Iron and steel industry load adjustable potential assessment method and system

Abstract

The invention discloses an iron and steel industry load adjustable potential assessment method and system, and relates to the technical field of power load management, and the method comprises the following steps: obtaining iron and steel industry related historical data, inputting the iron and steel industry related data into a pre-established resource task network RTN model, and outputting to obtain a minimum power cost data set; inputting the iron and steel industry related historical data and the minimum power cost data set into a pre-established FNN model for training to obtain a trained FNN model; and obtaining a steel industry load, inputting the steel industry load into the trained FNN model, and outputting to obtain an adjustable potential evaluation value.

Description

Technical Field

[0001] The invention relates to the technical field of power load management, and in particular to a method and system for evaluating the adjustable potential of load in the steel industry. Background Art

[0002] Globally, the steel industry is an energy-intensive industry that consumes a lot of electricity in the production process. Therefore, electricity cost has become a key factor affecting the economic benefits of steel companies. With the fluctuation of energy prices and the increase in environmental sustainability requirements, steel companies are in urgent need of improving their electricity efficiency and management strategies. Demand response (DR), as an emerging power system management strategy, encourages users to adjust their electricity consumption behavior according to electricity price signals or policy guidance to reduce electricity consumption during peak hours, balance the grid load, and improve overall energy efficiency.

[0003] Although demand response provides the steel industry with opportunities to reduce costs and improve energy efficiency, existing technologies face a series of challenges in practical application. First, many steel companies still use the traditional electricity management model, which is usually extensive and lacks the ability to deeply analyze and flexibly adjust the characteristics of electricity consumption. Secondly, the existing production scheduling models often only focus on production efficiency, but ignore the optimization potential of electricity costs and fail to fully utilize the advantages of demand response. In addition, steel companies often lack effective forecasting tools and methods when making load scheduling decisions, which limits their ability to make accurate scheduling decisions. Summary of the invention

[0004] In order to solve the deficiencies mentioned in the above background technology, the purpose of the present invention is to provide a method and system for evaluating the adjustable potential of load in the steel industry, which can adapt to changes in the electricity market, effectively reduce electricity costs, and improve the operating efficiency and stability of the power grid through refined modeling and intelligent scheduling.

[0005] In a first aspect, the purpose of the present invention can be achieved by the following technical solution: a method for evaluating the potential of load adjustment in the steel industry, the method comprising the following steps:

[0006] Obtain historical data related to the steel industry, input the data related to the steel industry into the pre-established resource task network RTN model, and output the minimized electricity cost data set;

[0007] Input relevant historical data of the steel industry and the data set of minimizing electricity cost into the pre-established FNN model for training to obtain a trained FNN model;

[0008] Obtain the steel industry load, input the steel industry load into the trained FNN model, and output the adjustable potential evaluation value.

[0009] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: the steel industry related historical data includes: power consumption characteristics and production constraints of electric arc furnaces, argon oxygen decarburization devices, ladle furnaces and continuous casting machines.

[0010] In combination with the first aspect, in some implementations of the first aspect, the method further includes: i,t Indicates the scheduling of task start, each task is indexed by i, and the binary variable N i,t Assign the start time of task i to time point t, N i,t =1 means task i starts in time period t, variable N i,t As the variables to be solved, t is used to index to a specific time of day, θ is a relative time index, and δ is used to represent the actual duration of an individual time slot.

[0011] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: the calculation process of inputting the steel industry related data into the pre-established resource task network RTN model:

[0012] The overall goal of optimization is to minimize the total energy cost, as follows:

[0013]

[0014] Among them, the time-of-use electricity price hr is a known quantity, input the objective function, and take T hr Indicates the set time slot in hours;

[0015] The formula for calculating power consumption is as follows:

[0016]

[0017] Among them, EL,t Equal to the power consumption of all possible tasks in time slot t; the total power consumption is obtained by adding up the power consumption of tasks in the whole time twice;

[0018] The constraints of the optimal scheduling scheme that minimizes electricity cost include: resource balancing constraints, task initiation constraints, task execution constraints, transfer time constraints, and product delivery constraints.

