process. Therefore, there is an urgent need to construct a collaborative optimization method that unifies variable batch steel production scheduling with power system operational constraints, ensuring production feasibility while generating a steel plant power consumption and unit output plan oriented towards wind power integration. Summary of the Invention

[0005] The purpose of this invention is to overcome the problems of pre-determined production furnace numbers, disconnect between steel plant load and internal production processes, and independent optimization of steel production scheduling and power system scheduling in existing technologies. This invention proposes a collaborative optimization method for steel-power scheduling based on an improved resource-task network model. This method embeds furnace production decisions into the RTN model, jointly determining production quantities, task arrangements, steel plant power consumption trajectories, thermal power unit output, and wind power unit output.

[0006] This invention proposes a collaborative optimization method for steel-electricity dispatch based on an improved resource-task network model, comprising the following steps: (1) Obtaining data on steel production and power system dispatching Acquire candidate furnace sets, processing equipment and product resource sets, processing and transfer task sets, equipment power and processing time range, resource generation and consumption coefficients, maximum transfer time, steel sales price, contract production target, production shortfall penalty coefficient, and time-of-use electricity price; simultaneously acquire the power system's busbars, lines, conventional loads, thermal power units, wind power units, unit operating costs, wind power deviation penalties, output upper and lower limits, ramping capability, and spinning reserve parameters.

[0007] (2) Establish the time scale correspondence between steel production and power system Let the time interval for steel production scheduling be . The power system dispatching time interval is and order This represents the number of steel production scheduling intervals contained within a power system scheduling interval; for each power system scheduling period, its corresponding set of steel production scheduling periods is determined: (1) in, Index for power system dispatching periods; To coincide with power system dispatching periods The corresponding set of steel production scheduling time period indexes; This refers to the number of steel production scheduling intervals contained within a single power system scheduling interval.

[0008] The corresponding steel production scheduling time index satisfies: (2) in, This is an index for steel production scheduling periods.

[0009] The above mapping aggregates the electricity consumption of steel equipment at fine time scales to the power system dispatch time scale.

[0010] (3) Establish a steel production scheduling model based on an improved resource task network. Electric arc furnaces, argon-oxygen decarburization units, ladle refining units, continuous casting machines, rolling mills, and products at each stage are represented as resources. Processing operations for each heat and transfer operations between adjacent processes are represented as tasks. Heat production decision variables, task initiation variables, task execution status variables, and resource status variables are set to ensure that whether a candidate heat is put into production, whether related tasks are executed, and the status of equipment and products are linked within the same resource-task network model.

[0011] (3-1) Constructing steelmaking efficiency targets The steelmaking profitability target is comprised of the energy costs of steel production, steel sales revenue, and penalties for failing to meet contract production targets. (3) in, For the steel mill's steelmaking efficiency; This refers to the set of time periods for steel production scheduling. and Time periods Time-of-use electricity pricing and electricity consumption at steel mills; For the candidate furnace set; h For furnace number index; For the production decision variable of each furnace, a value of 1 indicates the number of furnaces to be produced. h A value of 0 indicates that no batches will be produced. h ; This refers to the sales price of steel per furnace batch; To achieve the contract production target; This is the penalty coefficient for underproduction.

[0012] In the above formula, the first term is the electricity purchase cost for each steel production period, the second term is the steel sales revenue generated by the selected furnace, and the third term is the penalty for underproduction when the actual number of furnaces produced is lower than the contracted production target. Therefore, the number of furnaces is no longer a parameter given before optimization, but is determined simultaneously with task scheduling.

[0013] (3-2) Constructing resource balance and electricity consumption constraints The state of any resource in the current time period is determined by the state in the previous time period and the amount of resources generated or consumed by started tasks: (4) in, For time period resources State value; and These are the resource index and the task index, respectively. and These are respectively a resource set and a task set; This is a variable used to start the task; a value of 1 indicates the task is complete. During the period start up; This is an index of the latency after the task starts; For the task Duration of action; For the task After starting Resources in each time period The generation or consumption coefficient.

[0014] The electricity consumption of each processing task during the steel production scheduling period is as follows: (5) in, For the task After starting The energy consumption coefficient corresponding to each time period.

[0015] (3-3) Constructing furnace-level task activation and product delivery constraints When the production decision variable for a furnace batch is 1, all processing and transfer tasks corresponding to that furnace batch are executed once each; when the production decision variable for a furnace batch is 0, the tasks associated with that furnace batch are not executed. (6) in, In order to be compatible with the furnace Associated task index; For furnace A set of associated processing and transfer tasks.

