A hydrogen production data scheduling method and system for a metallurgical hydrogen working condition

Abstract

The present application relates to hydrogen production data scheduling technical field, disclose a kind of hydrogen production data scheduling method and system for metallurgical hydrogen working condition.The method comprises: constructing metallurgical hydrogen working condition image library;Metallurgical side working condition data and hydrogen production side resource data are obtained in rolling scheduling period;Data are time-aligned and effectively processed, and form scheduling sample sequence;Metallurgical hydrogen demand window and hydrogen production resource state matrix are generated based on scheduling sample sequence;Window response advance is calculated in combination with pipe network path, load climbing speed, quality stabilization time and hydrogen storage buffer state, and window reachability matrix is generated;Hydrogen production scheduling parameter is calculated according to metallurgical hydrogen demand window, hydrogen production resource state matrix and window reachability matrix;Hydrogen production scheduling parameter is issued to hydrogen production execution unit, and according to actual hydrogen supply data and actual furnace condition data, the next rolling scheduling period is corrected.The method can be aimed at furnace stage change in metallurgical hydrogen working condition, reducing atmosphere fluctuation and quality window difference Problem, realize the collaborative scheduling of hydrogen production load, purification drying, compression hydrogen storage and hydrogen supply flow.

Description

Technical Field

[0001] This invention relates to the field of hydrogen production data scheduling technology, and in particular to a hydrogen production data scheduling method and system for metallurgical hydrogen applications. Background Technology

[0002] Hydrogen or hydrogen-rich gases may be used in processes such as direct reduction ironmaking, strip annealing, metal powder reduction, and protective atmosphere heat treatment in the metallurgical industry. Unlike typical continuous hydrogen supply scenarios, hydrogen use in metallurgy does not only require meeting a fixed hydrogen production rate, but is coupled with parameters such as furnace cycle planning, furnace condition stage, material batch, furnace temperature, furnace pressure, tail gas hydrogen content, tail gas dew point, and reducing atmosphere intensity. Taking reduction processes as an example, hydrogen demand, pressure stability, dew point requirements, and hydrogen purity requirements will all change during stages such as heating, enhanced reduction, heat preservation, load reduction, and furnace cycle switching. If the hydrogen production side operates only at a fixed load, it is easy to encounter problems such as insufficient hydrogen supply response during peak hydrogen demand periods, or a mismatch between the hydrogen produced and the hydrogen storage and recovery capacity during low-load periods.

[0003] Existing hydrogen production control technologies typically focus on the hydrogen production unit itself as the primary control object. For example, they adjust the hydrogen production or purification power based on parameters such as electrolyzer voltage, current, temperature, pressure, hydrogen production rate, purification unit load, and hydrogen storage pressure. While this type of technology can maintain stable operation of the hydrogen production equipment, its data processing focus is usually concentrated on the internal state of the equipment, failing to fully express the impact of furnace stages, material reaction intensity, and tail gas conditions on hydrogen demand in the next period during metallurgical hydrogen use. Particularly during transitions in metallurgical furnace conditions, there are response times for hydrogen production load ramp-up, purification quality stabilization, and hydrogen storage pressurization. If scheduling is not done in advance based on furnace condition data, it can easily lead to a misalignment between hydrogen production output and the metallurgical hydrogen consumption window.

[0004] Existing hydrogen quality testing and purification adjustment technologies typically use equipment such as chromatography, spectroscopy, dew point meters, and oxygen content detectors to measure hydrogen purity, water content, and impurity content, and adjust purification or drying parameters based on the test results. While these technologies can improve the accuracy of hydrogen quality assessment, they often rely on uniform purity or dew point thresholds for control. In the metallurgical industry, different processes and furnace conditions have varying sensitivities to hydrogen quality. For example, the reduction enhancement stage focuses more on hydrogen reduction capacity and dew point margin, the protective atmosphere stage focuses more on trace oxygen, moisture, and pressure fluctuations, and the furnace switching stage focuses more on a smooth transition of hydrogen supply flow. Adjusting hydrogen production solely according to uniform quality standards makes it difficult to reflect the suitability of the produced hydrogen for specific metallurgical operating conditions.

[0005] Existing energy dispatching or hydrogen storage dispatching technologies typically determine hydrogen production capacity based on available electricity, energy storage status, time-of-use electricity pricing, or hydrogen storage pressure, which can improve the energy utilization efficiency of hydrogen production systems. However, hydrogen production and supply in metallurgical scenarios involve multiple execution stages, including electrolytic hydrogen production, hydrogen-containing gas purification, tail gas recovery, drying, compression, storage, and pipeline transportation. Each stage has different load adjustment rates and quality stabilization times. If dispatching is only performed from the energy side or the hydrogen storage side, problems may arise such as the total hydrogen supply being sufficient, but the pressure plateau, dew point margin, or purity level required for the target furnace cycle not being reached within the specified time, or high-quality hydrogen being allocated to low-priority operating conditions while critical reduction stages lack available hydrogen.

[0006] Therefore, hydrogen production for metallurgical applications still requires a method that considers data processing and scheduling control. This method would correlate metallurgical production plans, furnace conditions, material batches, furnace temperature and pressure, tail gas hydrogen content, tail gas dew point, hydrogen production equipment operating status, purification and drying status, hydrogen storage pressure, and pipeline pressure to form a metallurgical hydrogen demand window that describes time intervals, flow rates, pressures, purity, and dew point requirements. Furthermore, it would combine the hydrogen production capacity, load ramp-up capability, quality stabilization time, and hydrogen storage buffer status of hydrogen production resources to generate hydrogen production scheduling parameters. This would address the insufficient matching between the response characteristics of the hydrogen production side and the metallurgical hydrogen demand window. Summary of the Invention

[0007] This invention provides a hydrogen production data scheduling method and system for metallurgical hydrogen use, which addresses issues such as furnace condition stage changes, hydrogen quality window differences, hydrogen production execution unit response lag, and incoordination between hydrogen storage buffer and pipeline supply during the metallurgical hydrogen use process.