[0019] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the resource balancing constraint includes:

[0020]

[0021] where the value of resource r at time slot t is equal to its previous value at time t-1, adjusted by the amount produced / consumed by all related tasks, and the non-zero interaction parameter u r,i,θ This means that interactive task i interacts with resource r only when the task is in progress;

[0022] Task launch constraints:

[0023]

[0024] The above formula is the task initiation constraint. The task starts to meet two necessary conditions: the raw materials are available in the time period; the working equipment is available in the time period;

[0025] Task execution constraints:

[0026]

[0027] Task Execution: Use the following operational constraints to ensure that tasks are executed an appropriate number of times. Also, intermediate products should only be transferred once between each stage, defined as

[0028]

[0029] Transfer time constraints:

[0030]

[0031] The intermediate products generated by the transfer task after the task is executed need to be transferred immediately;

[0032] Assume that the transfer time of the intermediate products is w EA 、w AL 、w LC , regardless of the task production line and operation unit, the maximum allowable transportation time W EA ,W AL and W LC , the transmission time constraint is as follows:

[0033]

[0034] Wherein, δ is the time slot width. For intermediate products EA h The total time slot to wait before entering the next stage, the constraint states that for each intermediate product, the transfer time plus the waiting time should be an upper bound on the total transfer time;

[0035] Product delivery constraints:

[0036]

[0037] Where T represents the last time slot of the task cycle.

[0038] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: when the historical data related to the steel industry and the data set for minimizing electricity costs are input into the pre-established FNN model for training, evaluating the training effect by the mean square error and the coefficient of determination;

[0039] Mean square error MSE formula:

[0040]

[0041] In the formula, n is the number of samples, y is i is the actual value of the ith sample, is the predicted value of the i-th sample;

[0042] Coefficient of determination R 2 formula:

[0043]

[0044] In the formula, is the sample mean.

[0045] In combination with the first aspect, in some implementations of the first aspect, the method further includes: a calculation formula for the adjustable potential evaluation value:

[0046]

[0047] Where F is defined as the adjustable potential value of the steel industry load; To regulate the electricity consumption during the pre-peak period; It is the peak power consumption after dispatch; It is the electricity consumption in the off-peak period after dispatch; It is the electricity consumption during the off-peak period after dispatch.

[0048] In a second aspect, in order to achieve the above-mentioned object, the present invention discloses a steel industry load adjustable potential assessment system, comprising:

[0049] A data set generation module is used to obtain historical data related to the steel industry, input the data related to the steel industry into a pre-established resource task network RTN model, and output a data set for minimizing electricity costs;

[0050] A model training module is used to input the historical data related to the steel industry and the data set for minimizing electricity costs into the pre-established FNN model for training to obtain a trained FNN model;

[0051] The potential assessment module is used to obtain the steel industry load, input the steel industry load into the trained FNN model, and output an adjustable potential assessment value.

[0052] In conjunction with the second aspect, in certain implementations of the second aspect, the system further includes: the historical data related to the steel industry in the data set generation module includes: power consumption characteristics and production constraints of electric arc furnaces, argon oxygen decarburization devices, ladle furnaces, and continuous casting machines;

[0053] Or N in the RTN model in the dataset generation module i,t Indicates the scheduling of task start, each task is indexed by i, and the binary variable N i,t Assign the start time of task i to time point t, N i,t =1 means task i starts in time period t, variable N i,t As the variables to be solved, t is used to index to a specific time of the day, θ is a relative time index, and δ is used to represent the actual duration of an individual time slot;

[0054] Or the calculation process of inputting steel industry related data into the pre-established resource task network RTN model in the data set generation module:

[0055] The overall goal of optimization is to minimize the total energy cost, as follows:

[0056]

[0057] Among them, the time-of-use electricity price hr is a known quantity, input the objective function, and take T hr Indicates the set time slot in hours;

[0058] The formula for calculating power consumption is as follows:

[0059]

[0060] Among them, EL,t Equal to the power consumption of all possible tasks in time slot t; the total power consumption is obtained by adding up the power consumption of tasks in the whole time twice;

[0061] The constraints of the optimal scheduling scheme that minimizes electricity cost include: resource balancing constraints, task initiation constraints, task execution constraints, transfer time constraints, and product delivery constraints;

[0062] Or the resource balancing constraints within the dataset generation module include:

[0063]

[0064] where the value of resource r at time slot t is equal to its previous value at time t-1, adjusted by the amount produced / consumed by all related tasks, and the non-zero interaction parameter u r,i,θ This means that interactive task i interacts with resource r only when the task is in progress;

[0065] Task launch constraints:

[0066]

[0067] The above formula is the task initiation constraint. The task starts to meet two necessary conditions: the raw materials are available in the time period; the working equipment is available in the time period;

[0068] Task execution constraints:

[0069]

[0070] Task Execution: Use the following operational constraints to ensure that tasks are executed an appropriate number of times. Also, intermediate products should only be transferred once between each stage, defined as

[0071]

[0072] Transfer time constraints:

[0073]

[0074] The intermediate products generated by the transfer task after the task is executed need to be transferred immediately;

