CN113962050A - Oxygen scheduling calculation method combining production consumption prediction and pipe network calculation - Google Patents
Oxygen scheduling calculation method combining production consumption prediction and pipe network calculation Download PDFInfo
- Publication number
- CN113962050A CN113962050A CN202111088740.1A CN202111088740A CN113962050A CN 113962050 A CN113962050 A CN 113962050A CN 202111088740 A CN202111088740 A CN 202111088740A CN 113962050 A CN113962050 A CN 113962050A
- Authority
- CN
- China
- Prior art keywords
- oxygen
- consumption
- converter
- prediction
- production
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/18—Network design, e.g. design based on topological or interconnect aspects of utility systems, piping, heating ventilation air conditioning [HVAC] or cabling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/02—CAD in a network environment, e.g. collaborative CAD or distributed simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/14—Pipes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- General Factory Administration (AREA)
Abstract
The invention relates to an oxygen scheduling calculation method combining production and consumption prediction and pipe network calculation, and belongs to the technical field of industrial information. The calculation method comprises the following steps: s1: constructing an oxygen production and consumption prediction model, and predicting the oxygen consumption of a blast furnace, a converter and other users in a future period of time; correcting the prediction result of the oxygen production and consumption prediction model according to the actual production condition of the converter; s2: constructing an oxygen pipe network calculation model, inputting the corrected prediction result as a boundary condition into the oxygen pipe network calculation model, and calculating to obtain the condition of the pressure change of the oxygen pipe network in a period of time in the future; s3: oxygen scheduling: and scheduling the oxygen system according to the corrected prediction result and the pressure change condition of the oxygen network in a period of time in the future. According to the invention, oxygen production and consumption prediction and oxygen pipe network calculation are introduced on the basis of the traditional oxygen scheduling algorithm, and are organically combined, so that the oxygen scheduling has more sufficient data basis.
Description
Technical Field
The invention belongs to the technical field of industrial information, and relates to an oxygen scheduling calculation method combining production consumption prediction and pipe network calculation.
Background
Energy conservation and emission reduction are always one of the production targets of key attention of iron and steel enterprises. Oxygen, as one of the important energy media of iron and steel enterprises, has a great influence on the production of blast furnaces and converters. Oxygen in iron and steel enterprises is mainly prepared by an oxygen generator set and is sent to oxygen users through an oxygen pipe network. The main users of oxygen in iron and steel enterprises are blast furnaces in the iron making process and converters in the steel making process: the oxygen consumption in the blast furnace process is mainly used for oxygen-enriched iron making, the blast furnace process is stable in production, and the oxygen consumption is relatively stable, so that the blast furnace oxygen consumption prediction is the prediction of stable data, and the prediction is relatively easy. The consumption of oxygen in the converter process is mainly used for converter blowing, and the production rhythm of the converter is carried out according to the production schedule, so that the converter has the characteristics of discontinuous discontinuity, complex influence factors and the like. The blowing process in converter steelmaking is the main link of oxygen consumption of a steel mill, and the oxygen consumption of the converter accounts for more than 50% of the total amount of the oxygen gas in the gang height. Converter steelmaking is a genuine big consumer of oxygen consumption, and the oxygen supply mode has a great influence on the stability of the whole oxygen system.
Currently, a common oxygen consumption prediction is generally completed by a data-driven method, and according to historical data, a nuclear learning model represented by a neural network and a support vector machine, a prediction model based on a fuzzy system and the like are adopted to perform a prediction algorithm of an iterative mechanism (bus E, road w.iterative-decoding search with on-line tree size prediction [ J ]. Annals of Mathematics and architecture analysis, 2013, 69 (2): 183-. (Zhang L, Zhou W D, Chang P C, et a1. iterative time series prediction with multiple support vector regression models [ J ]. NeurocomPUting, 2013, 99: 411 and 422.) or taking the stage production characteristics into consideration, dividing the data into particles with unequal lengths, then carrying out fuzzy clustering, fuzzy reasoning and the like by taking the data segment as a basic analysis unit, and finally finishing the prediction of the oxygen load. (Korean. steelmaking process gas energy system prediction and scheduling method and application [ D ]. university of Connectorized 2016.). The prediction of the oxygen consumption of a single furnace is finished by adopting an SVM and other established models, but the prediction result exists in the form of a total amount point, cannot be predicted on the time granularity, and cannot meet the requirement of field actual production. (Jiangweig iron and steel enterprise oxygen system prediction and optimization scheduling model research [ D ]. Tianjin university of technology, 2017.)
However, these methods are only by analyzing historical oxygen production and consumption data and then using algorithms to make predictions. However, when a sudden change of the production plan occurs, a corresponding prediction cannot be made only by the history data. The change rule of oxygen consumption is closely related to the schedule of the production plan, and by extracting the schedule of the production plan and combining historical data, the change trend of prediction can be mechanically explained, and the accuracy is better. For example, patent publication No. CN111353656A, "a method for predicting oxygen load of iron and steel enterprises based on production planning", solves the above factors affecting the prediction of oxygen production and consumption data, but does not make real-time correction on the prediction model (because oxygen consumption is dynamically changed in real time during oxygen consumption), and does not provide a scheduling method for oxygen. And the traditional oxygen scheduling method does not consider oxygen production and consumption prediction and oxygen official network calculation. Therefore, it is necessary to design an oxygen scheduling calculation method combining production consumption prediction and pipe network calculation.