[0008] To achieve the above objectives, this invention provides a hydrogen production data scheduling method for metallurgical hydrogen use conditions, comprising: constructing a metallurgical hydrogen use condition profile library in a hydrogen production scheduling database, wherein the metallurgical hydrogen use condition profile library is used to record the baseline hydrogen consumption curve, baseline pressure curve, baseline quality requirements, tail gas response baseline value, and process priority of the metallurgical hydrogen use process under different furnace condition stages; acquiring metallurgical side operating condition data and hydrogen production side resource data in a rolling scheduling cycle; performing time alignment and data validity processing on the metallurgical side operating condition data and hydrogen production side resource data to form a scheduling sample sequence; identifying the current furnace condition stage based on the scheduling sample sequence, and generating a future predicted metallurgical hydrogen demand window in the time domain based on the current furnace condition stage, production plan, and metallurgical hydrogen use condition profile library; and generating a hydrogen production resource status matrix based on the hydrogen production side resource data in the scheduling sample sequence. Based on the pipeline route from the hydrogen production execution unit to the metallurgical hydrogen consumption process, the load ramp-up rate, the quality stabilization time, and the hydrogen storage buffer capacity, the window response lead of each hydrogen production execution unit relative to each metallurgical hydrogen demand window is calculated, and a window reachability matrix is ​​generated based on the window response lead and quality constraints. Based on the metallurgical hydrogen demand window, the hydrogen production resource status matrix, the window reachability matrix, and the hydrogen storage buffer status, hydrogen production scheduling parameters within the rolling scheduling cycle are calculated. These hydrogen production scheduling parameters are then distributed to the hydrogen production execution units, enabling them to prepare, purify, dry, compress, store, and supply hydrogen according to these parameters. Actual hydrogen supply data and actual metallurgical furnace condition data are collected, and the hydrogen consumption deviation feedback is calculated. Based on this feedback, the metallurgical hydrogen consumption condition profile library and the hydrogen production scheduling parameters for the next rolling scheduling cycle are corrected.

[0009] Furthermore, the construction of the metallurgical hydrogen operating condition profile library in the hydrogen production scheduling database includes: collecting metallurgical side operating condition data, hydrogen supply data, and hydrogen quality data from multiple historical production batches; classifying the collected data according to the metallurgical hydrogen use process type, furnace condition stage, and material batch category; calculating the baseline hydrogen use curve, baseline pressure curve, baseline hydrogen purity requirement, baseline hydrogen dew point requirement, baseline value of tail gas hydrogen content, baseline value of tail gas dew point, and allowable fluctuation range for each type of data; and storing the above data in association to form the metallurgical hydrogen operating condition profile library.

[0010] Furthermore, the process of performing time alignment and data validation on the metallurgical side operating data and hydrogen production side resource data to form a scheduling sample sequence includes: configuring a unified timestamp for the metallurgical side operating data and hydrogen production side resource data; determining the delay compensation amount based on pipeline length, average gas delivery speed, valve group action time, and sensor response time; performing time shifting on hydrogen supply-related data according to the delay compensation amount; resampling or maintaining valid values ​​for data with different sampling frequencies; removing abnormal data; and filling in missing data.

[0011] Furthermore, the step of identifying the current furnace condition stage based on the scheduling sample sequence and generating a future prediction window for metallurgical hydrogen demand based on the current furnace condition stage, production plan, and metallurgical hydrogen demand profile library includes: reading the process number, furnace start and end time, material batch, material loading amount, and target output within the future prediction time domain; normalizing and weighting the furnace temperature change rate, furnace pressure change rate, tail gas hydrogen content deviation, tail gas dew point deviation, and material batch coefficient to obtain a metallurgical hydrogen demand intensity value; determining the current furnace condition stage based on the metallurgical hydrogen demand intensity value and furnace event identifier; reading the corresponding baseline data from the metallurgical hydrogen demand profile library; correcting the baseline hydrogen demand curve based on the material loading amount, furnace temperature deviation, tail gas hydrogen content deviation, and tail gas dew point deviation, and forming a metallurgical hydrogen demand window.

[0012] Furthermore, the step of generating a hydrogen production resource status matrix based on the hydrogen production side resource data in the scheduling sample sequence includes: reading the current operating load, rated hydrogen production capacity, allowable load adjustment range, load ramp-up rate, hydrogen purity detection value, hydrogen dew point detection value, outlet pressure, quality stabilization time, hydrogen storage pressure, hydrogen storage volume, and unit hydrogen production energy consumption of each hydrogen production execution unit; calculating the available hydrogen production capacity, effective hydrogen supply capacity, and hydrogen storage buffer capacity; and generating the hydrogen production resource status matrix with the hydrogen production execution units as matrix rows and the corresponding capacity and status parameters as matrix columns.

[0013] Furthermore, the window response advance is determined according to the maximum value among the load ramp-up time, purification stabilization time, drying stabilization time, compression pressurization time, and pipeline transportation time. When the sum of the current time and the window response advance is not later than the start time of hydrogen consumption in the metallurgical hydrogen demand window, and the corresponding hydrogen production execution unit meets the requirements of flow rate, pressure, purity, and dew point, the corresponding element in the window reachability matrix is ​​marked as reachable; otherwise, it is marked as unreachable and the gap is recorded.

[0014] Furthermore, the calculation of hydrogen production scheduling parameters within the rolling scheduling cycle includes: generating a set of candidate scheduling parameters; predicting the hydrogen supply flow rate, hydrogen supply pressure, hydrogen purity, hydrogen dew point, hydrogen storage pressure change, and energy consumption corresponding to each candidate scheduling parameter; evaluating each candidate scheduling parameter according to the scheduling evaluation value G; and selecting the candidate scheduling parameter with the smallest scheduling evaluation value G that satisfies the reachability state in the window reachability matrix as the hydrogen production scheduling parameter.

[0015] Furthermore, before sending the hydrogen production scheduling parameters to the hydrogen production execution unit, the method further includes: reconstructing the hydrogen storage pressure trajectory, pipeline pressure trajectory, and hydrogen supply flow trajectory in the future prediction time domain based on the hydrogen production scheduling parameters; and adjusting the hydrogen production load target, purification load target, compressor operation target, or hydrogen storage pressure target when the hydrogen storage pressure trajectory or pipeline pressure trajectory does not meet the constraints.