[0075] Assume that the transfer time of the intermediate products is w EA 、w AL 、w LC , regardless of the task production line and operation unit, the maximum allowable transportation time W EA ,W AL and W LC , the transmission time constraint is as follows:

[0076]

[0077] Wherein, δ is the time slot width. For intermediate products EA h The total time slot to wait before entering the next stage, the constraint states that for each intermediate product, the transfer time plus the waiting time should be an upper bound on the total transfer time;

[0078] Product delivery constraints:

[0079]

[0080] In the formula, T represents the last time slot of the task cycle;

[0081] Among them, when the historical data related to the steel industry and the minimum electricity cost data set are input into the pre-established FNN model for training in the model training module, the training effect is evaluated by the mean square error and determination coefficient;

[0082] Mean square error MSE formula:

[0083]

[0084] In the formula, n is the number of samples, y is i is the actual value of the ith sample, is the predicted value of the i-th sample;

[0085] Coefficient of determination R 2 formula:

[0086]

[0087] In the formula, is the sample mean.

[0088] The calculation formula of the adjustable potential evaluation value in the potential evaluation module is:

[0089]

[0090] Where F is defined as the adjustable potential value of the steel industry load; To regulate the electricity consumption during the pre-peak period; It is the peak power consumption after dispatch; It is the electricity consumption in the off-peak period after dispatch; It is the electricity consumption during the off-peak period after dispatch.

[0091] In another aspect of the present invention, in order to achieve the above-mentioned purpose, a terminal device is disclosed, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the memory stores a computer program capable of running on the processor, and when the processor loads and executes the computer program, a method for evaluating the potential for load adjustment in the steel industry as described above is adopted.

[0092] Beneficial effects of the present invention:

[0093] Through refined modeling and intelligent scheduling, the present invention can adapt to changes in the power market, effectively reduce power costs, and improve the operating efficiency and stability of the power grid. By combining advanced mathematical models and artificial intelligence technology, the present invention not only improves the intelligent level of load management in the steel industry, but also provides strong technical support for demand response in the power market. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0095] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0096] Figure 2 This is a schematic diagram of the steel load process flow for testing of the present invention;

[0097] Figure 3 A schematic diagram of a resource task network constructed for the present invention;

[0098] Figure 4 This is a task scheduling diagram under a certain province's time-of-use electricity price calculated by the present invention;

[0099] Figure 5 A schematic diagram of the usage status of each device in the RTN mode calculated by the present invention;

[0100] Figure 6 This is a schematic diagram of the FNN structure used in the present invention;

[0101] Figure 7 A schematic diagram showing the comparison of the training set and test set results calculated by the present invention;

[0102] Figure 8 This is a schematic diagram of sensitivity analysis of each link calculated by the present invention;

[0103] Fig. 9 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0104] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0105] Embodiment 1:

[0106] The following is an introduction to the relevant terms involved in the embodiments of the present application:

[0107] like Figure 1 As shown, a method for evaluating the potential of load adjustment in the steel industry comprises the following steps:

[0108] S101: Obtain historical data related to the steel industry, input the data related to the steel industry into a pre-established resource task network RTN model, and output a minimized electricity cost data set;

[0109] First, a resource task network model is constructed based on the production process of the steel industry. The model describes in detail the power consumption characteristics and production constraints of key equipment such as electric arc furnaces, argon oxygen decarburization devices, ladle furnaces and continuous casting machines. The main implementation steps are as follows:

[0110] Specifically, the present invention is further described below by way of embodiments:

[0111] The model describes the complete steel industry process flow model. The tasks considered are four main production stages, namely melting in an electric arc furnace, decarburization in an argon oxygen decarburization device, refining in a ladle furnace, casting in a continuous caster, and transfer tasks between stages. First, the raw materials are melted in an electric arc furnace to obtain products, which are then transported to the argon oxygen decarburization device, decarburized, and transported to the ladle furnace equipment. After the refining task, the products are transported to the continuous caster for the final casting task, and finally the finished product H is obtained. It should be noted that the process of performing tasks in the equipment consumes electrical energy (EL)

[0112] The resource-task network flow diagram shows how each task interacts with each resource. Since the factory consumes electricity all the time, the interaction between the process task and the power resource is continuous, but the interaction with other resources is discrete. Continuous interaction means that the task continuously consumes or generates resources during the processing. Based on the simplification of the model, this paper assumes that the melting task consumes electricity continuously during the entire task. Discrete interaction means that the interaction only occurs at very obvious time points during the task process, which means that the use of some resources only changes significantly at a specific stage of the task. For example, the melting task occupies an arc furnace at the beginning and releases an arc furnace only at the end of the melting process. The melting task may last for several time slots, but it only interacts with the arc furnace resource in two time slots.