Disclosure of Invention
In view of the above, the present invention provides an oxygen scheduling calculation method combining production consumption prediction and pipe network calculation, which solves the problem that the conventional oxygen scheduling method based on production consumption balance only is inaccurate in oxygen scheduling.
In order to achieve the purpose, the invention provides the following technical scheme:
an oxygen scheduling calculation method combining production and consumption prediction and pipe network calculation specifically comprises the following steps:
s1: constructing an oxygen production and consumption prediction model, and predicting the oxygen consumption of a blast furnace, a converter and other users in a future period of time; correcting the prediction result of the oxygen production and consumption prediction model according to the actual production condition of the converter;
s2: constructing an oxygen pipe network calculation model, inputting the prediction result corrected in the step S1 as a boundary condition into the oxygen pipe network calculation model, and calculating to obtain the condition of oxygen pipe network pressure change in a period of time in the future;
s3: oxygen scheduling: and performing optimized scheduling on the oxygen system according to the corrected prediction result obtained in the step S1 and the oxygen network pressure change situation in the future period of time calculated in the step S2.
Further, in step S1, the constructed oxygen consumption prediction model includes prediction of the oxygen-rich flow of the blast furnace, and the prediction is performed by using a moving average autoregressive method or a BP neural network.
Furthermore, the prediction is performed by adopting a moving average autoregressive method, specifically, the number of regression terms, the number of moving average terms and the number of difference terms are determined by training a model through historical data.
Furthermore, the BP neural network is adopted for prediction, and specifically, a model is trained through historical data to determine weights inside the neural network.
Further, in step S1, the constructed oxygen production and consumption prediction model further includes analysis and statistics of oxygen consumption of the converter, and an oxygen consumption calculation formula is fitted by finding a relationship between production process data of the converter and oxygen consumption of a corresponding furnace; the method specifically comprises the following steps: the method comprises the steps of firstly collecting a large amount of historical data of the production process of the converter, then determining the correlation between parameters such as the amount of molten iron, the amount of scrap steel and the composition of molten iron and the oxygen consumption through data analysis, secondly selecting the parameter which has the largest influence on the oxygen consumption according to the correlation, and finally fitting a calculation formula between the selected process parameters and the oxygen consumption through a formula fitting method.
Further, in step S1, the constructed oxygen generation and consumption prediction model further includes converter beat prediction; the method comprises the steps of introducing converter historical data to carry out beat prediction in converter beat prediction and carrying out real-time correction on converter beat and consumption according to actual production conditions; the method specifically comprises the following steps: firstly, a converter beat prediction model predicts smelting time of each furnace and interval time between the furnaces according to a converter production schedule, and then calculates oxygen consumption of each furnace according to molten iron data in a test system and a weighing system; when the data in the converter production scheduling, the testing system or the weighing system is missing or not updated timely, the system completes the data required in the rhythm prediction by reading the historical data of the converter production and in a weighted moving average mode; secondly, the converter beat prediction model compares whether the oxygen consumption prediction data and the actual oxygen consumption data of the converter production exist deviation in real time, if so, the converter oxygen consumption beat and the consumption are corrected to ensure that the prediction value is consistent with the actual production all the time.
Furthermore, when the predicted oxygen consumption data and the actual production consumption data of the converter are deviated, the correction of the oxygen consumption beat and consumption of the converter specifically comprises the following steps: judging and correcting the prediction condition, firstly judging whether the prediction condition is different at the beginning of a period or different at the end of the period;
the first case: if the difference occurs when a period starts, the actual period starts to occur first, and the predicted period starts to occur later, namely the actual blowing starts in advance, at the moment, not only the predicted value needs to be adjusted to be aligned with the actual value, but also the calculated value of Gap _ t needs to be updated by combining the scheduling data or the historical production data of the converter due to the occurrence of new Gap _ t; wherein Gap _ t represents the time interval between two smelting heats;
the second case: if the difference occurs at the beginning of a cycle, the predicted cycle occurs first, and the actual cycle occurs later, i.e., the predicted blow starts earlier, at which time the time for measurement is earlier, then Q _ t and Q _ O are measured2Reassigning, and pushing the prediction time backwards; wherein Q _ t represents the oxygen blowing duration of a single heat, Q _ O2Represents the total oxygen consumption of a single heat;
in a third case, if at the end of a cycle, the actual cycle ends first and the predicted cycle does not end, i.e., the actual blow ends earlier, this is the caseNot only does it need to adjust the predicted value to align with the actual value, but also new Q _ t and Q _ O appear2Therefore, it is necessary to combine the scheduling data or the converter production history data with the Q _ t and Q _ O pairs2Updating the calculated value;
in the fourth case, if the predicted period ends first and the actual period does not end yet, i.e., the predicted blowing ends earlier, at the end of a period, Q _ t and Q _ O are needed to be adjusted2And re-assigning values and pushing the cycle end time backwards.