[0016] The corresponding transfer task is triggered immediately after the previous processing task is completed: (7) in, This is the product resource index corresponding to the starting location of the transfer task.

[0017] The total transfer and waiting time between adjacent processing stages shall not exceed the maximum allowable transfer time. (8) in, This refers to the time interval for steel production scheduling. This is the product resource index corresponding to the termination location of the transfer task; For resources Transit time; For resources Maximum allowed transit and total waiting time.

[0018] The final products of the selected heats are produced before the end of the steel production scheduling cycle: (9) in, For furnace Product resource index corresponding to the final production stage; This refers to the final period of the steel production scheduling cycle.

[0019] (3-4) Construct constraints on equipment power, processing time and energy demand. For equipment with power regulation capabilities, the power consumption during processing is within the allowable range: (10) in, For the task During the period The active power; This is a task execution status variable; a value of 1 indicates the task is running. During the period In the process of processing; and Tasks The lower and upper limits of power; This is the set of processing tasks corresponding to the power adjustable device.

[0020] The task execution status is determined by the relationship between task initiation and the initiation of subsequent transfer tasks, thus ensuring that the task continues to execute until completion once it begins. (11) in, For processing tasks Subsequent transfer task index.

[0021] The duration of the processing task corresponding to the adjustable equipment meets the allowable upper and lower bounds: (12) in, and Tasks The shortest and longest processing times; The steel production scheduling time period index used for summation.

[0022] Its maximum processing time and minimum processing time satisfy: (13) To ensure processing quality, the cumulative energy consumption for the task shall not be less than the rated energy requirement: (14) in, To complete the task Minimum cumulative energy required.

[0023] When the equipment power is adjustable, its power consumption during the time period is calculated according to the actual power trajectory of the task: (15) in, To incorporate the actual power trajectory of adjustable equipment into the steel mill's time period Electricity consumption; and The difference set represents the set of tasks in the task set excluding tasks for power-adjustable devices. Furthermore, in this invention, the electric arc furnace is a device with power adjustment capabilities.

[0024] (4) Establish the optimal dispatch model of the power system By injecting the aggregated electricity consumption of steel mills during various power system dispatch periods as bus load, an optimal power system dispatch model is established, comprising thermal power units, wind power units, conventional loads, transmission lines, and spinning reserves. The power system objective is to deduct the operating costs of thermal power units and the penalty costs for unused wind power from electricity sales revenue. (16) in, For the benefit of the power system; For the set of power system dispatch periods; Index for thermal power units or wind power units; , and These are respectively the thermal power unit set, the wind power unit set, and the conventional load set; and Time periods Operating costs of thermal power units and penalty costs for unutilized wind power; For time period The electricity sales price; For regular load indexing; For normal load During the period Electricity consumption; For steel mills during the period The equivalent electricity consumption.

[0025] The operating cost of thermal power units is expressed as a quadratic function related to the active power output of the unit: (17) in, For thermal power units During the period Those who have made meritorious contributions; , and thermal power units The coefficients of the constant term, linear term, and quadratic term of the operating cost function.

[0026] The penalty cost for unutilized wind power is determined by the difference between the predicted available wind power and the actual output: (18) in, For wind turbines The predicted available power; For wind turbines During the period The actual contribution of the individual; This represents the penalty coefficient for unutilized wind power.

[0027] The steel mill's electricity consumption at fine time scales is aggregated into equivalent electricity consumption during power system dispatch periods according to time mapping: (19) Among them, the power system dispatch period and the corresponding steel production dispatch period are associated through time scale mapping; the equivalent electricity consumption of steel plants is the aggregate value of electricity consumption of each steel production dispatch period within a power system dispatch interval.

[0028] Power system constraints include power balance constraints, line transmission capacity constraints, output constraints of thermal power units and wind power units, thermal power unit ramping constraints, and system spinning reserve constraints. (20) in, To connect load nodes The steel mill during the period The equivalent active power; For normal load During the period The active power.

[0029] (twenty one) in, and These are the transmission line index and the bus index, respectively. For the set of busbars; For the line Relative to the busbar DC power transfer factor; For the line Transmission capacity.

[0030] (twenty two) in, and thermal power units The minimum and maximum active power outputs.

[0031] (twenty three) (twenty four) in, This refers to the power system dispatch time interval; and thermal power units The limits for downward and upward ramp rates.

[0032] (25) in, and thermal power units During the period Provided up-and-down rotational reserve capacity.

[0033] (26) (27) in, and Time periods The system rotates up and down to meet backup requirements.