[0016] Furthermore, the process of collecting actual hydrogen supply data and actual metallurgical furnace condition data, calculating hydrogen consumption deviation feedback, and correcting the metallurgical hydrogen consumption condition profile library and the hydrogen production scheduling parameters for the next rolling scheduling cycle based on the hydrogen consumption deviation feedback includes: collecting actual hydrogen supply data and actual furnace condition data after the metallurgical hydrogen demand window ends; calculating flow deviation, pressure deviation, purity deviation, dew point deviation, tail gas hydrogen deviation, and tail gas dew point deviation; obtaining the hydrogen consumption deviation feedback by weighted summation; and correcting the metallurgical hydrogen consumption condition profile library according to an exponential weighting method when the hydrogen consumption deviation feedback is greater than a preset feedback threshold.

[0017] Furthermore, the present invention also provides a hydrogen production data scheduling system for metallurgical hydrogen applications, including a memory, a processor, a hydrogen production scheduling database, and a hydrogen production control interface. The memory stores a computer program, which, when executed by the processor, implements the above-mentioned method.

[0018] This invention has at least the following beneficial effects: First, by incorporating the metallurgical hydrogen operating condition profile library and the metallurgical hydrogen demand window, the furnace stage, material batch, tail gas hydrogen content, and tail gas dew point are included in the hydrogen production scheduling basis, so that the hydrogen production process no longer operates solely based on the state of the hydrogen production equipment itself; Second, by using the window response lead time, load ramp-up, purification stabilization, drying stabilization, compression pressurization, and pipeline transportation are unified into time accessibility constraints, enabling advance handling of hydrogen production preparation issues before the peak of metallurgical hydrogen demand; Third, by using the window accessibility matrix and scheduling evaluation values, matching can be performed between multiple hydrogen production execution units, hydrogen storage units, and metallurgical hydrogen processes, reducing the risk that the total hydrogen supply is met but the quality or pressure window is not. Fourth, by correcting the metallurgical hydrogen operating condition profile library through hydrogen usage deviation feedback, the next rolling scheduling cycle can be adaptively corrected based on the actual furnace condition response. Attached Figure Description

[0019] Figure 1 This is a flowchart of a hydrogen production data scheduling method for metallurgical hydrogen applications, according to the present invention. Detailed Implementation

[0020] The present invention will be described below with reference to the accompanying drawings and embodiments. It should be understood that the embodiments are used to illustrate the technical solution of the present invention and do not limit the specific application scope of the present invention. The hydrogen production execution unit in the present invention can be one or more combinations of an electrolytic hydrogen production unit, a hydrogen-containing tail gas recovery and purification unit, a purification and drying unit, a compression hydrogen storage unit, and a hydrogen supply valve group. The metallurgical hydrogen-using process can be a process that requires hydrogen, such as direct reduction ironmaking, strip steel annealing, metal powder reduction, or protective atmosphere heat treatment.

[0021] like Figure 1As shown, this embodiment provides a data scheduling method for hydrogen production in metallurgical hydrogen applications. This method executes on a rolling scheduling cycle, which can be from 1 to 15 minutes, and the future prediction time domain can be from 30 to 180 minutes. In one example, the rolling scheduling cycle is 5 minutes, and the future prediction time domain is 60 minutes. It is worth noting that the rolling scheduling cycle and the future prediction time domain can be set according to the metallurgical furnace volume, the length of the hydrogen supply pipeline, the load ramp-up capability of the hydrogen production execution unit, and the hydrogen storage capacity.

[0022] Furthermore, a metallurgical hydrogen usage profile database is constructed within the hydrogen production scheduling database. Specifically, metallurgical-side operating condition data, hydrogen supply data, and hydrogen quality data are collected from multiple historical production batches. The metallurgical-side operating condition data includes furnace number, process number, material batch, material loading amount, furnace temperature, furnace pressure, tail gas hydrogen content, and tail gas dew point; the hydrogen supply data includes hydrogen supply flow rate, hydrogen supply pressure, valve group opening degree, and pipeline pressure; and the hydrogen quality data includes hydrogen purity and hydrogen dew point. Subsequently, the collected data is categorized according to the metallurgical hydrogen usage process type, furnace condition stage, and material batch category. The furnace condition stage can include the heating stage, reduction enhancement stage, heat preservation stage, load reduction stage, and furnace switching stage.

[0023] In one example, statistical analysis of historical batches of a direct reduction furnace yielded a baseline hydrogen consumption curve of 900 Nm³ / h to 1200 Nm³ / h for the reduction enhancement stage, a baseline hydrogen supply pressure of 0.55 MPa, a baseline hydrogen purity requirement of no less than 99.90%, a baseline hydrogen dew point requirement of no higher than -60°C, a baseline tail gas hydrogen content of 42%, and a baseline tail gas dew point of -35°C. By associating these parameters with process priorities and storing them, a profile entry for that process during the reduction enhancement stage can be created. Furthermore, the profile library does not simply record fixed thresholds; rather, it binds furnace condition stages with hydrogen consumption curves, quality windows, and tail gas response baseline values, enabling subsequent scheduling to differentiate between different metallurgical hydrogen consumption conditions.

[0024] Furthermore, time alignment and data validation are performed on the metallurgical side operating data and the hydrogen production side resource data to form a scheduling sample sequence. Since there is a pipeline delivery delay in the arrival of hydrogen supply flow, pressure, and quality detection values ​​at the furnace end from the hydrogen production side, this embodiment first configures a unified timestamp for all data, and then determines the delay compensation amount based on pipeline length, average gas delivery speed, valve group action time, and sensor response time. The delay compensation amount can be expressed as: Δtij equals Sij divided by vij plus tv plus ts.

[0025] Where Δtij represents the delay compensation amount from the i-th hydrogen production execution unit to the j-th metallurgical hydrogen process, Sij represents the pipeline path length, vij represents the average gas delivery speed, tv represents the valve group action time, and ts represents the sensor response time.

[0026] In one example, the pipeline path length from the hydrogen production unit to the direct reduction furnace is 180m, the average gas delivery speed is 12m / s, the valve group action time is 3s, and the sensor response time is 2s. Therefore, the delay compensation is 180 / 12 + 3 + 2, which equals 20s. The system performs a 20-second time shift on the hydrogen supply flow rate, pipeline pressure, hydrogen purity, and hydrogen dew point to align them with the furnace condition data. For data with a sampling frequency higher than the rolling scheduling cycle, the mean, median, or last effective value within the cycle can be used; for data with a sampling frequency lower than the rolling scheduling cycle, effective value preservation can be implemented. Anomaly removal can employ equipment range constraints, adjacent sampling rate of change constraints, and historical distribution constraints of similar furnace conditions. For example, if the hydrogen supply pressure sensor range is 0 to 1.6MPa, a sampled value of 2.2MPa can be considered an anomaly.