Further, in step S2, the constructed oxygen pipe network calculation model includes two parts: one part is the hydraulic calculation of the pipeline, and the other part is the pressure calculation of the oxygen spherical tank.
Furthermore, the hydraulic calculation of the pipeline is solved by adopting a node method, the hydraulic calculation of the branch pipe network and the ring pipe network is solved, and the solving process comprises 3 equations: a) a node flow continuous equation; b) a pipe section pressure drop equation; c) a pipe section flow equation;
and calculating the pressure of the oxygen spherical tank, and obtaining the pressure change of the oxygen spherical tank caused by the unbalance amount of the oxygen production consumption by using a gas state equation.
Further, in step S3, the oxygen scheduling specifically includes: according to the corrected prediction result and the result calculated by the oxygen pipe network calculation model, whether the oxygen pipe network has larger pressure fluctuation in a future period of time is judged, if the pressure is suddenly increased, oxygen is diffused to cause energy use to be uneconomical, and if the pressure is suddenly reduced, the user requirements cannot be met to influence the process production; through judgment, if the unbalance amount of oxygen production consumption can cause larger pressure fluctuation of a pipe network in a period of time in the future, the oxygen dispatching system needs to intervene in advance, firstly, the load regulation of the oxygen generator set is used for meeting the requirement change of an oxygen user, but when the load regulation of the oxygen generator set is not enough for meeting the regulation requirement, the oxygen dispatching system can also give an adjustment suggestion of the oxygen blowing beat of the converter to the steelmaking process, so that the severe pressure fluctuation of the oxygen pipe network caused by simultaneous blowing or simultaneous non-blowing of multiple converters for a long time is avoided.
The invention has the beneficial effects that: the invention introduces oxygen production and consumption prediction and oxygen pipe network calculation on the basis of the traditional oxygen scheduling algorithm and organically combines the oxygen production and consumption prediction and the oxygen pipe network calculation. An oxygen production and consumption prediction function is introduced, and the amount of unbalance of oxygen production and consumption in a future period of time can be predicted, so that a scheduling system is intervened in advance, and the purpose of better stabilizing the pressure of an oxygen system is achieved; furthermore, aiming at the particularity of oxygen blowing of the converter, the accuracy of oxygen consumption prediction is improved by using a prediction method combining oxygen prediction and process parameters, and a method for correcting the oxygen prediction according to the actual production condition of the converter is provided, so that the condition of accumulated prediction errors is effectively avoided. And finally, an oxygen pipe network calculation function is introduced, so that the pressure and flow fluctuation of the pipe network and the oxygen storage tank caused by unbalanced oxygen production and consumption can be calculated more clearly, and the influence of the overlarge pressure or flow fluctuation on the normal production of oxygen users is avoided, so that the oxygen scheduling has more sufficient data basis.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
FIG. 1 is a general diagram of an oxygen optimization scheduling calculation process according to the present invention;
FIG. 2 is a flow chart of oxygen production and consumption prediction;
FIG. 3 is a flow chart of the converter oxygen consumption analysis statistics;
FIG. 4 is a flowchart of converter beat prediction;
FIG. 5 is a schematic diagram of predicted oxygen consumption data and parameters for a converter;
FIG. 6 is a schematic diagram of the real-time correction of oxygen beat and consumption of a converter;
FIG. 7 is a flow chart of an oxygen pipe network calculation model;
fig. 8 is a flow chart of oxygen schedule calculation.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Referring to fig. 1 to 8, before the oxygen optimization scheduling model is used, the oxygen consumption of the blast furnace, the converter and other users in a future period of time needs to be predicted, and the stable prediction of the oxygen consumption data of the blast furnace and other users is accurate and reliable in the prediction process, but the production rhythm of the converter is relatively complex, so that the production data related to the oxygen consumption of the converter needs to be further analyzed according to the data information of the molten iron amount, the steel scrap amount, the molten iron temperature, the molten iron components and the like in the historical data of the converter production detection and test system and the weighing system. The prediction of the oxygen demand of the converter is also required to be combined with a converter production scheduling plan, the furnace blowing time and the furnace interval time in a future period of time are predicted according to the scheduling plan, and when the actual production and the converter production scheduling are different, the system can automatically correct the predicted data according to the actual production condition, so that the continuous accumulation of prediction errors is avoided, and the data are ensured to be matched with the actual production all the time.
And entering the oxygen pipe network for calculation according to the oxygen consumption prediction results of the blast furnace, the converter and other users as boundary input conditions, and quantitatively analyzing whether the oxygen pipe network and the spherical tank have overpressure or decompression or not through the calculation of the oxygen pipe network. When the pressure or the flow in the pipeline exceeds a certain range, the pipe network calculation function gives corresponding alarm information according to the calculation result. And finally, the oxygen optimization scheduling function gives a scheduling instruction to the oxygen system according to the production and consumption balance condition of the prediction result and the condition that whether the alarm exists in the pipe network calculation: for example, when the oxygen production is larger than the oxygen consumption, a suggestion for properly reducing the load of the oxygen generator set is given, and when the oxygen consumption is larger than the production thought, a condition that the production rhythm of the converter is properly adjusted to avoid simultaneous blowing of a plurality of converters is given. The flow of the oxygen optimization scheduling calculation method is shown in fig. 1, and specifically includes the following steps:
step 1: constructing an oxygen production and consumption prediction model, and predicting the oxygen consumption of a blast furnace, a converter and other users in a future period of time; and correcting the prediction result of the oxygen production and consumption prediction model according to the actual production condition of the converter.