[0034] (5) Establish a collaborative optimization scheduling model for the steel-power system. The sum of the steelmaking benefits of the steel plant and the revenue of the power system is used as the synergistic optimization objective: (28) in, This represents the total revenue of the steel plant and the power system.

[0035] The collaborative optimization constraints consist of the steel production scheduling constraints of the improved RTN model in step (3) and the optimal power system scheduling constraints in step (4). The electricity consumption of the steel plant is simultaneously incorporated into the steelmaking energy cost, power system power balance, and line power flow constraints, so that furnace selection, production tasks, electric arc furnace power, thermal power output, and wind power output are linked in the same optimization problem.

[0036] (6) Solve and generate a cooperative scheduling plan. The above model is formed into a mixed integer optimization problem and solved. The output includes the selected production furnace, the start and end times of processing and transfer tasks for each furnace, equipment occupancy status, steel plant time-series electricity consumption, thermal power unit output, wind power unit output, line power and spinning reserve plan. The equipment-level task plan is then sent to the steel plant production scheduling system, and the unit output plan is sent to the power system scheduling system.

[0037] The present invention has the following beneficial effects: This invention incorporates production quantity into the RTN (Resource-to-Number) by using furnace production decision variables, reconstructing the relationship between task execution and product completion. This enables steel mills to simultaneously determine the furnaces to be put into production and the corresponding tasks. Furthermore, it extends the optimization of energy costs under a given production volume to the optimization of steelmaking efficiency under the combined effects of steel sales revenue, energy consumption costs, and penalties for underproduction.

[0038] This invention does not simplify steel plants into aggregated adjustable loads detached from the production process. Instead, it utilizes resource balancing, continuous task execution, transfer time limits, product delivery, equipment power, and energy demand constraints to generate executable load trajectories, thereby ensuring that the power consumption plans of steel plants participating in power system dispatch are consistent with multi-stage production requirements.

[0039] This invention places steel production scheduling and power system active power scheduling within the same optimization framework. It can adjust the production sequence based on the available wind power output and system operating status, coordinating thermal power and wind power output while meeting the steel production quality and delivery requirements, and providing an feasible scheduling scheme for industrial loads to participate in wind power consumption. Detailed Implementation

[0040] The following example, using a short-process steel plant and a power system comprising thermal power units and wind power units, further illustrates the present invention. These embodiments are used to illustrate the technical approach of the present invention and do not constitute a limitation on the scope of protection.

[0041] A collaborative optimization method for steel-electricity dispatch based on an improved resource-task network model includes the following steps: (1) Obtaining data on steel production and power system dispatching Acquire candidate furnace sets, processing equipment and product resource sets, processing and transfer task sets, equipment power and processing time range, resource generation and consumption coefficients, maximum transfer time, steel sales price, contract production target, production shortfall penalty coefficient, and time-of-use electricity price; simultaneously acquire the power system's busbars, lines, conventional loads, thermal power units, wind power units, unit operating costs, wind power deviation penalties, output upper and lower limits, ramping capability, and spinning reserve parameters.

[0042] (2) Establish the time scale correspondence between steel production and power system Let the time interval for steel production scheduling be . The power system dispatching time interval is and order This represents the number of steel production scheduling intervals contained within a power system scheduling interval; for each power system scheduling period, its corresponding set of steel production scheduling periods is determined: (1) in, Index for power system dispatching periods; To coincide with power system dispatching periods The corresponding set of steel production scheduling time period indexes; This refers to the number of steel production scheduling intervals contained within a single power system scheduling interval.

[0043] The corresponding steel production scheduling time index satisfies: (2) in, This is an index for steel production scheduling periods.

[0044] The above mapping aggregates the electricity consumption of steel equipment at fine time scales to the power system dispatch time scale.

[0045] (3) Establish a steel production scheduling model based on an improved resource task network. Electric arc furnaces, argon-oxygen decarburization units, ladle refining units, continuous casting machines, rolling mills, and products at each stage are represented as resources. Processing operations for each heat and transfer operations between adjacent processes are represented as tasks. Heat production decision variables, task initiation variables, task execution status variables, and resource status variables are set to ensure that whether a candidate heat is put into production, whether related tasks are executed, and the status of equipment and products are linked within the same resource-task network model.

[0046] (3-1) Constructing steelmaking efficiency targets The steelmaking profitability target is comprised of the energy costs of steel production, steel sales revenue, and penalties for failing to meet contract production targets. (3) in, For the steel mill's steelmaking efficiency; This refers to the set of time periods for steel production scheduling. and Time periods Time-of-use electricity pricing and electricity consumption at steel mills; The set of candidate furnace batches; h is the furnace batch index; For the production decision variable of furnace number, a value of 1 indicates that furnace number h is produced, and a value of 0 indicates that furnace number h is not produced. This refers to the sales price of steel per furnace batch; To achieve the contract production target; This is the penalty coefficient for underproduction.