[0027] Furthermore, the system identifies the current furnace condition stage based on the scheduling sample sequence and generates a future prediction window for metallurgical hydrogen demand based on the current furnace condition stage, production plan, and metallurgical hydrogen demand profile library. The system reads the process number, furnace start and end times, material batches, material loading amounts, and target output for the future prediction time domain from the production plan. Subsequently, it performs normalized weighting on the furnace temperature change rate, furnace pressure change rate, tail gas hydrogen content deviation, tail gas dew point deviation, and material batch coefficient to obtain the metallurgical hydrogen demand intensity value.

[0028] The intensity value K of hydrogen in metallurgical applications can be expressed as: K equals wT multiplied by xT plus wP multiplied by xP plus wH multiplied by xH plus wD multiplied by xD plus wM multiplied by xM.

[0029] Where xT represents the normalized furnace temperature change rate, xP represents the normalized furnace pressure change rate, xH represents the normalized tail gas hydrogen content deviation, xD represents the normalized tail gas dew point deviation, xM represents the material batch coefficient, and wT, wP, wH, wD, and wM are weighting coefficients, with the sum of all weighting coefficients being 1. The tail gas hydrogen content deviation can be obtained by normalizing the difference between the tail gas hydrogen content benchmark value and the actual tail gas hydrogen content; the tail gas dew point deviation can be obtained by normalizing the difference between the actual tail gas dew point and the tail gas dew point benchmark value.

[0030] In one example, the normalized value xT corresponding to the furnace temperature change rate is 0.65, the normalized value xP corresponding to the furnace pressure change rate is 0.30, the tail gas hydrogen content deviation xH is 0.70, the tail gas dew point deviation xD is 0.55, and the material batch coefficient xM is 0.40, with weights of 0.25, 0.15, 0.30, 0.20, and 0.10 respectively. Therefore, K is 0.25 × 0.65 + 0.15 × 0.30 + 0.30 × 0.70 + 0.20 × 0.55 + 0.10 × 0.40, resulting in 0.5675. If the system presets the operating intensity value range of the reduction enhancement stage to 0.50 to 0.80, and the furnace event identifier indicates that it is currently in the reduction stage, then the current furnace condition stage can be identified as the reduction enhancement stage.

[0031] When generating the target hydrogen supply flow rate, the baseline hydrogen supply curve can be corrected using the following formula: Qjreq equals Qbase multiplied by the sum of the numbers in parentheses: 1 + αmδm + αTδT + αHδH + αDδD.

[0032] Where Qjreq represents the target hydrogen supply flow rate for the j-th metallurgical hydrogen demand window, Qbase represents the baseline hydrogen curve value for the corresponding furnace condition stage in the metallurgical hydrogen operating condition profile library, δm represents the material charging deviation, δT represents the furnace temperature deviation, δH represents the tail gas hydrogen content deviation, δD represents the tail gas dew point deviation, and αm, αT, αH, and αD are correction coefficients. The corrected target hydrogen supply flow rate can be limited between the minimum and maximum flow rates allowed for the corresponding process to avoid the calculated results exceeding the achievable range of the equipment.

[0033] In one example, Qbase is 1000 Nm³ / h, δm is 0.08, δT is 0.04, δH is 0.06, δD is 0.05, and αm, αT, αH, and αD are 0.50, 0.30, 0.40, and 0.20, respectively. Therefore, Qjreq is 1000 × (1 within parentheses) + 0.04 + 0.012 + 0.024 + 0.010, resulting in 1086 Nm³ / h. The system encapsulates 1086 Nm³ / h, 0.55 MPa, 99.90% hydride content, -60°C, and the allowable flow rate fluctuation range as a metallurgical hydrogen demand window.

[0034] Furthermore, a hydrogen production resource status matrix is ​​generated based on the hydrogen production side resource data in the scheduling sample sequence. For each hydrogen production execution unit, the system reads the current operating load, rated hydrogen production capacity, allowable load adjustment range, load ramp-up rate, hydrogen purity detection value, hydrogen dew point detection value, outlet pressure, quality stabilization time, hydrogen storage pressure, hydrogen storage volume, and unit hydrogen production energy consumption.

[0035] The available hydrogen production capacity Ci can be determined by the rated hydrogen production capacity, the allowable load adjustment limit, and the current operational status. Effective hydrogen supply capacity can further consider the load ramp-up rate and quality stabilization time within the future forecast time domain. Hydrogen storage buffer capacity can be determined based on hydrogen storage pressure, hydrogen storage volume, and pipeline pressure. It is worth noting that the quality stabilization time is not a fixed detection delay, but rather refers to the time required for the hydrogen production unit to restore hydrogen purity and hydrogen dew point to the target quality window after a load change.

[0036] In one example, the first hydrogen production unit is an electrolysis hydrogen production unit with a rated hydrogen production capacity of 800 Nm³ / h, a current operating load of 520 Nm³ / h, a load ramp-up rate of 35 Nm³ / h per minute, a hydrogen purity of 99.95%, a hydrogen dew point of -65°C, a quality stabilization time of 6 minutes, an outlet pressure of 0.8 MPa, and a unit hydrogen production energy consumption of 4.7 kWh / Nm³. The second hydrogen production unit is a hydrogen-containing tail gas purification unit with a current effective hydrogen supply capacity of 350 Nm³ / h, a quality stabilization time of 12 minutes, a hydrogen purity of 99.80%, and a hydrogen dew point of -55°C. The system organizes the above data into a hydrogen production resource status matrix by rows and columns.

[0037] Furthermore, based on the pipeline route from the hydrogen production execution unit to the metallurgical hydrogen demand process, the load ramp-up rate, the quality stabilization time, and the hydrogen storage buffer capacity, the window response lead is calculated, and a window reachability matrix is ​​generated. For the i-th hydrogen production execution unit and the j-th metallurgical hydrogen demand window, the load ramp-up time τijramp can be expressed as: τijramp equals the difference between 0 (within the parentheses) and Qjreq minus Qicur, divided by ri.

[0038] Where Qjreq represents the target hydrogen supply flow rate for the j-th metallurgical hydrogen demand window, Qicur represents the current available flow rate for the i-th hydrogen production execution unit, and ri represents the load ramp-up rate of the i-th hydrogen production execution unit. When Qjreq is not greater than Qicur, the load ramp-up time is 0.

[0039] The compression repressurization time τijcomp can be calculated based on the target hydrogen supply pressure, the current hydrogen storage pressure, the equivalent buffer volume, and the compressor capacity. In one example, it can be expressed as: τijcomp equals the difference between 0 (within the parentheses) and Pjreq minus Ps, multiplied by Veq and then divided by kcqc.

[0040] Where Pjreq represents the target hydrogen supply pressure, Ps represents the current hydrogen storage pressure, Viq represents the equivalent buffer volume, kc represents the compression efficiency correlation coefficient, and qc represents the compressor's equivalent pressure compensation capability.

[0041] The window response lead Lij is: Lij equals max(τijramp, τipur, τidry, τijcomp, and τijpipe).

[0042] Where τipur represents the purification stabilization time, τidry represents the drying stabilization time, and τijpipe represents the pipeline transportation time. If a certain time item cannot be calculated due to equipment downtime or unavailable parameters, the element can be marked as unreachable, and the reason for unreachability can be recorded.

[0043] In one example, a reduction enhancement window is scheduled to begin in 30 minutes, with a target hydrogen supply flow rate of 1086 Nm³ / h. The current available flow rate is 760 Nm³ / h, the load ramp-up rate is 35 Nm³ / h per minute, and the load ramp-up time is 9.31 minutes. The purification stabilization time is 6 minutes, the drying stabilization time is 4 minutes, the compression pressurization time is 8 minutes, and the pipeline delivery time is 1 minute, resulting in a window response lead of 9.31 minutes. Since the current time plus 9.31 minutes is earlier than the start time of the window, and the purity, dew point, and pressure all meet the requirements, the corresponding element in the window reachability matrix is ​​marked as reachable.

[0044] Furthermore, based on the hydrogen demand window for metallurgical applications, the hydrogen production resource status matrix, the window reachability matrix, and the hydrogen storage buffer status, the system calculates the hydrogen production scheduling parameters within the rolling scheduling cycle. The system first generates a set of candidate scheduling parameters within the allowable operating range of the hydrogen production execution unit. These candidate scheduling parameters include candidate hydrogen production load, candidate purification load, candidate drying switchover time, candidate compressor operating target, candidate hydrogen storage pressure target, and candidate hydrogen supply flow rate target. Subsequently, based on each candidate scheduling parameter, the system predicts the future hydrogen supply flow rate, hydrogen supply pressure, hydrogen purity, hydrogen dew point, hydrogen storage pressure change, and energy consumption within the prediction time domain.

[0045] The scheduling evaluation value G can be expressed as: G equals Σρj (within parentheses) βQEjQ plus βPEjP plus βUEjU plus βDEjD, plus βEEE plus βSES plus βBEB plus βAEA.

[0046] Where j represents the metallurgical hydrogen demand window number, ρj represents the process priority weight of the j-th metallurgical hydrogen demand window, EjQ represents the flow deviation item, EjP represents the pressure deficiency item, EjU represents the purity deficiency item, EjD represents the dew point exceedance item, EE represents the energy consumption item, ES represents the equipment load switching item, EB represents the hydrogen storage pressure deviation item, EA represents the unreachable penalty item, and βQ, βP, βU, βD, βE, βS, βB and βA represent the corresponding weights.

[0047] The flow deviation term EjQ can be the squared relative deviation between the predicted and target hydrogen supply flow rates; the pressure deficiency term EjP can be the difference when the target hydrogen supply pressure is higher than the predicted hydrogen supply pressure, and is set to 0 if the predicted hydrogen supply pressure is not lower than the target hydrogen supply pressure; the purity deficiency term EjU can be the difference when the target hydrogen purity is higher than the predicted hydrogen purity; and the dew point exceedance term EjD can be the difference when the predicted hydrogen dew point is higher than the target hydrogen dew point. Since a lower dew point usually represents a lower moisture content, when the target hydrogen dew point is -60℃ and the predicted hydrogen dew point is -55℃, the dew point exceedance term is the normalized value corresponding to 5℃.

[0048] In one example, candidate scheduling parameter A corresponds to a predicted hydrogen supply flow rate of 1070 Nm³ / h, a predicted hydrogen supply pressure of 0.56 MPa, a predicted hydrogen purity of 99.93%, and a predicted hydrogen dew point of -63°C for the reduction enhancement window; candidate scheduling parameter B corresponds to a predicted hydrogen supply flow rate of 1100 Nm³ / h, a predicted hydrogen supply pressure of 0.54 MPa, a predicted hydrogen purity of 99.91%, and a predicted hydrogen dew point of -58°C. If the reduction enhancement window has a high weighting for pressure and dew point, candidate scheduling parameter A may receive a lower scheduling evaluation value because both pressure and dew point requirements are met. The system selects the candidate scheduling parameter with the smallest scheduling evaluation value G that satisfies the reachability state in the window reachability matrix as the hydrogen production scheduling parameter.

[0049] Furthermore, before distributing the hydrogen production scheduling parameters to the hydrogen production execution unit, the hydrogen storage pressure trajectory, pipeline pressure trajectory, and hydrogen supply flow trajectory within the future prediction time domain are reconstructed based on the hydrogen production scheduling parameters. The hydrogen storage pressure trajectory can be estimated based on changes in the gas volume within the hydrogen storage tank, while the pipeline pressure trajectory can be estimated based on the hydrogen supply flow rate, valve group opening degree, and pipeline resistance.

[0050] In one example, the lower limit of hydrogen storage pressure is 0.45 MPa, and the upper limit is 1.20 MPa. If it is predicted that the hydrogen storage pressure will drop to 0.42 MPa after 25 minutes, and a reduction enhancement window exists after 30 minutes, the system will increase the hydrogen production load target or compressor operation target in advance to ensure that the hydrogen storage pressure rises to no less than 0.45 MPa before the window begins. If it is predicted that the hydrogen storage pressure will rise to 1.25 MPa during the low-load phase, the system will reduce the hydrogen production load target or adjust the hydrogen storage pressure target to prevent the hydrogen storage pressure from exceeding the set range.

[0051] Furthermore, the hydrogen production scheduling parameters are distributed to the hydrogen production execution unit. The hydrogen production execution unit produces hydrogen according to the hydrogen production load target, purifies hydrogen according to the purification load target, switches drying towers according to the drying switching time, performs compression and pressurization according to the compressor operation target, and supplies hydrogen to the corresponding metallurgical hydrogen-using processes according to the hydrogen supply flow target and valve group control target. It is worth noting that the distribution of hydrogen production scheduling parameters can be achieved through an industrial control network, a distributed control system, or a programmable logic controller interface; the specific communication method does not affect the data scheduling logic of this invention.

[0052] Furthermore, actual hydrogen supply data and actual metallurgical furnace condition data are collected to calculate the hydrogen consumption deviation feedback, and to correct the metallurgical hydrogen consumption condition profile database and the hydrogen production scheduling parameters for the next rolling scheduling cycle. After the metallurgical hydrogen demand window ends, the system collects actual hydrogen supply flow rate, actual hydrogen supply pressure, actual hydrogen purity, actual hydrogen dew point, actual furnace temperature, actual furnace pressure, actual tail gas hydrogen content, and actual tail gas dew point.

[0053] The hydrogen deviation feedback quantity E can be expressed as: E equals γQeQ plus γPeP plus γUeU plus γDeD plus γHeH plus γReR.

[0054] Where eQ represents flow rate deviation, eP represents pressure deviation, eU represents purity deviation, eD represents dew point deviation, eH represents tail gas hydrogen deviation, eR represents tail gas dew point deviation, and γQ, γP, γU, γD, γH and γR are the corresponding weights.

[0055] In one example, the target hydrogen supply flow rate is 1086 Nm³ / h, and the actual hydrogen supply flow rate is 1040 Nm³ / h, so the flow rate deviation is 46 / 1086; the target hydrogen supply pressure is 0.55 MPa, and the actual hydrogen supply pressure is 0.53 MPa, so the pressure deviation is 0.02 / 0.55; the target hydrogen dew point is -60°C, and the actual hydrogen dew point is -57°C, so the dew point deviation is the normalized value corresponding to 3°C. If the weighted hydrogen consumption deviation feedback is greater than the preset feedback threshold, the system considers that the baseline hydrogen consumption curve or quality margin in the current profile library needs to be corrected.

[0056] The correction can be performed using an exponentially weighted method. For example, the baseline is updated using the hydrogen curve as follows: Qbasenew equals 1 minus η multiplied by Qbaseold plus η multiplied by Qactualaligned.

[0057] Where Qbasenew represents the updated baseline hydrogen supply curve value, Qbaseold represents the original baseline hydrogen supply curve value, Qactualaligned represents the actual hydrogen supply flow rate after time alignment, and η represents the update coefficient, which can range from 0.05 to 0.30. In one example, Qbaseold is 1000 Nm³ / h, Qactualaligned is 1060 Nm³ / h, and η is 0.10, then Qbasenew is 1006 Nm³ / h. Similarly, the baseline pressure curve, baseline quality requirements, and exhaust gas response baseline values ​​can be updated.

[0058] This embodiment also provides a hydrogen production data scheduling system for metallurgical hydrogen use conditions, including a memory, a processor, a hydrogen production scheduling database, and a hydrogen production control interface. The memory stores a computer program, which, when executed by the processor, implements the aforementioned method. The hydrogen production scheduling database stores a metallurgical hydrogen use condition profile library, scheduling sample sequences, metallurgical hydrogen demand windows, a hydrogen production resource status matrix, a window reachability matrix, and hydrogen production scheduling parameters. The hydrogen production control interface sends the hydrogen production scheduling parameters to the hydrogen production execution unit.

[0059] Through the above implementation methods, this invention combines metallurgical furnace condition stage identification, hydrogen demand window generation, hydrogen production resource status matrix, window response lead time, window reachability matrix, and hydrogen consumption deviation feedback into a continuous data scheduling chain. This enables the hydrogen production side to prepare for hydrogen production, purification, drying, compression, and supply in advance based on specific metallurgical hydrogen consumption conditions, thereby improving the time adaptability, quality adaptability, and pressure adaptability between hydrogen production output and the metallurgical hydrogen consumption window.

Claims

1. A data scheduling method for hydrogen production in metallurgical applications, characterized in that, include: A metallurgical hydrogen production condition profile database is constructed in the hydrogen production scheduling database. The metallurgical hydrogen production condition profile database is used to record the benchmark hydrogen production curve, benchmark pressure curve, benchmark quality requirements, tail gas response benchmark value and process priority of the metallurgical hydrogen production process under different furnace conditions. The metallurgical side operating condition data and hydrogen production side resource data are obtained by rolling scheduling cycle. The metallurgical side operating condition data includes production plan, furnace number identification, material batch, furnace temperature, furnace pressure, tail gas hydrogen content and tail gas dew point. The hydrogen production side resource data includes hydrogen production execution unit operating load, hydrogen purity, hydrogen dew point, hydrogen storage pressure, pipeline pressure, hydrogen supply flow rate and hydrogen supply valve group status. Time alignment and data validation processing are performed on the metallurgical side operating condition data and hydrogen production side resource data to form a scheduling sample sequence; Based on the scheduling sample sequence, the current furnace condition stage is identified, and a future prediction window for metallurgical hydrogen demand is generated according to the current furnace condition stage, production plan, and metallurgical hydrogen demand profile library. The metallurgical hydrogen demand window includes the start time of hydrogen use, the end time of hydrogen use, the target hydrogen supply flow rate, the target hydrogen supply pressure, the target hydrogen purity, the target hydrogen dew point, and the allowable flow fluctuation range. Based on the hydrogen production side resource data in the scheduling sample sequence, a hydrogen production resource status matrix is ​​generated. The hydrogen production resource status matrix is ​​used to characterize the available hydrogen production capacity, load ramp-up speed, quality stabilization time, outlet pressure capacity, hydrogen storage buffer capacity and unit hydrogen production energy consumption of each hydrogen production execution unit. Based on the pipeline route from the hydrogen production execution unit to the metallurgical hydrogen process, the load ramp-up rate, the quality stabilization time, and the hydrogen storage buffer capacity, calculate the window response lead of each hydrogen production execution unit relative to each metallurgical hydrogen demand window, and generate a window reachability matrix based on the window response lead and quality constraints. Based on the aforementioned metallurgical hydrogen demand window, hydrogen production resource status matrix, window accessibility matrix, and hydrogen storage buffer status, hydrogen production scheduling parameters within the rolling scheduling cycle are calculated. These hydrogen production scheduling parameters include hydrogen production load target, purification load target, drying switching time, compressor operation target, hydrogen storage pressure target, and hydrogen supply flow target. The hydrogen production scheduling parameters are sent to the hydrogen production execution unit, so that the hydrogen production execution unit can produce, purify, dry, compress, store and supply hydrogen according to the hydrogen production scheduling parameters; Collect actual hydrogen supply data and actual metallurgical furnace condition data, calculate the hydrogen consumption deviation feedback amount, and correct the metallurgical hydrogen consumption condition profile library and the hydrogen production scheduling parameters for the next rolling scheduling cycle based on the hydrogen consumption deviation feedback amount.

2. The hydrogen production data scheduling method for metallurgical hydrogen applications according to claim 1, characterized in that, The construction of a metallurgical hydrogen usage condition profile database within the hydrogen production scheduling database includes: Collect metallurgical side operating condition data, hydrogen supply data, and hydrogen quality data from multiple historical production batches; The collected data are classified according to the type of hydrogen processing in metallurgy, furnace condition stage, and material batch category. The furnace condition stage includes at least two of the following: heating stage, reduction enhancement stage, heat preservation stage, load reduction stage, and furnace switching stage. For each type of data, calculate the baseline hydrogen curve, baseline pressure curve, baseline hydrogen purity requirement, baseline hydrogen dew point requirement, baseline value of tail gas hydrogen content, baseline value of tail gas dew point, and allowable fluctuation range. The reference hydrogen curve, reference pressure curve, reference hydrogen purity requirement, reference hydrogen dew point requirement, reference value of tail gas hydrogen content, reference value of tail gas dew point, allowable fluctuation range and process priority are associated and stored to form the metallurgical hydrogen operating condition profile library.

3. The hydrogen production data scheduling method for metallurgical hydrogen applications according to claim 1, characterized in that, The process of performing time alignment and data validation on the metallurgical side operating condition data and hydrogen production side resource data to form a scheduling sample sequence includes: Configure a unified timestamp for the metallurgical side operating condition data and the hydrogen production side resource data; The delay compensation amount of hydrogen supply data relative to furnace condition data is determined based on the pipeline length from the hydrogen production unit to the metallurgical hydrogen process, the average gas delivery speed, the valve group action time, and the sensor response time. The hydrogen supply flow rate, pipeline pressure, hydrogen purity, and hydrogen dew point are shifted over time according to the aforementioned delay compensation amount. Data with a sampling frequency higher than the rolling scheduling period is resampled, while data with a sampling frequency lower than the rolling scheduling period retains its valid value. Data that exceeds the equipment's measurement range, violates the adjacent sampling rate of change constraint, or deviates from the historical distribution of similar furnace conditions will be anomaly removed. The missing data is filled in by interpolating adjacent valid values ​​and using the benchmark values ​​of the same stage in the metallurgical hydrogen operating condition profile library to obtain the scheduling sample sequence.

4. The hydrogen production data scheduling method for metallurgical hydrogen applications according to claim 1, characterized in that, The step of identifying the current furnace condition stage based on the scheduling sample sequence and generating a future prediction window for metallurgical hydrogen demand based on the current furnace condition stage, production plan, and metallurgical hydrogen operating condition profile library includes: Read the process number, furnace start and end time, material batch, material loading amount and target output in the future prediction time domain from the production plan; The furnace temperature change rate, furnace pressure change rate, tail gas hydrogen content deviation, tail gas dew point deviation and material batch coefficient are normalized and weighted to obtain the metallurgical hydrogen working condition intensity value. The current furnace condition stage is determined based on the metallurgical hydrogen operating condition intensity value and the furnace event identifier. Read the benchmark hydrogen curve, benchmark pressure curve and benchmark quality requirements corresponding to the process number, material batch category and current furnace condition stage from the metallurgical hydrogen operating condition profile library; The baseline hydrogen curve is corrected based on the material loading amount, furnace temperature deviation, tail gas hydrogen content deviation, and tail gas dew point deviation to obtain the target hydrogen supply flow rate. The target hydrogen supply flow rate, target hydrogen supply pressure, target hydrogen purity, target hydrogen dew point, and allowable flow fluctuation range are packaged into the metallurgical hydrogen demand window according to the start and end times of hydrogen use.

5. The hydrogen production data scheduling method for metallurgical hydrogen applications according to claim 1, characterized in that, The step of generating a hydrogen production resource status matrix based on hydrogen production-side resource data in the scheduling sample sequence includes: For each hydrogen production execution unit, read the current operating load, rated hydrogen production capacity, allowable load adjustment range, load ramp-up rate, hydrogen purity detection value, hydrogen dew point detection value, outlet pressure, quality stabilization time, hydrogen storage pressure, hydrogen storage volume, and unit hydrogen production energy consumption. The available hydrogen production capacity is calculated based on the current operating load, rated hydrogen production capacity, and allowable load adjustment range. The effective hydrogen supply capacity that can be achieved in the future prediction time domain is calculated based on the load ramp-up rate and quality stabilization time. The hydrogen storage buffer capacity is calculated based on the hydrogen storage pressure, hydrogen storage volume, and pipeline pressure. The hydrogen production resource status matrix is ​​generated by using each hydrogen production execution unit as a matrix row and the available hydrogen production capacity, effective hydrogen supply capacity, load ramp-up rate, quality stabilization time, hydrogen purity detection value, hydrogen dew point detection value, outlet pressure, hydrogen storage buffer capacity, and unit hydrogen production energy consumption as matrix columns.

6. The hydrogen production data scheduling method for metallurgical hydrogen applications according to claim 1, characterized in that, The calculation of the window response lead of each hydrogen production execution unit relative to each metallurgical hydrogen demand window, based on the pipeline route from the hydrogen production execution unit to the metallurgical hydrogen consumption process, the load ramp-up rate, the quality stabilization time, and the hydrogen storage buffer capacity, and the generation of a window reachability matrix based on the window response lead and quality constraints, includes: For the i-th hydrogen production execution unit and the j-th metallurgical hydrogen demand window, calculate the load ramp-up time, purification stabilization time, drying stabilization time, compression pressurization time, and pipeline transportation time. Calculate the window response lead using the following formula: Lij is equal to the maximum value among the load ramp-up time, purification stabilization time, drying stabilization time, compression pressurization time, and pipeline transportation time; When the sum of the current time and the window response advance is not later than the start time of hydrogen use in the j-th metallurgical hydrogen demand window, and the i-th hydrogen production execution unit meets the flow rate, pressure, purity, and dew point requirements of the j-th metallurgical hydrogen demand window, the corresponding element in the window reachability matrix is ​​marked as reachable. When the reachability conditions are not met, record the corresponding flow gap, pressure gap, purity gap, dew point exceedance and time gap, and mark the corresponding element in the window reachability matrix as unreachable.

7. The hydrogen production data scheduling method for metallurgical hydrogen applications according to claim 1, characterized in that, The calculation of hydrogen production scheduling parameters within the rolling scheduling cycle, based on the metallurgical hydrogen demand window, hydrogen production resource status matrix, window reachability matrix, and hydrogen storage buffer status, includes: Within the allowable operating range of the hydrogen production execution unit, a set of candidate scheduling parameters is generated. Each candidate scheduling parameter includes candidate hydrogen production load, candidate purification load, candidate drying switching time, candidate compressor operating target, candidate hydrogen storage pressure target, and candidate hydrogen supply flow target. Based on each candidate scheduling parameter, predict the hydrogen supply flow rate, hydrogen supply pressure, hydrogen purity, hydrogen dew point, hydrogen storage pressure change and energy consumption in the future prediction time domain. Each candidate scheduling parameter is evaluated according to the scheduling evaluation value G, which is the sum of the weighted flow deviation, weighted pressure insufficiency, weighted purity insufficiency and weighted dew point exceedance items for all metallurgical hydrogen demand windows, plus the energy consumption item, equipment load switching item, hydrogen storage pressure deviation item and unreachable penalty item. The candidate scheduling parameter that has the smallest scheduling evaluation value G and satisfies the reachability state in the window reachability matrix is ​​selected as the hydrogen production scheduling parameter.

8. The hydrogen production data scheduling method for metallurgical hydrogen applications according to claim 1, characterized in that, Before sending the hydrogen production scheduling parameters to the hydrogen production execution unit, the method further includes: Based on the hydrogen production scheduling parameters, the hydrogen storage pressure trajectory, pipeline pressure trajectory, and hydrogen supply flow trajectory in the future prediction time domain can be reconstructed. The hydrogen storage pressure trajectory is compared with the upper limit and lower limit of hydrogen storage pressure, and the pipeline pressure trajectory is compared with the target hydrogen supply pressure in the metallurgical hydrogen demand window. When the hydrogen storage pressure trajectory is lower than the lower limit of the hydrogen storage pressure, or the pipeline pressure trajectory is lower than the target hydrogen supply pressure, the hydrogen production load target, the purification load target, or the compressor operation target should be increased in advance. When the hydrogen storage pressure trajectory is higher than the upper limit of the hydrogen storage pressure, reduce the hydrogen production load target or adjust the hydrogen storage pressure target. After the adjusted hydrogen production scheduling parameters meet the constraints of hydrogen storage pressure, pipeline pressure, and quality, the adjusted hydrogen production scheduling parameters will be sent to the hydrogen production execution unit.

9. A hydrogen production data scheduling method for metallurgical hydrogen applications according to claim 1, characterized in that, The process of collecting actual hydrogen supply data and actual metallurgical furnace condition data, calculating hydrogen consumption deviation feedback, and correcting the metallurgical hydrogen consumption condition profile database and the hydrogen production scheduling parameters for the next rolling scheduling cycle based on the hydrogen consumption deviation feedback includes: After the demand window for hydrogen in metallurgy ends, the actual hydrogen supply flow rate, actual hydrogen supply pressure, actual hydrogen purity, actual hydrogen dew point, actual furnace temperature, actual furnace pressure, actual tail gas hydrogen content, and actual tail gas dew point are collected. Calculate the flow deviation between the actual hydrogen supply flow rate and the target hydrogen supply flow rate, the pressure deviation between the actual hydrogen supply pressure and the target hydrogen supply pressure, the purity deviation between the actual hydrogen purity and the target hydrogen purity, the dew point deviation between the actual hydrogen dew point and the target hydrogen dew point, the tail gas hydrogen content deviation between the actual tail gas hydrogen content and the tail gas hydrogen content benchmark value, and the tail gas dew point deviation between the actual tail gas dew point and the tail gas dew point benchmark value. The hydrogen usage deviation feedback quantity is obtained by weighted summing of the flow rate deviation, pressure deviation, purity deviation, dew point deviation, tail gas hydrogen deviation, and tail gas dew point deviation. When the hydrogen usage deviation feedback amount is greater than the preset feedback threshold, the benchmark hydrogen usage curve, benchmark pressure curve, benchmark quality requirements and tail gas response benchmark value under the corresponding metallurgical hydrogen usage process and corresponding furnace condition stage are corrected in an exponential weighting manner. The revised baseline hydrogen curve, baseline pressure curve, baseline quality requirements, and baseline tail gas response values ​​are written into the metallurgical hydrogen operating condition profile library and used for calculating hydrogen production scheduling parameters in the next rolling scheduling cycle.

10. A hydrogen production data scheduling system for metallurgical hydrogen applications, characterized in that, It includes a memory, a processor, a hydrogen production scheduling database, and a hydrogen production control interface. The memory stores a computer program, which, when executed by the processor, implements a hydrogen production data scheduling method for metallurgical hydrogen production as described in any one of claims 1 to 9. The hydrogen production scheduling database is used to store a metallurgical hydrogen operating condition profile library, scheduling sample sequences, metallurgical hydrogen demand windows, hydrogen production resource status matrix, window reachability matrix, and hydrogen production scheduling parameters. The hydrogen production control interface is used to send hydrogen production scheduling parameters to the hydrogen production execution unit. The hydrogen production execution unit includes at least one of an electrolysis hydrogen production unit and a hydrogen-containing gas purification unit, and further includes a purification and drying unit, a compression and storage hydrogen unit, and a hydrogen supply valve assembly.

Images