The method for constructing the oxygen production and consumption prediction model specifically comprises the following steps:
1) oxygen production and consumption prediction
The oxygen production and consumption prediction refers to the function of predicting the oxygen consumption of the blast furnace and other oxygen users. The oxygen enrichment of the blast furnace is used as a large user of oxygen consumption, the relative stability of the oxygen consumption is determined due to the characteristics of the production process of the blast furnace, and the data characteristics of the oxygen enrichment flow data of a certain blast furnace are shown as that the average fluctuation amount per minute is below 1.8 percent according to the statistical data of the oxygen enrichment flow of the certain blast furnace for 48 hours and can be regarded as stable data. The prediction of the oxygen-rich flow of the blast furnace can adopt a method of moving average autoregressive or BP neural network, and the calculation flow is shown in figure 2.
If the prediction is carried out by utilizing a moving average autoregressive method, training a model through historical data to determine the regression term number, the moving average term number and the difference term number; if a BP neural network is used for prediction, the model is trained through historical data to determine weights inside the neural network. Due to the stability of the oxygen consumption of the blast furnace, the prediction result can obtain better effect no matter the prediction is carried out by adopting a moving average autoregressive method or a BP neural network.
2) Analysis and statistics of oxygen consumption of converter
The statistical analysis of the oxygen consumption of the converter is realized by finding the relationship between the production process data of the converter and the oxygen consumption of the corresponding furnace and fitting an oxygen consumption calculation formula. During the oxygen blowing period of the converter production, the oxygen blowing time and the total oxygen consumption are greatly related to the components of molten iron. The oxygen prediction process may take into account the amount of molten iron, the amount of scrap (molten iron ratio), the components of molten iron (carbon, silicon, manganese, phosphorus, sulfur), and the like. In the process of predicting oxygen consumption in converter blowing, firstly, a large amount of historical data of the production process of the converter is required to be collected, then, the correlation between parameters such as the amount of molten iron, the amount of scrap steel and the composition of molten iron and the oxygen consumption is determined through data analysis, secondly, the parameter which has the largest influence on the oxygen consumption is selected according to the correlation, and finally, a calculation formula is fitted between the selected process parameters and the oxygen consumption by a formula fitting method, wherein the calculation process is shown in fig. 3.
The historical data of 230 times of heavy steel are used in the testing process of the functional module, wherein the historical data comprises 10 data of scrap steel weight, molten iron silicon content, molten iron manganese content, molten iron phosphorus content, molten iron sulfur content, molten iron temperature, molten iron carbon content, molten iron titanium content and oxygen consumption.
In the data analysis process, not only the correlation between a single parameter and the oxygen consumption is considered, but also the relationship between the combination of a plurality of parameters and the oxygen consumption is considered. For example, when the relationship between the molten iron content and the oxygen consumption amount is considered alone, the correlation is not high, but if the data obtained by combining the molten iron weight and the molten iron content (molten iron weight x molten iron content) is considered, the correlation with the oxygen consumption amount increases. This can be interpreted from a process point of view as: the correlation between the actual oxygen consumption and the total mass of various components in the molten iron is high (molten iron weight. molten iron component), and the correlation between the actual oxygen consumption and the percentage of various components in the unit molten iron (molten iron component) is low.
The calculated correlation values in table 1 can be used to represent the magnitude of the correlation between the corresponding parameter and the oxygen consumption per furnace: positive values indicate positive correlation, negative values indicate negative correlation, and larger absolute values indicate greater correlation. Before fitting the oxygen consumption calculation formula, parameters with high correlation need to be selected, and in a certain calculation case, the selected parameters include: weight of molten iron x molten iron temperature, weight of molten iron x molten iron Si, weight of molten iron x molten iron Mn, weight of molten iron x molten iron P, weight of molten iron x molten iron S, weight of molten iron x molten iron C, weight of molten iron x molten iron Ti.
TABLE 1 correlation analysis results of process parameters and oxygen consumption
The fitting formula is selected as a linear function, and the fitting formula is as follows:
Qoxygen consumptionWeight of molten iron (a) weight of molten iron (b) weight of molten iron (Si + C) weight of molten iron (Mn + d) weight of molten iron (P + e) weight of molten iron (S + f) weight of molten iron (C + g) weight of molten iron (Ti)
The a, b, c, d, e, f and g are respectively calculation coefficients of each parameter, and the results show that the fitted curve can basically reflect the change of oxygen consumption in the smelting process of each furnace, for example, after 150 furnaces, the oxygen consumption is obviously increased due to the change of molten iron components, and the average error value between the calculated oxygen consumption and the actual oxygen consumption of each furnace is 4.6% according to error statistics. The fitting calculation result can be used as the basis of oxygen prediction calculation and can also be used as the basis of converter heat characteristic classification.