[0047] In the above formula, the first term is the electricity purchase cost for each steel production period, the second term is the steel sales revenue generated by the selected furnace, and the third term is the penalty for underproduction when the actual number of furnaces produced is lower than the contracted production target. Therefore, the number of furnaces is no longer a parameter given before optimization, but is determined simultaneously with task scheduling.

[0048] (3-2) Constructing resource balance and electricity consumption constraints The state of any resource in the current time period is determined by the state in the previous time period and the amount of resources generated or consumed by started tasks: (4) in, For time period resources State value; and These are the resource index and the task index, respectively. and These are respectively a resource set and a task set; This is a variable used to start the task; a value of 1 indicates the task is complete. During the period start up; This is an index of the latency after the task starts; For the task Duration of action; For the task After starting Resources in each time period The generation or consumption coefficient.

[0049] The electricity consumption of each processing task during the steel production scheduling period is as follows: (5) in, For the task After starting The energy consumption coefficient corresponding to each time period.

[0050] (3-3) Constructing furnace-level task activation and product delivery constraints When the production decision variable for a furnace batch is 1, all processing and transfer tasks corresponding to that furnace batch are executed once each; when the production decision variable for a furnace batch is 0, the tasks associated with that furnace batch are not executed. (6) in, In order to be compatible with the furnace Associated task index; For furnace A set of associated processing and transfer tasks.

[0051] The corresponding transfer task is triggered immediately after the previous processing task is completed: (7) in, This is the product resource index corresponding to the starting location of the transfer task.

[0052] The total transfer and waiting time between adjacent processing stages shall not exceed the maximum allowable transfer time. (8) in, This refers to the time interval for steel production scheduling. This is the product resource index corresponding to the termination location of the transfer task; For resources Transit time; For resources Maximum allowed transit and total waiting time.

[0053] The final products of the selected heats are produced before the end of the steel production scheduling cycle: (9) in, For furnace Product resource index corresponding to the final production stage; This refers to the final period of the steel production scheduling cycle.

[0054] (3-4) Construct constraints on equipment power, processing time and energy demand. For equipment with power regulation capabilities, the power consumption during processing is within the allowable range: (10) in, For the task During the period The active power; This is a task execution status variable; a value of 1 indicates the task is running. During the period In the process of processing; and Tasks The lower and upper limits of power; This is the set of processing tasks corresponding to the power adjustable device.

[0055] The task execution status is determined by the relationship between task initiation and the initiation of subsequent transfer tasks, thus ensuring that the task continues to execute until completion once it begins. (11) in, For processing tasks Subsequent transfer task index.

[0056] The duration of the processing task corresponding to the adjustable equipment meets the allowable upper and lower bounds: (12) in, and Tasks The shortest and longest processing times; The steel production scheduling time period index used for summation.

[0057] Its maximum processing time and minimum processing time satisfy: (13) To ensure processing quality, the cumulative energy consumption for the task shall not be less than the rated energy requirement: (14) in, To complete the task Minimum cumulative energy required.

[0058] When the equipment power is adjustable, its power consumption during the time period is calculated according to the actual power trajectory of the task: (15) in, To incorporate the actual power trajectory of adjustable equipment into the steel mill's time period Electricity consumption; and The difference set represents the set of tasks in the task set excluding tasks for power-adjustable devices. Furthermore, in this invention, the electric arc furnace is a device with power adjustment capabilities.

[0059] (4) Establish the optimal dispatch model of the power system By injecting the aggregated electricity consumption of steel mills during various power system dispatch periods as bus load, an optimal power system dispatch model is established, comprising thermal power units, wind power units, conventional loads, transmission lines, and spinning reserves. The power system objective is to deduct the operating costs of thermal power units and the penalty costs for unused wind power from electricity sales revenue. (16) in, For the benefit of the power system; For the set of power system dispatch periods; Index for thermal power units or wind power units; , and These are respectively the thermal power unit set, the wind power unit set, and the conventional load set; and Time periods Operating costs of thermal power units and penalty costs for unutilized wind power; For time period The electricity sales price; For regular load indexing; For normal load During the period Electricity consumption; For steel mills during the period The equivalent electricity consumption.

[0060] The operating cost of thermal power units is expressed as a quadratic function related to the active power output of the unit: (17) in, For thermal power units During the period Those who have made meritorious contributions; , and thermal power units The coefficients of the constant term, linear term, and quadratic term of the operating cost function.