[0113] The settings of the relevant variables are outlined below.

[0114] N i,t Indicates the scheduling of task start. Each task is indexed by i, and the binary variable N i,t Assign the start time of task i to time point t, that is, N i,t =1 means task i starts in time period t, variable N i,t As a variable to be solved, it is one of the solution objectives and needs to be obtained through the optimization solver to solve the optimal scheduling problem under the corresponding objective function.

[0115] The types of resources included in the model are equipment units (EAF, AOD, LF and CC), electric energy (EL), intermediate products (EA s ,EA d ,AL s ,ALd ,LC s , and LC d ) and the final product (H). In the steel production process, as a continuous industrial process, the transfer task must be executed immediately after the previous processing task is completed, in order to avoid affecting the solution of the optimal scheduling situation due to irrelevant waiting time. Any delay may lead to increased costs or reduced product quality.

[0116] Non-negative continuous variable R r,t Represents the availability of resource r. For example, R EAF,t =1 means that there is an EAF device available at time t, R EAF,t =0 means that the current arc furnace equipment is in working order and the previous products need to wait.

[0117] Resource interaction parameter u r,i,θ is a core parameter that quantifies how much resource r is consumed or generated in a time slot θ in task i (θ refers to the θth time slot after the task starts). Parameter u is a known quantity and can be initially set by the time and power consumption of each stage of the actual industrial process to assist in subsequent scheduling solutions.

[0118] Variable Π EL,t Used to summarize the power usage of all tasks in time period t.

[0119] In terms of time, there are three types of time bases. The first is the actual time of 24 hours a day, which is used to couple with actual industrial scheduling; t is used to index to a specified time of the day; the relative time index θ is used for the time index of independent tasks, which is related to the task start time and discretizes time into time slots for more convenient representation of the interaction between tasks and resources. δ is used to represent the actual duration of a single time slot. For example, if δ is set to 10 minutes, there are 144 time slots in a day, and the discretization of time is completed in this way.

[0120] The calculation process of inputting relevant data of the steel industry into the pre-established Resource Task Network RTN model:

[0121] The overall goal of optimization is to minimize the total energy cost, as follows:

[0122]

[0123] Among them, the time-of-use electricity price hr is a known quantity, input the objective function, and take T hr Indicates the set time slot in hours;

[0124] The formula for calculating power consumption is as follows:

[0125]

[0126] Among them, EL,t Equal to the power consumption of all possible tasks in time slot t; the total power consumption is obtained by adding up the power consumption of tasks in the whole time twice;

[0127] The constraints of the optimal scheduling scheme that minimizes electricity cost include: resource balancing constraints, task initiation constraints, task execution constraints, transfer time constraints, and product delivery constraints.

[0128] The constraints of the optimal dispatching scheme to minimize electricity cost are as follows:

[0129] Resource balancing constraints:

[0130]

[0131] Here, the value of resource r at time slot t is equal to its previous value at time t-1, adjusted by the amount produced / consumed by all related tasks. The above constraints apply to all resources described in this paper, except power resources.

[0132] Non-zero interaction parameter u r,i,θ This means that interactive task i interacts with resource r only when the task is in progress. That is, only when task i starts time slot θ earlier than t(N i,t-θ =1) and θ is less than τ i ——The duration of task i, the interaction action will occur in time slot t.

[0133] Task launch constraints:

[0134]

[0135] The above formula is the task initiation constraint. The task starts only when two necessary conditions are met: (1) raw materials are available during this period; (2) working equipment is available during this period.

[0136] Task execution constraints:

[0137]

[0138] Task Execution: Use the following operational constraints to ensure that tasks are executed an appropriate number of times. Also, intermediate products should only be transferred once between each stage, defined as

[0139]

[0140] Transfer time constraints:

[0141]

[0142] Transfer tasks are tasks where the intermediate products generated after the task is executed need to be transferred immediately. This is common in the steel manufacturing industry and is achieved by enforcing constraints. Either 0 or 1. If it is equal to 1, it means the intermediate product Waiting in time slot t, the above constraint requires that any intermediate product with superscript s has no waiting time.

[0143] Assume that the transfer time of the intermediate products is w EA 、w AL 、w LC , regardless of the task production line and operation unit. The maximum allowable transportation time W EA ,W AL and W LC According to this similar definition, since the quality of the product may be affected by the low temperature caused by the process idleness, and the quality must be compensated by secondary heating which consumes more energy, it is ensured that the cooling effect during transportation will not have an adverse effect on the processing of the intermediate product in the next stage. Therefore, the transfer time constraint is as follows:

[0144]

[0145] Wherein, δ is the time slot width. For intermediate products EA h The total time slot to wait before entering the next stage. The constraint states that for each intermediate product, the transfer time plus the waiting time should be an upper bound on the total transfer time to prevent adverse cooling effects.

[0146] Product delivery constraints:

[0147]

[0148] Where T represents the last time slot of the task cycle. The final product should be delivered before the end of the deadline. Since the objective function is to minimize the electricity cost, the solution is to avoid factory shutdown.

[0149] S102: inputting relevant historical data of the steel industry and the minimum electricity cost data set into a pre-established FNN model for training to obtain a trained FNN model;

[0150] In order to evaluate the adjustable potential of the steel industry load, it is necessary to first construct a sample set. The sample set comes from the resource task network model of the steel industry load, and the characteristic value is determined by the parameters and solution values ​​in the model. The working time and operating power of four types of equipment, including electric arc furnace, argon oxygen decarburization device, ladle furnace and continuous casting machine, are used as input characteristic values, and the output value is to minimize the dispatching power cost and the adjustable potential value. With 50% to 150% of the rated value of each input parameter as the limit, the parameters are randomly obtained and input into the model to solve the output value. The model is run in a loop to obtain 824 sets of data as sample data sets for subsequent neural network training, in which the training set and test set are divided into 2:8.

[0151] A multi-layer feedforward neural network (FNN) is chosen, which has significant advantages. FNN has a strong nonlinear mapping capability, which can effectively capture the complex nonlinear relationship between input features and output results. During the operation of the load scheduling model for the steel industry based on RTN established in this paper, there are highly nonlinear interactions between factors such as equipment working time, power consumption, and task scheduling. Traditional linear models are difficult to accurately describe this complex relationship. FNN can fit these complex nonlinear relationships well through multi-layer neurons and nonlinear activation functions, thereby improving the prediction accuracy of the model.

[0152] At the same time, in order to prevent overfitting during model training, this paper introduces two widely used techniques: regularization term and early stopping method.

[0153] Regularization prevents overfitting by adding a penalty term to the loss function to limit the complexity of the model. Early stopping is another important technique to prevent overfitting. It prevents the model from overfitting on the training set by stopping the training early when the validation error starts to increase.

[0154] In order to evaluate the training results, the mean square error (MSE) and the coefficient of determination (R 2 ).

[0155] The mean square error quantifies the prediction accuracy of the model by calculating the average of the squares and the average of the prediction errors. The smaller the MSE value, the smaller the gap between the predicted value and the actual value, and the better the prediction performance of the model. i is the actual value of the ith sample, is the predicted value of the i-th sample, then the formula for the mean square error is as follows:

[0156]

[0157] Coefficient of Determination (R 2 ) is an indicator to measure the goodness of fit of the model to the data. It reflects the explanatory power of the model, that is, the proportion of variance that the model can explain. 2 The value of is between 0 and 1. The closer the value is to 1, the stronger the explanatory power of the model. The formula is as follows:

[0158]

[0159]

[0160] In the formula, is the sample mean.

[0161] S103: Obtain the steel industry load, input the steel industry load into the trained FNN model, and output an adjustable potential evaluation value.

[0162] In order to explore the adjustable potential of the steel industry load and achieve the purpose of peak shaving and valley filling to optimize the operation of the power system, the relative adjustable potential value is set. The peak and valley time periods are set as shown in Table 3.1. The peak period is 8:00-11:00 in the morning and 17:00-22:00 in the afternoon, and the valley period is 0:00-8:00. The pre-dispatch mode is based on the Basic mode, and the steel industry production tasks are carried out from 8:00. The formula for the dispatchable potential value is as follows:

[0163]

[0164] Where F is defined as the adjustable potential value of the steel industry load; To regulate the electricity consumption during the pre-peak period; It is the peak power consumption after dispatch; It is the electricity consumption in the off-peak period after dispatch; It is the electricity consumption during the off-peak period after dispatch.

[0165] The training results are analyzed. Regarding the potential evaluation value, the overall trend is relatively consistent in the training set, indicating that the model can better capture the trend of potential evaluation values. In the test set, the deviation between the predicted value and the actual value is small, and the model prediction effect is good, especially in the high potential evaluation value range, the model prediction accuracy is high. Considering the mean square error and determination coefficient, they all meet the characteristics of a strong prediction ability model, and the model has good performance.

[0166] In order to explore which link in the steel industry process has greater adjustable potential space, this paper introduces sensitivity analysis. The sensitivity analysis of this paper aims to study the impact of changes in various parameters in the steel production process on the potential evaluation value, by making small quantitative adjustments to the model input characteristic values, calculating the change in the output value, and finally solving the sensitivity of each link:

[0167]

[0168] Based on a multi-layer feedforward neural network, an adjustable potential evaluation model for the steel industry load was established, and model evaluation parameters such as mean square error and determination coefficient were introduced. After training, the model has good prediction ability and can solve the minimum electricity cost and potential evaluation value by adjusting the parameters of the steel production process, which is helpful to achieve the demand response goal. At the same time, through sensitivity analysis, it is found that the production links that have a greater impact on the adjustable potential evaluation value provide a basis for the scheduling and operation of the steel industry.

[0169] Embodiment 2: In the second aspect, as Fig. 9 As shown, in order to achieve the above-mentioned purpose, the present invention discloses a steel industry load adjustable potential assessment system, comprising:

[0170] The data set generation module 11 is used to obtain historical data related to the steel industry, input the data related to the steel industry into a pre-established resource task network RTN model, and output a data set for minimizing electricity costs;

[0171] The model training module 12 is used to input the historical data related to the steel industry and the data set for minimizing the electricity cost into the pre-established FNN model for training to obtain a trained FNN model;

[0172] The potential assessment module 13 is used to obtain the steel industry load, input the steel industry load into the trained FNN model, and output an adjustable potential assessment value.

[0173] In conjunction with the second aspect, in certain implementations of the second aspect, the system further includes: the historical data related to the steel industry in the data set generation module 11 includes: power consumption characteristics and production constraints of electric arc furnaces, argon oxygen decarburization devices, ladle furnaces and continuous casting machines;

[0174] Or N in the RTN model in the data set generation module 11 i,t Indicates the scheduling of task start, each task is indexed by i, and the binary variable N i,t Assign the start time of task i to time point t, N i,t =1 means task i starts in time period t, variable N i,t As the variables to be solved, t is used to index to a specific time of the day, θ is a relative time index, and δ is used to represent the actual duration of an individual time slot;

[0175] Or the calculation process of inputting the steel industry related data into the pre-established resource task network RTN model in the data set generation module 11:

[0176] The overall goal of optimization is to minimize the total energy cost, as follows:

[0177]

[0178] Among them, the time-of-use electricity price hr is a known quantity, input the objective function, and take T hr Indicates the set time slot in hours;

[0179] The formula for calculating power consumption is as follows:

[0180]

[0181] Among them, EL,t Equal to the power consumption of all possible tasks in time slot t; the total power consumption is obtained by adding up the power consumption of tasks in the whole time twice;

[0182] The constraints of the optimal scheduling scheme that minimizes electricity cost include: resource balancing constraints, task initiation constraints, task execution constraints, transfer time constraints, and product delivery constraints;

[0183] Or the resource balancing constraints in the data set generation module 11 include:

[0184]

[0185] where the value of resource r at time slot t is equal to its previous value at time t-1, adjusted by the amount produced / consumed by all related tasks, and the non-zero interaction parameter u r,i,θ This means that interactive task i interacts with resource r only when the task is in progress;

[0186] Task launch constraints:

[0187]

[0188] The above formula is the task initiation constraint. The task starts to meet two necessary conditions: the raw materials are available in the time period; the working equipment is available in the time period;

[0189] Task execution constraints:

[0190]

[0191] Task Execution: Use the following operational constraints to ensure that tasks are executed an appropriate number of times. Also, intermediate products should only be transferred once between each stage, defined as

[0192]

[0193] Transfer time constraints:

[0194]

[0195] The intermediate products generated by the transfer task after the task is executed need to be transferred immediately;

[0196] Assume that the transfer time of the intermediate products is w EA 、w AL 、w LC , regardless of the task production line and operation unit, the maximum allowable transportation time W EA ,W AL and W LC , the transmission time constraint is as follows:

[0197]

[0198] Wherein, δ is the time slot width. For intermediate products EA h The total time slot to wait before entering the next stage, the constraint states that for each intermediate product, the transfer time plus the waiting time should be an upper bound on the total transfer time;

[0199] Product delivery constraints:

[0200]

[0201] In the formula, T represents the last time slot of the task cycle;

[0202] Among them, when the historical data related to the steel industry and the minimum electricity cost data set are input into the pre-established FNN model for training in the model training module 12, the training effect is evaluated by the mean square error and the determination coefficient;

[0203] Mean square error MSE formula:

[0204]

[0205] In the formula, n is the number of samples, y is i is the actual value of the ith sample, is the predicted value of the i-th sample;

[0206] Coefficient of determination R 2 formula:

[0207]

[0208] In the formula, is the sample mean.

[0209] The calculation formula of the adjustable potential evaluation value in the potential evaluation module 13 is:

[0210]

[0211] Where F is defined as the adjustable potential value of the steel industry load; To regulate the electricity consumption during the pre-peak period; It is the peak power consumption after dispatch; It is the electricity consumption in the off-peak period after dispatch; It is the electricity consumption during the off-peak period after dispatch.

[0212] Based on the same inventive concept, the present invention also provides a computer device, which includes: one or more processors, and a memory for storing one or more computer programs; the program includes program instructions, and the processor is used to execute the program instructions stored in the memory. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is used to implement one or more instructions, specifically used to load and execute one or more instructions in a computer storage medium to implement the above method.

[0213] It needs to be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium, on which a computer program is stored, and the computer program is executed by a processor to execute the above method. The storage medium can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electrical, magnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0214] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0215] The above shows and describes the basic principles, main features and advantages of the present disclosure. Those skilled in the art should understand that the present disclosure is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present disclosure. Without departing from the spirit and scope of the present disclosure, the present disclosure may have various changes and improvements, and these changes and improvements fall within the scope of the present disclosure to be protected.

Claims

1. A method for evaluating the potential of load adjustment in the steel industry, characterized in that: The method comprises the following steps: Obtain historical data related to the steel industry, input the data related to the steel industry into the pre-established resource task network RTN model, and output the minimized electricity cost data set; Input relevant historical data of the steel industry and the data set of minimizing electricity cost into the pre-established FNN model for training to obtain a trained FNN model; Obtain the steel industry load, input the steel industry load into the trained FNN model, and output the adjustable potential evaluation value.

2. A method for evaluating the potential for load adjustment in the steel industry according to claim 1, characterized in that: The steel industry related historical data includes: power consumption characteristics and production constraints of electric arc furnaces, argon oxygen decarburization devices, ladle furnaces and continuous casters.

3. The method for evaluating the potential of load adjustment in the steel industry according to claim 1, characterized in that: The RTN model contains N i,t Indicates the scheduling of task start, each task is indexed by i, and the binary variable N i,t Assign the start time of task i to time point t, N i,t =1 means task i starts in time period t, variable N i,t As the variables to be solved, t is used to index to a specific time of day, θ is a relative time index, and δ is used to represent the actual duration of an individual time slot.

4. The method for evaluating the potential of load adjustment in the steel industry according to claim 1, characterized in that: The calculation process of inputting the steel industry related data into the pre-established resource task network RTN model: The overall goal of optimization is to minimize the total energy cost, as follows: Among them, the time-of-use electricity price hr is a known quantity, input the objective function, and take T hr Indicates the set time slot in hours; The formula for calculating power consumption is as follows: Among them, EL,t Equal to the power consumption of all possible tasks in time slot t; the total power consumption is obtained by adding up the power consumption of tasks in the whole time twice; The constraints of the optimal scheduling scheme that minimizes electricity cost include: resource balancing constraints, task initiation constraints, task execution constraints, transfer time constraints, and product delivery constraints.

5. A method for evaluating the potential for load adjustment in the steel industry according to claim 4, characterized in that: The resource balancing constraints include: where the value of resource r at time slot t is equal to its previous value at time t-1, adjusted by the amount produced / consumed by all related tasks, and the non-zero interaction parameter u r,i,θ This means that interactive task i interacts with resource r only when the task is in progress; Task launch constraints: The above formula is the task initiation constraint. The task starts to meet two necessary conditions: the raw materials are available in the time period; the working equipment is available in the time period; Task execution constraints: Task Execution: Use the following operational constraints to ensure that tasks are executed an appropriate number of times. Also, intermediate products should only be transferred once between each stage, defined as Transfer time constraints: The intermediate products generated by the transfer task after the task is executed need to be transferred immediately; Assume that the transfer time of the intermediate products is w EA 、w AL 、w LC , regardless of the task production line and operation unit, the maximum allowable transportation time W EA ,W AL and W LC , the transmission time constraint is as follows: Where δ is the time slot width, For intermediate products EA h The total time slot to wait before entering the next stage, the constraint states that for each intermediate product, the transfer time plus the waiting time should be an upper bound on the total transfer time; Product delivery constraints: Where T represents the last time slot of the task cycle.

6. A method for evaluating the potential for load adjustment in the steel industry according to claim 1, characterized in that: When the steel industry related historical data and the minimum electricity cost data set are input into the pre-established FNN model for training, the training effect is evaluated by the mean square error and the determination coefficient; Mean square error MSE formula: In the formula, n is the number of samples, y is i is the actual value of the ith sample, is the predicted value of the i-th sample; Coefficient of determination R 2 formula: In the formula, is the sample mean.

7. A method for evaluating the potential of load adjustment in the steel industry according to claim 1, characterized in that: The calculation formula of the adjustable potential evaluation value is: Where F is defined as the adjustable potential value of the steel industry load; To regulate the pre-peak electricity consumption; It is the peak power consumption after dispatch; It is the electricity consumption in the off-peak period after dispatch; It is the electricity consumption during the off-peak period after dispatch.

8. A steel industry load adjustable potential assessment system, characterized in that: include: A data set generation module is used to obtain historical data related to the steel industry, input the data related to the steel industry into a pre-established resource task network RTN model, and output a data set for minimizing electricity costs; A model training module is used to input the historical data related to the steel industry and the data set for minimizing electricity costs into the pre-established FNN model for training to obtain a trained FNN model; The potential assessment module is used to obtain the steel industry load, input the steel industry load into the trained FNN model, and output an adjustable potential assessment value.

9. The steel industry load adjustable potential assessment system according to claim 8, characterized in that: The historical data related to the steel industry in the data set generation module include: power consumption characteristics and production constraints of electric arc furnaces, argon oxygen decarburization devices, ladle furnaces and continuous casting machines; Or N in the RTN model in the dataset generation module i,t Indicates the scheduling of task start, each task is indexed by i, and the binary variable N i,t Assign the start time of task i to time point t, N i,t =1 means task i starts in time period t, variable N i,t As the variables to be solved, t is used to index to a specific time of the day, θ is a relative time index, and δ is used to represent the actual duration of an individual time slot; Or the calculation process of inputting steel industry related data into the pre-established resource task network RTN model in the data set generation module: The overall goal of optimization is to minimize the total energy cost, as follows: Among them, the time-of-use electricity price hr is a known quantity, input the objective function, and take T hr Indicates the set time slot in hours; The formula for calculating power consumption is as follows: Among them, EL,t Equal to the power consumption of all possible tasks in time slot t; the total power consumption is obtained by adding up the power consumption of tasks in the whole time twice; The constraints of the optimal scheduling scheme that minimizes electricity cost include: resource balancing constraints, task initiation constraints, task execution constraints, transfer time constraints, and product delivery constraints; Or the resource balancing constraints within the dataset generation module include: where the value of resource r at time slot t is equal to its previous value at time t-1, adjusted by the amount produced / consumed by all related tasks, and the non-zero interaction parameter u r,i,θ This means that interactive task i interacts with resource r only when the task is in progress; Task launch constraints: The above formula is the task initiation constraint. The task starts to meet two necessary conditions: the raw materials are available in the time period; the working equipment is available in the time period; Task execution constraints: Task Execution: Use the following operational constraints to ensure that tasks are executed an appropriate number of times. Also, intermediate products should only be transferred once between each stage, defined as Transfer time constraints: The intermediate products generated by the transfer task after the task is executed need to be transferred immediately; Assume that the transfer time of the intermediate products is w EA 、w AL 、w LC , regardless of the task production line and operation unit, the maximum allowable transportation time W EA ,W AL and W LC , the transmission time constraint is as follows: Where δ is the time slot width, For intermediate products EA h The total time slot to wait before entering the next stage, the constraint states that for each intermediate product, the transfer time plus the waiting time should be an upper bound on the total transfer time; Product delivery constraints: In the formula, T represents the last time slot of the task cycle; Among them, when the historical data related to the steel industry and the minimum electricity cost data set are input into the pre-established FNN model for training in the model training module, the training effect is evaluated by the mean square error and determination coefficient; Mean square error MSE formula: In the formula, n is the number of samples, y is i is the actual value of the ith sample, is the predicted value of the i-th sample; Coefficient of determination R 2 formula: In the formula, is the mean of the sample; The calculation formula of the adjustable potential evaluation value in the potential evaluation module is: Where F is defined as the adjustable potential value of the steel industry load; To regulate the pre-peak electricity consumption; It is the peak power consumption after dispatch; It is the electricity consumption in the off-peak period after dispatch; It is the electricity consumption during the off-peak period after dispatch.

10. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: The memory stores a computer program that can be run on the processor. When the processor loads and executes the computer program, a method for assessing the potential for load adjustment in the steel industry according to any one of claims 1 to 7 is adopted.

Images