3) Converter beat prediction
The converter is a main user of an oxygen system, the oxygen consumption of the converter is closely related to the production cycle of the converter, so the oxygen consumption prediction needs to be closely combined with the prediction of the production cycle of the converter. Fig. 4 shows a flowchart of the converter beat prediction function.
In the production process of iron and steel enterprises, the steel-making process schedules the production of the converter according to production scheduling, and if the converter smelting is strictly executed according to the production scheduling, the prediction of the converter tempo can be completed only by taking the production scheduling as the input of a program. However, in many cases, the steel-making process may have a missing production schedule or an untimely update, and may have a deviation between the actual tempo of the converter and the production schedule due to an emergency situation of actual production or a habit of personal operation. In order to meet the requirement that the converter beat prediction function can normally operate and work under any condition, the functions of carrying out beat prediction through converter historical data and carrying out real-time correction on the converter beat and the consumption according to the actual production condition are introduced into the converter beat prediction.
Firstly, the converter beat prediction function predicts the smelting time of each furnace and the interval time between the furnaces according to the production schedule of the converter, and then calculates the oxygen consumption of each furnace according to the molten iron data in the detection and test system and the weight detection system. When the data in the converter production scheduling, the testing system or the weighing system is missing or not updated timely, the system can complement the data required in the beat prediction by reading the historical data of the converter production and in a weighted moving average mode. Secondly, the converter beat prediction function can compare whether the oxygen consumption prediction data and the actual oxygen consumption data of the converter production have deviation in real time, and if the deviation exists, the converter oxygen consumption beat and the consumption can be corrected to ensure that the prediction value is always consistent with the actual production.
Fig. 5 is a data and parameter description for predicting the oxygen tact and consumption of the converter, which actually predicts three parameters, as shown in fig. 5, which are: gap _ t, Q _ t and Q _ O2Wherein Gap _ t represents the time interval between two smelting heats, Q _ t represents the oxygen blowing duration of a single heat, Q _ O2The total oxygen consumption of a single furnace (i.e. the curve area of the curve of the instantaneous flow rate of each furnace and the surrounding of the time axis in the figure) is represented, wherein the total oxygen consumption of the single furnace is equal to the oxygen blowing duration multiplied by the instantaneous flow rate of oxygen of the single furnace.
Finally, the oxygen prediction beat and consumption result is compared with the actually produced oxygen consumption to see whether the predicted value is deviated from the actually generated value, if so, the actually generated value is judged to be the deviationIf deviation exists, the converter blowing beat and the oxygen consumption amount need to be corrected in real time. The function of real-time correction of the prediction result according to the actual situation is only used in the converter oxygen prediction, because the converter oxygen consumption is closely related to the converter heat beat, if the prediction quantity is found to have deviation from the actual production situation and error correction is not carried out, the error quantity can be accumulated continuously, so that the prediction value is seriously deviated from the actual production situation. Fig. 6 shows a schematic diagram of real-time correction of oxygen beat and consumption of a converter, in which the abscissa of fig. 6 represents a time axis, and the ordinate represents oxygen consumption, wherein the black vertical line represents the current time, the dark gray line represents predicted oxygen consumption, and the light gray line represents actual oxygen consumption. The data in fig. 6 are used for functional development and testing, so that the oxygen consumption can be clarified for a future period of time. When the predicted data and the actual production consumption data have deviation, namely the dark gray line and the light gray line are not coincident, correction is needed, and a corresponding correction method is determined by analyzing all deviation conditions which are predicted and may actually occur in the function development. The judgment and correction of the prediction situation need to first judge whether the difference occurs at the beginning of a cycle or at the end of a cycle. The first case: if the difference occurs when a period starts, the actual period starts to occur first, and the predicted period starts to occur later, namely the actual blowing starts in advance, the predicted value does not need to be adjusted at this moment to be aligned with the actual value, and the calculated value of Gap _ t needs to be updated by combining scheduling data or converter production historical data due to the occurrence of new Gap _ t; in the second case, if the difference occurs at the beginning of a cycle, the predicted cycle occurs first, and the actual cycle occurs later, i.e., the predicted blowing begins earlier, at which time the time of measurement is earlier, then Q _ t and Q _ O are measured2Reassigning, and pushing the prediction time backwards; in the third case, if the actual period ends first and the predicted period does not end yet, i.e., the actual blowing ends earlier, the predicted value does not need to be adjusted to align with the actual value at this time, and new Q _ t and Q _ O appear2Therefore, a knot is requiredThe scheduled data or converter production history data pair Q _ t and Q _ O2Updating the calculated value; in the fourth case, if the predicted period ends first and the actual period does not end yet, i.e., the predicted blowing ends earlier, at the end of a period, Q _ t and Q _ O are needed to be adjusted2And re-assigning values and pushing the cycle end time backwards.
Step 2: and (3) constructing an oxygen pipe network calculation model, inputting the prediction result corrected in the step (1) as a boundary condition into the oxygen pipe network calculation model, and calculating to obtain the condition of the pressure change of the oxygen pipe network in a period of time in the future.