[0061] The penalty cost for unutilized wind power is determined by the difference between the predicted available wind power and the actual output: (18) in, For wind turbines The predicted available power; For wind turbines During the period The actual contribution of the individual; This represents the penalty coefficient for unutilized wind power.

[0062] The steel mill's electricity consumption at fine time scales is aggregated into equivalent electricity consumption during power system dispatch periods according to time mapping: (19) Among them, the power system dispatch period and the corresponding steel production dispatch period are associated through time scale mapping; the equivalent electricity consumption of steel plants is the aggregate value of electricity consumption of each steel production dispatch period within a power system dispatch interval.

[0063] Power system constraints include power balance constraints, line transmission capacity constraints, output constraints of thermal power units and wind power units, thermal power unit ramping constraints, and system spinning reserve constraints. (20) in, To connect load nodes The steel mill during the period The equivalent active power; For normal load During the period The active power.

[0064] (twenty one) in, and These are the transmission line index and the bus index, respectively. For the set of busbars; For the line Relative to the busbar DC power transfer factor; For the line Transmission capacity.

[0065] (twenty two) in, and thermal power units The minimum and maximum active power outputs.

[0066] (twenty three) (twenty four) in, This refers to the power system dispatch time interval; and thermal power units The limits for downward and upward ramp rates.

[0067] (25) in, and thermal power units During the period Provided up-and-down rotational reserve capacity.

[0068] (26) (27) in, and Time periods The system rotates up and down to meet backup requirements.

[0069] (5) Establish a collaborative optimization scheduling model for the steel-power system. The sum of the steelmaking benefits of the steel plant and the revenue of the power system is used as the synergistic optimization objective: (28) in, This represents the total revenue of the steel plant and the power system.

[0070] The collaborative optimization constraints consist of the steel production scheduling constraints of the improved RTN model in step (3) and the optimal power system scheduling constraints in step (4). The electricity consumption of the steel plant is simultaneously incorporated into the steelmaking energy cost, power system power balance, and line power flow constraints, so that furnace selection, production tasks, electric arc furnace power, thermal power output, and wind power output are linked in the same optimization problem.

[0071] (6) Solve and generate a cooperative scheduling plan. The above model is formed into a mixed integer optimization problem and solved. The output includes the selected production furnace, the start and end times of processing and transfer tasks for each furnace, equipment occupancy status, steel plant time-series electricity consumption, thermal power unit output, wind power unit output, line power and spinning reserve plan. The equipment-level task plan is then sent to the steel plant production scheduling system, and the unit output plan is sent to the power system scheduling system.

[0072] The invention is illustrated below with specific embodiments. The short-process production line in these embodiments includes an electric arc furnace, an argon-oxygen decarburization unit, a ladle refining unit, a continuous casting machine, and a rolling mill. Scrap steel is first melted in the electric arc furnace, and then sequentially undergoes decarburization, refining, continuous casting, and rolling. Each heat of molten steel constitutes a production batch, and the processing stages and inter-stage transfer processes of each heat correspond to the processing tasks and transfer tasks in the resource task network model, respectively.

[0073] (1) Establish candidate furnaces and resource-task relationships A candidate furnace set is pre-defined instead of forcing all candidate furnaces to be mandatory for production. A binary production decision variable is set for each candidate furnace; when this variable is 1, the melting, decarburizing, refining, continuous casting, rolling, and all transfer tasks for that furnace are activated; when this variable is 0, none of the related tasks are scheduled. Equipment resources are used to limit task occupancy during the same time period, product resources are used to connect upstream and downstream processes, and the resource-task interaction coefficient is used to describe the generation or consumption of resources under different time delays after task initiation.

[0074] The steelmaking benefits are calculated according to formula (3).

[0075] After setting the contract production target, the model weighs the electricity price, steel selling price, and penalties for underproduction. If reducing a certain furnace batch can lower the cost of electricity purchase, but the resulting loss of sales revenue and penalties for underproduction are greater, then that furnace batch is still selected; otherwise, the model can choose not to produce that furnace batch, thus making the production quantity part of the scheduling outcome.

[0076] (2) Forming a steel plant load trajectory that satisfies the production coupling relationship. Resource balancing for each time period is performed according to formula (4).

[0077] The execution of tasks for each furnace is linked to the furnace production decision according to formula (6).

[0078] Then apply immediate transfer constraints, maximum transfer time constraints and final product completion constraints according to Equations (7) to (9) respectively to maintain process connection.