The oxygen pipe network calculation model is an important reference basis for oxygen scheduling, when the consumption of oxygen users sharply increases, the oxygen stored in the spherical tank in the oxygen pipe network is not enough to meet the demand of the users, the pressure of the oxygen pipe network is reduced along with the increase, and the normal production of user units is even influenced when the pressure is serious; on the contrary, when the consumption of oxygen users is sharply reduced, the oxygen stored in the spherical tank in the oxygen pipe network is increased, the pressure of the oxygen pipe network is also increased, and the oxygen pipe network is diffused seriously, so that the energy is not used economically.
The boundary condition of oxygen pipe network calculation is prediction data (after correction) of oxygen production and consumption, and the condition of oxygen pipe network pressure change in a period of time in the future is calculated by combining pipe network calculation and oxygen production and consumption prediction.
The calculation of the oxygen pipe network can be divided into two parts, one part is the hydraulic calculation of the pipeline, and the other part is the pressure calculation of the oxygen spherical tank. In the calculation of the pipeline, a node method is mainly adopted for solving, the hydraulic calculation of a branch pipe network and a ring pipe network can be solved, and the solving process mainly comprises 3 equations: a) a node flow continuous equation; b) a pipe section pressure drop equation; c) a pipe segment flow equation. In the calculation of the pressure change of the oxygen spherical tank, a gas state equation is used, and the pressure change of the oxygen spherical tank caused by the unbalance amount of oxygen production and consumption can be obtained.
The oxygen production and consumption prediction data are used as the flow and pressure data obtained by calculating the input conditions of the pipe network calculation, and on the premise of ensuring the prediction accuracy, the oxygen production and consumption prediction data can be regarded as the pressure and flow change condition of the oxygen pipe network in a period of time in the future, and the pressure and flow change condition is used as a reference basis for oxygen scheduling.
And step 3: an oxygen scheduling module: and (4) carrying out optimized scheduling on the oxygen system according to the oxygen consumption prediction value corrected in the step (1) and the oxygen network pressure change situation in the future period of time calculated in the step (2).
The air source in the oxygen system is an oxygen generator set, users are a blast furnace, a converter and other users, and the blast furnace, the converter and other users are nonadjustable users to the oxygen system, so that only the oxygen generator set can be used as an adjustable unit in the oxygen scheduling process. Only when a plurality of converters blow oxygen simultaneously and the oxygen generator set cannot meet the oxygen consumption requirement even under full load, the oxygen scheduling module can provide a scheduling suggestion for avoiding simultaneous oxygen blowing of the plurality of converters for the steelmaking process, and the flow schematic diagram of the whole oxygen scheduling module is shown in fig. 7.
The oxygen scheduling module has the main functions of judging whether the oxygen pipe network has larger pressure fluctuation in a future period of time according to the results of prediction and pipe network calculation: if the pressure is suddenly increased, oxygen is diffused, so that the energy is uneconomical, and if the pressure is suddenly reduced, the process is affected because the user requirements are not satisfied. Through judgment, if the unbalance amount of oxygen production consumption can cause larger pressure fluctuation of a pipe network in a period of time in the future, the oxygen dispatching system needs to intervene in advance, firstly, the load regulation of the oxygen generator set is used for meeting the requirement change of an oxygen user, but when the load regulation of the oxygen generator set is not enough for meeting the regulation requirement, the oxygen dispatching system can also give an adjustment suggestion of the oxygen blowing beat of the converter to the steelmaking process, so that the severe pressure fluctuation of the oxygen pipe network caused by simultaneous blowing or simultaneous non-blowing of multiple converters for a long time is avoided.
Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.
Claims (10)
1. An oxygen scheduling calculation method combining production and consumption prediction and pipe network calculation is characterized by comprising the following steps:
s1: constructing an oxygen production and consumption prediction model, and predicting the oxygen consumption of a blast furnace, a converter and other users in a future period of time; correcting the prediction result of the oxygen production and consumption prediction model according to the actual production condition of the converter;
s2: constructing an oxygen pipe network calculation model, inputting the prediction result corrected in the step S1 as a boundary condition into the oxygen pipe network calculation model, and calculating to obtain the condition of oxygen pipe network pressure change in a period of time in the future;
s3: oxygen scheduling: and performing optimized scheduling on the oxygen system according to the oxygen consumption predicted value corrected in the step S1 and the oxygen network pressure change situation in the future period of time calculated in the step S2.
2. The oxygen scheduling calculation method of claim 1, wherein the step S1 is implemented by constructing an oxygen consumption prediction model including prediction of the oxygen-rich flow of the blast furnace by using a moving average autoregressive method or a BP neural network.
3. The oxygen scheduling calculation method according to claim 2, wherein the prediction is performed by a moving average autoregressive method, specifically, a model is trained by historical data to determine the regression term number, the moving average term number and the difference term number.
4. The oxygen scheduling calculation method according to claim 2, wherein the prediction is performed by using a BP neural network, specifically, a model is trained by historical data to determine weights inside the neural network.
5. The oxygen scheduling calculation method according to claim 1, wherein in step S1, the constructed oxygen production and consumption prediction model further comprises analysis statistics of oxygen consumption of the converter, and an oxygen consumption calculation formula is fitted by finding a relationship between process data of production of the converter and oxygen consumption of a corresponding furnace; the method specifically comprises the following steps: the method comprises the steps of firstly collecting a large amount of historical data of the production process of the converter, then determining the correlation among the molten iron amount, the scrap steel amount, the molten iron components and the oxygen consumption amount through data analysis, secondly selecting parameters with the largest influence on the oxygen consumption amount according to the correlation, and finally fitting a calculation formula between the selected process parameters and the oxygen consumption amount through a formula fitting method.
6. The oxygen scheduling calculation method according to claim 1, wherein in step S1, the constructed oxygen consumption prediction model further includes converter beat prediction; the method comprises the steps of introducing converter historical data to carry out beat prediction in converter beat prediction and carrying out real-time correction on converter beat and consumption according to actual production conditions; the method specifically comprises the following steps: firstly, a converter beat prediction model predicts smelting time of each furnace and interval time between the furnaces according to a converter production schedule, and then calculates oxygen consumption of each furnace according to molten iron data in a test system and a weighing system; when the data in the converter production scheduling, the testing system or the weighing system is missing or not updated timely, the system completes the data required in the rhythm prediction by reading the historical data of the converter production and in a weighted moving average mode; secondly, the converter beat prediction model compares whether the oxygen consumption prediction data and the actual oxygen consumption data of the converter production exist deviation in real time, if so, the converter oxygen consumption beat and the consumption are corrected to ensure that the prediction value is consistent with the actual production all the time.
7. The oxygen scheduling calculation method according to claim 6, wherein when there is a deviation between the predicted oxygen consumption data and the actual production consumption data of the converter, the method for correcting the oxygen consumption tempo and consumption of the converter comprises: judging and correcting the prediction condition, firstly judging whether the prediction condition is different at the beginning of a period or different at the end of the period;
the first case: if the difference occurs when a period starts, the actual period starts to occur first, and the predicted period starts to occur later, namely the actual blowing starts in advance, at the moment, not only the predicted value needs to be adjusted to be aligned with the actual value, but also the calculated value of Gap _ t needs to be updated by combining the scheduling data or the historical data of the converter production; wherein Gap _ t represents the time interval between two smelting heats;
the second case: if the difference occurs at the beginning of a cycle, and the predicted cycle occurs first, but the actual cycle occurs later, i.e., the blowing is predicted to begin earlier, for Q _ t and Q _ O2Reassigning, and pushing the prediction time backwards; wherein Q _ t represents the oxygen blowing duration of a single heat, Q _ O2Represents the total oxygen consumption of a single heat;
in the third case, if the actual period ends first and the predicted period does not end yet, i.e., the actual blowing ends earlier, then not only the predicted value needs to be adjusted to be aligned with the actual value, but also the Q _ t and Q _ O pairs need to be combined with the scheduling data or the historical data of the converter production2Updating the calculated value;
in the fourth case, if the predicted period ends first and the actual period does not end yet, i.e., the predicted blowing ends earlier, at the end of a period, Q _ t and Q _ O are needed to be adjusted2And re-assigning values and pushing the cycle end time backwards.
8. The oxygen scheduling calculation method according to claim 1, wherein in step S2, the constructed oxygen pipe network calculation model includes two parts: one part is the hydraulic calculation of the pipeline, and the other part is the pressure calculation of the oxygen spherical tank.
9. The oxygen scheduling calculation method according to claim 8, wherein the hydraulic calculation of the pipeline is solved by using a node method, and the solution process includes 3 equations: a) a node flow continuous equation; b) a pipe section pressure drop equation; c) a pipe section flow equation;
and calculating the pressure of the oxygen spherical tank, and obtaining the pressure change of the oxygen spherical tank caused by the unbalance amount of the oxygen production consumption by using a gas state equation.