[0079] The upper and lower limits of power, processing duration and cumulative energy of adjustable equipment such as electric arc furnaces are constrained according to Equations (10), (12) and (14), respectively.

[0080] The actual electricity consumption for each time period is calculated according to formula (15).

[0081] The steel plant load trajectory obtained in this way is not given externally, but is determined by the selection of furnace batches, equipment occupancy, task coordination, and equipment power.

[0082] (3) Unify the time scale for steel production and power system dispatch. In one embodiment, the steel production scheduling interval is 5 minutes, and the power system scheduling interval is 15 minutes. Therefore, each power system scheduling period corresponds to three consecutive steel production scheduling periods. The correspondence between the two types of time period indexes is established using equations (1) and (2).

[0083] And according to formula (19), the electricity consumption of steel in the three sub-time periods is aggregated into the equivalent load of a power system dispatch period.

[0084] (4) Establish a power system operation model The power system consists of three thermal power units and one wind power unit. The equivalent load of the steel plant and the conventional load are substituted into equation (20) to satisfy the system power balance.

[0085] The active power transmission of the line adopts the DC power flow model shown in equation (21) and meets the corresponding line capacity limit.

[0086] Further, set the output range of thermal power units and wind power units, the climbing constraint of thermal power units and the rotational standby constraint of the system according to Equations (22) to (27).

[0087] The revenue of the power system is calculated according to formula (16).

[0088] Among them, the cost of thermal power and the penalty for unused wind power are calculated according to formula (17) and formula (18), respectively.

[0089] (5) Collaboratively solve and generate an execution plan According to formula (28), the benefits of steelmaking and the benefits of the power system are combined to form a synergistic optimization objective.

[0090] During the solution process, periods with higher available wind power output can influence the steel plant's task timing through power balancing and wind power penalty terms; steel plant production constraints limit the feasible range of load transfer. The results simultaneously provide the number of production furnaces, equipment tasks, and power output, eliminating the need to fix the steel plant load before power-side scheduling.

[0091] In the embodiment with a scheduling cycle from 00:00 to 06:00, the contract production target is set at 8 heats, and the model selects 6 heats for production, with each heat sequentially passing through five processing stages. Compared to optimizing the steel mill and power system separately, the total revenue increased from US$28,578.54 to US$47,892.92 after adopting the collaborative optimization of this invention, an increase of US$19,314.38; the steel mill's production tasks are shifted to periods with higher available wind power output, while reducing the penalty for unused wind power and the operating burden on thermal power units. The above results show that this invention can leverage the role of steel load in wind power consumption while maintaining the feasibility of the production plan.

Claims

1. A collaborative optimization method for steel-electricity dispatch based on an improved resource-task network model, characterized in that, Includes the following steps: (1) Obtain steel production and power system dispatch data, including candidate furnace batches, processing equipment, product resources, processing and transfer tasks, equipment power range, task duration, energy demand, process transfer time limit, steel sales price, time-of-use electricity price, contract production target and underproduction penalty parameters, as well as conventional load, thermal power unit and wind power unit operating parameters, wind power available output, transmission network topology, line impedance, line capacity and system reserve requirements; (2) Establish the time scale correspondence between steel production and power system, set steel production scheduling intervals and power system scheduling intervals respectively, determine the number of steel production scheduling periods included in each power system scheduling period, and establish the index mapping and data alignment relationship between the two types of scheduling periods; (3) Establish a steel production scheduling model based on an improved resource task network. Electric arc furnace, argon-oxygen decarburization unit, ladle refining unit, continuous casting machine, rolling mill and products at each stage are represented as resources. The processing operations of each furnace and the transfer operations between adjacent processes are represented as tasks. Set furnace production decision variables, task initiation variables, task execution status variables and resource status variables so that whether the candidate furnace is put into production, whether the relevant tasks are executed and the status of equipment and products are linked within the same resource task network model. (4) Aggregate the time-series electricity consumption of steel plants generated by the steel production scheduling model into the power system scheduling period, and construct an optimal power system scheduling model that includes thermal power units, wind power units, conventional loads and transmission lines; (5) Establish a collaborative optimization scheduling model for the steel-power system, with the sum of steelmaking benefits and power system revenue as the optimization objective. Incorporate steel sales revenue, steel production energy consumption costs, production shortfall penalties, thermal power unit operating costs and wind power penalty costs into the objective function. Combine steel production scheduling constraints with power system operating constraints to ensure that furnace selection, task sequence, equipment power, thermal power output, wind power output and line transmission power are coordinated and determined in the same optimization problem. (6) Solve the collaborative optimization scheduling model to obtain the candidate furnace commissioning results, the start and end times of processing and transfer tasks, equipment power trajectory, thermal power unit output, wind power unit output, line transmission power and spinning reserve plan, and generate a collaborative scheduling plan for the steel production scheduling system and the power system scheduling system.

2. The steel-power dispatching collaborative optimization method based on an improved resource-task network model according to claim 1, characterized in that, Step (1) includes: Acquire candidate furnace sets, processing equipment and product resource sets, processing and transfer task sets, equipment power and processing time range, resource generation and consumption coefficients, maximum transfer time, steel sales price, contract production target, production shortfall penalty coefficient, and time-of-use electricity price; simultaneously acquire the power system's busbars, lines, conventional loads, thermal power units, wind power units, unit operating costs, wind power deviation penalties, output upper and lower limits, ramping capability, and spinning reserve parameters.

3. The steel-power dispatching collaborative optimization method based on an improved resource task network model according to claim 2, characterized in that, Step (2) includes: setting the steel production scheduling time interval as... The power system dispatching time interval is and order This represents the number of steel production scheduling intervals contained within a power system scheduling interval; for each power system scheduling period, its corresponding set of steel production scheduling periods is determined: (1) in, Index for power system dispatching periods; To coincide with power system dispatching periods The corresponding set of steel production scheduling time period indexes; This refers to the number of steel production scheduling intervals contained within a single power system scheduling interval. The corresponding steel production scheduling time index satisfies: (2) in, Index for steel production scheduling periods; The above mapping aggregates the electricity consumption of steel equipment at fine time scales to the power system dispatch time scale.

4. The steel-power dispatching collaborative optimization method based on an improved resource task network model according to claim 3, characterized in that, Step (3) includes: (3-1) Constructing steelmaking efficiency targets The steelmaking profitability target is comprised of the energy costs of steel production, steel sales revenue, and penalties for failing to meet contract production targets. (3) in, For the steel mill's steelmaking efficiency; This is a set of time periods for steel production scheduling. and Time periods Time-of-use electricity pricing and electricity consumption at steel mills; The set of candidate furnace batches; h is the furnace batch index; For the production decision variable of furnace number, a value of 1 indicates that furnace number h is produced, and a value of 0 indicates that furnace number h is not produced. This refers to the sales price of steel per furnace batch; To achieve the contract production target; This is the penalty coefficient for underproduction; In equation (3), the first term is the electricity purchase cost for each steel production period, the second term is the steel sales revenue generated by the selected furnace, and the third term is the penalty for underproduction when the actual number of furnaces produced is lower than the contract production target. Thus, the number of furnaces is no longer a parameter given before optimization, but is determined simultaneously with the task scheduling. (3-2) Constructing resource balance and electricity consumption constraints The state of any resource in the current time period is determined by the state in the previous time period and the amount of resources generated or consumed by started tasks: (4) in, For time period resources State value; and These are the resource index and the task index, respectively. and These are respectively a resource set and a task set; This is a variable used to start the task; a value of 1 indicates the task is complete. During the period start up; This is an index of the latency after the task starts; For the task Duration of action; For the task After starting Resources in each time period The generation or consumption coefficient; The electricity consumption of each processing task during the steel production scheduling period is as follows: (5) in, For the task After starting Energy consumption coefficients for each time period; (3-3) Constructing furnace-level task activation and product delivery constraints When the production decision variable for a furnace batch is 1, all processing and transfer tasks corresponding to that furnace batch are executed once each; when the production decision variable for a furnace batch is 0, the tasks associated with that furnace batch are not executed. (6) in, In order to be related to the furnace Associated task index; For furnace A set of associated processing and transfer tasks; The corresponding transfer task is triggered immediately after the previous processing task is completed: (7) in, This is the product resource index corresponding to the starting location of the transfer task; The total transfer and waiting time between adjacent processing stages shall not exceed the maximum allowable transfer time. (8) in, This refers to the time interval for steel production scheduling. This is the product resource index corresponding to the termination location of the transfer task; For resources Transit time; For resources Maximum allowed transit and total waiting time. The final products of the selected heats are produced before the end of the steel production scheduling cycle: (9) in, For furnace Product resource index corresponding to the final production stage; This refers to the final period of the steel production scheduling cycle. (3-4) Construct constraints on equipment power, processing time and energy demand. For equipment with power regulation capabilities, the power consumption during processing is within the allowable range: (10) in, For the task During the period The active power; This is a task execution status variable; a value of 1 indicates the task is running. During the period In the process of processing; and Tasks The lower and upper limits of power; This refers to the set of processing tasks corresponding to power-adjustable devices. The task execution status is determined by the relationship between task initiation and the initiation of subsequent transfer tasks, thus ensuring that the task continues to execute until completion once it begins. (11) in, For processing tasks Subsequent transit task index; The duration of the processing task corresponding to the adjustable equipment meets the allowable upper and lower bounds: (12) in, and Tasks The shortest and longest processing times; The steel production scheduling time period index used for summation; Its maximum processing time and minimum processing time satisfy: (13) To ensure processing quality, the cumulative energy consumption for the task shall not be less than the rated energy requirement: (14) in, To complete the task Minimum cumulative energy required; When the equipment power is adjustable, its power consumption during the time period is calculated according to the actual power trajectory of the task: (15) in, To incorporate the actual power trajectory of adjustable equipment into the steel mill's time period Electricity consumption; and The difference set represents the set of tasks in the task set excluding tasks for power-adjustable equipment; the electric arc furnace is a device with power adjustment capability.

5. The steel-power dispatching collaborative optimization method based on an improved resource task network model according to claim 4, characterized in that, Step (4) includes: By injecting the aggregated electricity consumption of steel plants during various power system dispatch periods as bus load, an optimal power system dispatch model is established, consisting of thermal power units, wind power units, conventional loads, transmission lines, and spinning reserves. The power system objective is to deduct the operating costs of thermal power units and the penalty costs for unused wind power from the electricity sales revenue. (16) in, For the benefit of the power system; For the set of power system dispatch periods; Index for thermal power units or wind power units; , and These are respectively the thermal power unit set, the wind power unit set, and the conventional load set; and Time periods Operating costs of thermal power units and penalty costs for unutilized wind power; For time period Electricity sales price; For regular load indexing; For normal load During the period Electricity consumption; For steel mills during the period Equivalent electricity consumption; The operating cost of thermal power units is expressed as a quadratic function related to the active power output of the unit: (17) in, For thermal power units During the period Those who have made meritorious contributions; , and thermal power units The coefficients of the constant term, linear term, and quadratic term of the operating cost function; The penalty cost for unutilized wind power is determined by the difference between the predicted available wind power and the actual output: (18) in, For wind turbines The predicted available power; For wind turbines During the period The actual contribution of the individual; This is the penalty coefficient for unutilized wind power. The steel mill's electricity consumption at fine time scales is aggregated into equivalent electricity consumption during power system dispatch periods according to time mapping: (19) Among them, the power system dispatching period and the corresponding steel production dispatching period are associated through time scale mapping; the equivalent electricity consumption of steel plants is the aggregate value of electricity consumption of each steel production dispatching period within a power system dispatching interval; Power system constraints include power balance constraints, line transmission capacity constraints, output constraints of thermal power units and wind power units, thermal power unit ramping constraints, and system spinning reserve constraints. (20) in, To connect load nodes The steel mill during the period The equivalent active power; For normal load During the period The active power; (21) in, and These are the transmission line index and the bus index, respectively. For the set of busbars; For the line Relative to the busbar DC power transfer factor; For the line The transmission capacity; (22) in, and thermal power units The minimum and maximum active power output; (23) (24) in, This refers to the power system dispatch time interval; and thermal power units Downward and upward ramp rate limits; (25) in, and thermal power units During the period Provided rotational spare capacity up and down; (26) (27) in, and Time periods The system rotates up and down to meet backup requirements.

6. The steel-power dispatching collaborative optimization method based on an improved resource-task network model according to claim 5, characterized in that, Step (5) includes: using the sum of the steelmaking benefits of the steel plant and the revenue of the power system as the synergistic optimization objective. (28) in, The total revenue for the steel plant and the power system; The collaborative optimization constraints consist of the steel production scheduling constraints of the improved RTN model in step (3) and the optimal scheduling constraints of the power system in step (4). The electricity consumption of the steel plant is simultaneously included in the steelmaking energy consumption cost, power system power balance and line power flow constraints, so that the furnace selection, production tasks, electric arc furnace power, thermal power output and wind power output are linked in the same optimization problem.

7. The steel-power dispatching collaborative optimization method based on an improved resource-task network model according to claim 6, characterized in that, The step (6) includes: forming the above model into a mixed integer optimization problem and solving it, outputting the selected production furnace, the start and end times of processing and transfer tasks for each furnace, equipment occupancy status, steel plant time-series electricity consumption, thermal power unit output, wind power unit output, line power and spinning reserve plan, and sending the equipment-level task plan to the steel plant production scheduling system and the unit output plan to the power system scheduling system.

Images