10. The oxygen scheduling calculation method according to claim 1, wherein in step S3, the oxygen scheduling specifically includes: according to the corrected prediction result and the result calculated by the oxygen pipe network calculation model, whether the oxygen pipe network has larger pressure fluctuation in a period of time in the future is judged, if the oxygen production consumption unbalance amount in the period of time in the future is judged, the oxygen dispatching system intervenes in advance, firstly, the requirement change of an oxygen user is met through the load regulation of the oxygen generator set, but when the load regulation of the oxygen generator set is not enough to meet the regulation requirement, the oxygen dispatching system also gives a regulation suggestion of the oxygen blowing beat of the converter to the steelmaking process, and therefore the severe pressure fluctuation of the oxygen pipe network caused by simultaneous blowing or simultaneous non-blowing of multiple converters for a long time is avoided.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111088740.1A CN113962050A (en) | 2021-09-16 | 2021-09-16 | Oxygen scheduling calculation method combining production consumption prediction and pipe network calculation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111088740.1A CN113962050A (en) | 2021-09-16 | 2021-09-16 | Oxygen scheduling calculation method combining production consumption prediction and pipe network calculation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113962050A true CN113962050A (en) | 2022-01-21 |
Family
ID=79461765
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111088740.1A Pending CN113962050A (en) | 2021-09-16 | 2021-09-16 | Oxygen scheduling calculation method combining production consumption prediction and pipe network calculation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113962050A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115232907A (en) * | 2022-09-23 | 2022-10-25 | 北京科技大学 | Method and system for predicting oxygen blowing amount in converter steelmaking |
| CN115331415A (en) * | 2022-10-14 | 2022-11-11 | 河北省科学院应用数学研究所 | Oxygen concentration early warning method and device, electronic equipment and readable storage medium |
| CN116085685A (en) * | 2023-03-15 | 2023-05-09 | 上海叁零肆零科技有限公司 | Method and system for guaranteeing stable gas supply of natural gas in peak period of gas consumption |
| CN116640906A (en) * | 2023-07-27 | 2023-08-25 | 江苏永钢集团有限公司 | Ladle bottom blowing carbon dioxide smelting method and system based on 5G technology |
| WO2023179380A1 (en) * | 2022-03-23 | 2023-09-28 | 乔治洛德方法研究和开发液化空气有限公司 | Method and apparatus for controlling gas supply of gas supply system |
| CN116977115A (en) * | 2023-08-04 | 2023-10-31 | 重庆大学 | Web interactive steel production flow management configuration method |
-
2021
- 2021-09-16 CN CN202111088740.1A patent/CN113962050A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023179380A1 (en) * | 2022-03-23 | 2023-09-28 | 乔治洛德方法研究和开发液化空气有限公司 | Method and apparatus for controlling gas supply of gas supply system |
| CN115232907A (en) * | 2022-09-23 | 2022-10-25 | 北京科技大学 | Method and system for predicting oxygen blowing amount in converter steelmaking |
| CN115331415A (en) * | 2022-10-14 | 2022-11-11 | 河北省科学院应用数学研究所 | Oxygen concentration early warning method and device, electronic equipment and readable storage medium |
| CN116085685A (en) * | 2023-03-15 | 2023-05-09 | 上海叁零肆零科技有限公司 | Method and system for guaranteeing stable gas supply of natural gas in peak period of gas consumption |
| CN116640906A (en) * | 2023-07-27 | 2023-08-25 | 江苏永钢集团有限公司 | Ladle bottom blowing carbon dioxide smelting method and system based on 5G technology |
| CN116640906B (en) * | 2023-07-27 | 2023-10-20 | 江苏永钢集团有限公司 | Ladle bottom blowing carbon dioxide smelting method and system based on 5G technology |
| CN116977115A (en) * | 2023-08-04 | 2023-10-31 | 重庆大学 | Web interactive steel production flow management configuration method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113962050A (en) | Oxygen scheduling calculation method combining production consumption prediction and pipe network calculation | |
| CN108676955B (en) | A kind of BOF Steelmaking Endpoint carbon content and temprature control method | |
| CN111353656B (en) | Steel enterprise oxygen load prediction method based on production plan | |
| CN104630410B (en) | A kind of pneumatic steelmaking quality real-time dynamic forecast method based on data parsing | |
| CN109447346B (en) | Converter oxygen consumption prediction method based on gray prediction and neural network combined model | |
| Liu et al. | A dynamic analytics method based on multistage modeling for a BOF steelmaking process | |
| CN106096788B (en) | Converter steelmaking process cost control method and system based on PSO _ ELM neural network | |
| CN109359723A (en) | Based on the converter terminal manganese content prediction technique for improving regularization extreme learning machine | |
| CN112036081B (en) | Method for determining addition amount of silicon-manganese alloy in converter tapping based on yield prediction | |
| CN112215464A (en) | Prediction balance scheduling system for blast furnace gas under multiple working conditions | |
| CN100465294C (en) | Intelligent control method for bottom-blowing argon in refining furnace | |
| CN115522012B (en) | Method for controlling TSC stage oxygen blowing amount of large converter | |
| CN111664470A (en) | Automatic energy-saving combustion control system and equipment for hot blast stove | |
| CN106650944A (en) | Metallurgical enterprise converter gas scheduling method based on knowledge | |
| CN102952912A (en) | Blast furnace smelting method and system based on expert system and knowledge base | |
| CN111444942A (en) | Intelligent forecasting method and system for silicon content of blast furnace molten iron | |
| CN113239482A (en) | Dynamic prediction method and device for converter post-blowing carbon content | |
| Xu et al. | Reducing the fluctuation of oxygen demand in a steel plant through optimal production scheduling | |
| JP2005036289A (en) | Model parameter decision method and program therefor, and model prediction method and program therefor | |
| CN104714519A (en) | Method for setting and online optimizing technological parameters of production process of continuous annealing unit | |
| CN116911057A (en) | Real-time prediction method for temperature of molten pool in converter steelmaking converting process | |
| CN114997722A (en) | Steel enterprise byproduct gas balance optimization regulation and control system and method | |
| EP3989013B1 (en) | Method for controlling process, operation guidance method, method for operating blast furnace, method for producing hot metal, and device for controlling process | |
| CN115271153A (en) | Multi-objective planning-based regional industry structure adjustment optimization method and system | |
| CN105242545B (en) | A kind of Forecasting Methodology of the coverter pig oxygen flow based on process batch model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |