A method, system, equipment, and medium for optimizing the scheduling of by-product coal gas in steel enterprises.

Abstract

This invention belongs to the fields of artificial intelligence and by-product gas technology, specifically disclosing a method, system, equipment, and medium for optimizing the scheduling of by-product gas in steel enterprises. The method includes the following steps: S1, establishing the objective function: minimizing the penalty cost of gas holders and electricity costs under a peak-valley electricity pricing mechanism; S2, constraints: including quality balance constraints, energy balance constraints, and operating condition constraints; S3, setting the penalty factor: based on the peak-valley electricity pricing mechanism, setting the penalty factor during non-peak periods to a higher level than during other periods. This invention employs the aforementioned method, system, equipment, and medium for optimizing the scheduling of by-product gas in steel enterprises, combined with a MINLP model based on a peak-valley electricity pricing mechanism and operational optimization, to achieve a balance between reducing energy costs and ensuring the operational stability of gas holders.

Description

Technical Field

[0001] This invention relates to the fields of artificial intelligence and by-product gas technology, and in particular to a method, system, equipment and medium for optimizing the scheduling of by-product gas in steel enterprises. Background Technology

[0002] With increasingly fierce market competition, the profit margins of steel companies are being continuously squeezed, and energy costs account for a significant proportion of operating costs. Reducing energy costs in the production process through refined energy management has become a common concern in the industry. By-product gas is an important secondary energy source for steel companies, accounting for about 40% of total energy consumption. It is mainly used in plant heating furnaces, with the remainder used for power generation. After the implementation of peak-valley electricity pricing mechanisms, utilizing the cross-cycle adjustment capabilities of self-owned power plants and gas holders to save on purchased electricity costs has become an important means for companies to reduce costs.

[0003] However, traditional steel companies have a relatively crude approach to energy management, relying heavily on experience for gas dispatching and gas holder location control. This leads to inflexible gas dispatching, unreasonable energy consumption, and low energy utilization. In practice, this manifests primarily in the use of fixed penalty factors during both peak-shaving and off-peak periods. During peak-shaving periods, the high penalty factor and objective function weights make it difficult for the model to respond effectively to electricity pricing mechanisms, while during off-peak periods, the low penalty factor leads to unnecessary costs and resource waste, ultimately hindering the release of greater peak-shaving potential.

[0004] To better address these issues, steel companies have been exploring optimal coordination mechanisms between by-product gas dispatching and generator operation in recent years. Various mathematical models applied in this area have provided important insights for gas dispatching and management. By applying mathematical models, a balance between gas and electricity can be achieved, enabling efficient energy utilization.

[0005] In existing technologies, for example, some steel companies introduce an initial efficiency model into their gas dispatching model, then convert different types of energy media in the model into equivalent electrical energy in the same unit to obtain a simplified efficiency model. Next, a multi-objective model is constructed based on the simplified efficiency model, and the TOPSIS method is used to make decisions on the non-dominated solution set on the Pareto front to obtain the optimal dispatching scheme. While this dispatching method can improve the rationality of gas-steam-electricity system dispatching, it does not consider the different dispatching methods during peak and off-peak periods, and is insufficient to cope with peak-valley electricity pricing mechanisms.

[0006] A more advanced technology utilizes artificial intelligence for gas dispatching. It employs a deep deterministic strategy gradient algorithm combined with deep neural networks and Actor-Critic networks to make the dispatching process more intelligent. This ensures that the power generation of generators at different times better matches the electricity demand of the steel industry, reducing the amount of electricity purchased from the grid and maximizing the value of generated electricity, while also taking into account time-of-use pricing. However, even with time-of-use pricing considered, the objective function of this dispatching method still uses a fixed penalty factor, so the revenue from power generation that fails to fully unleash peak-shaving potential can still be improved. Summary of the Invention

[0007] The purpose of this invention is to provide a method, system, equipment and medium for optimizing the scheduling of by-product coal gas in steel enterprises, which combines the MINLP model of peak-valley electricity pricing mechanism with operation optimization to achieve a balance between reducing energy costs and ensuring the stability of gas holder operation.

[0008] To achieve the above objectives, this invention provides a method, system, equipment, and medium for optimizing the scheduling of by-product coal gas in steel enterprises, comprising the following steps: S1. Establishment of the objective function: Minimize the gas storage penalty cost and electricity cost under the peak-valley electricity pricing mechanism; S2. Constraints: These include mass balance constraints, energy balance constraints, and operating condition constraints. S3. Penalty Factor Setting: Based on the peak-valley electricity pricing mechanism, the penalty factor during non-peak periods is set to a higher level than that during other periods.

[0009] Preferably, S1 is as follows: To minimize the gas storage penalty cost and electricity cost under the peak-valley electricity pricing mechanism, the following formula is used: ; in, GPC To incur penalties for gas holders, EPC For electricity costs; Gas holder penalty cost GPC It is the absolute value of the difference between the gas holder position offset and the reference position and the corresponding penalty factor. Pf The sum of the products over all time intervals, where the penalty factor is... Pf The gas holder penalty cost varies with time and the degree of deviation from the reference position. GPC The calculation method is shown in the following formula: ; in, t It is a time period. g It refers to a specific type of gas. G It is a collection of different types of gas; Electricity costs EPCIt is the sum of the difference between total electricity demand and self-generated electricity, multiplied by the electricity price for the current period and the prices for all periods. EPC The calculation method is shown in the following formula: .

[0010] Preferably, in S2, the mass balance constraint is the flow and volume balance of by-product gas and steam in the steel plant system, and the change of the gas holder liquid level precisely matches the difference between the by-product gas production and the consumption of the steel system and the boiler. The volume of by-product gas consumed by the boiler is determined by the number of burners turned on and the unit load, and the gas input is consistent with the burner operating status. The steam produced by the boiler simultaneously meets the direct needs of the steel system and the input needs of the turbine power generation.

[0011] Preferably, in S2, the energy balance constraint is that the energy in the system is converted only between the chemical energy of coal gas, the thermal energy of steam, and the electrical energy, and the conversion process needs to be related to the equipment characteristics and the energy properties of the medium.

[0012] Preferably, for a boiler, the steam volume is matched with the gas energy input, and the steam volume is determined by the total heat released from the combustion of by-product gas, while taking into account the inherent characteristics formed by the boiler's historical operation. For steam turbines, the amount of electricity generated is determined by the energy contained in the steam consumed, the amount of steam used, and the steam turbine's own energy conversion efficiency.

[0013] Preferably, in S2, the operating condition constraints are used to set safe operating boundaries for the gas holder equipment and boiler equipment; For gas holder equipment, the liquid level is limited to not be higher than the maximum value or lower than the minimum value, and the rate of change of liquid level between adjacent time periods is limited; For boiler equipment, the heat load should not exceed the maximum limit or fall below the minimum limit, the consumption of a single type of by-product gas should not exceed the upper limit or fall below the lower limit, and the rate of change of gas consumption in adjacent time periods should be limited.

[0014] Preferably, in S3, peak power generation is performed 1-3 hours before the electricity price changes from low to high, and supply and demand are kept stable during non-peak periods.

[0015] This invention also provides a system for optimizing the dispatch of by-product coal gas in steel enterprises, wherein the method described includes: The model module is constructed based on minimizing the gas holder penalty cost and electricity cost under the peak-valley electricity price mechanism. The objective function is constructed, the constraints are constructed, and the time-varying penalty factor strategy is set. The electricity price changes by 1-3 hours before and after the peak power generation operation, and the supply and demand are kept stable during the non-peak period. The solution module constructs and solves the gas scheduling optimization model based on the objective function, constraints, and penalty factors.

[0016] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the aforementioned method for optimizing the scheduling of by-product coal gas in steel enterprises.

[0017] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned method for optimizing the scheduling of by-product coal gas in steel enterprises.

[0018] Therefore, the present invention employs the above-mentioned method, system, equipment, and medium for optimizing the scheduling of by-product coal gas in steel enterprises, and the beneficial effects are as follows: This invention employs a time-varying penalty factor strategy to fully unleash the volatility potential of gas holders during peak-shaving periods to adapt to peak-valley electricity pricing mechanisms. Simultaneously, during off-peak periods, it stabilizes the gas holder position near a reference level, effectively balancing the dual needs of "utilizing electricity price differences to increase revenue" and "ensuring the gas holder's resilience to price fluctuations." This resolves the core contradiction that traditional fixed penalty factor models cannot balance. Compared to traditional models and manual operation, this model responds more proactively to peak-valley electricity pricing mechanisms, increases the proportion of power generation during peak hours, reduces reliance on purchased electricity, and lowers overall energy costs for enterprises from an energy dispatch perspective, providing a more efficient solution for energy conservation and cost reduction in steel companies.

[0019] During off-peak periods, this invention enables the model to stably control the gas holder positions, preventing significant fluctuations in holder positions from impacting energy supply and demand balance. This ensures the continuity and reliability of energy supply during steel production and reduces production risks that may result from energy instability. The model is built upon the actual operating conditions of equipment in steel enterprises, and the time-varying penalty factor is tailored to actual scheduling needs. Compared to traditional fixed-parameter models, it can more flexibly address differences in scheduling objectives across different time periods, enhancing the model's application value and practicality in real-world production scenarios.

[0020] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0021] Figure 1 This is a flowchart of an embodiment of the present invention: a method, system, equipment, and medium for optimizing the scheduling of by-product coal gas in steel enterprises. Figure 2 This is a cabinet position control level diagram and cabinet position control diagram for different time periods in Embodiment 2 of the present invention, which describes a method, system, equipment and medium for optimizing the scheduling of by-product coal gas in steel enterprises. Figure 3This is a diagram showing the total power generation and load of each boiler before and after optimization in each time period of a method, system, equipment and medium for optimizing the dispatch of by-product coal gas in steel enterprises according to the present invention, in Example 2. Where a is the total power generation of the generator set in each time period under different penalty factor settings, b is the power generation load of boiler #1 in each time period under different penalty factor settings, c is the power generation load of boiler #2 in each time period under different penalty factor settings, d is the power generation load of boiler #3 in each time period under different penalty factor settings, e is the power generation load of boiler #4 in each time period under different penalty factor settings, and f is the power generation load of boiler #5 in each time period under different penalty factor settings. Figure 4 This is a benefit analysis diagram under different penalty factor settings in Implementation Example 2 of the present invention, which describes a method, system, equipment and medium for optimizing the dispatch of by-product coal gas in steel enterprises. In this diagram, a represents the peak-valley power generation ratio under different penalty factor settings, and b represents the total power generation and total power generation revenue of the unit under different penalty factor settings. Figure 5 This is a diagram showing the change in gas holder positions under different penalty factor settings in Embodiment 2 of the present invention, which describes a method, system, equipment, and medium for optimizing the scheduling of by-product gas in steel enterprises. In this diagram, a represents the gas holder positions of blast furnace gas holders at different time periods under different penalty factor settings; b represents the gas holder positions of coke oven gas holders at different time periods under different penalty factor settings; c represents the gas holder positions of converter gas holders at different time periods under different penalty factor settings; and d represents the sum of the standard deviations of the volume over the entire cycle, the peak-shaving period, and the non-peak-shaving period under different penalty factor settings. Detailed Implementation

[0022] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0023] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0024] Example 1 like Figure 1 As shown, a method, system, equipment, and medium for optimizing the scheduling of by-product coal gas in steel enterprises includes the following steps: S1. Establishment of the objective function: Minimize the gas storage penalty cost and electricity cost under the peak-valley electricity pricing mechanism, as shown in the following formula: ; in, GPC To incur penalties for gas holders, EPC For electricity costs.

[0025] Gas holder penalty cost ( GPC() is the absolute value of the difference between the gas holder position offset and the reference position, and the corresponding penalty factor. Pf The sum of the product over all time intervals. Where the penalty factor is... Pf It varies with time and the degree of gas holder offset from the reference position. GPC The calculation method is shown in the following formula: ; in, t It is a time period. t =1 represents the first time segment. g It refers to a specific type of gas, such as blast furnace gas, converter gas, or coke oven gas. G It is a set of gas types G = {blast furnace gas, coke oven gas, converter gas}.

[0026] Electricity costs ( EPC The sum of the difference between total electricity demand and self-generated electricity, multiplied by the electricity price for the current period, over all periods. EPC The calculation method is shown in the following formula: .

[0027] S2. Constraints: These include mass balance constraints, energy balance constraints, and operating condition constraints.

[0028] Quality balance constraints refer to the flow and volume balance of by-product gas (BFG, COG, LDG) and steam within the steel plant system, with the core objective of preventing material accumulation or shortage. For gas holders: changes in the gas holder level must precisely match the difference between "by-product gas production - steel system consumption - boiler consumption" to ensure dynamic adjustment of the level according to production and consumption, preventing abnormal levels caused by imbalances in production and consumption. For boilers: The volume of by-product gas consumed by the boiler is determined by the number of burners turned on and the unit load, ensuring that the gas input is consistent with the burner operating status; For steam: The steam produced by the boiler must simultaneously meet the direct needs of the steel system and the input needs of turbine power generation, so as to achieve a seamless and non-redundant process in the entire process of steam "production-use-conversion".

[0029] Energy balance constraints refer to the conservation of energy conversion within a system. Energy is only converted between different forms (chemical energy of coal gas → thermal energy of steam → electrical energy), and cannot be created or destroyed out of thin air. Furthermore, the conversion process must be linked to equipment characteristics and the energy properties of the medium. For boilers, the steam production capacity (i.e., steam volume) is determined by the total heat released from the combustion of by-product coal gas (which needs to be considered in conjunction with the lower calorific value of various types of coal gas to reflect its energy density). Simultaneously, the inherent characteristics formed during the boiler's historical operation (such as the heat-steam conversion patterns of different boilers) should be taken into account to ensure that the steam volume matches the coal gas energy input and avoid energy conversion imbalance. For steam turbines, power generation is determined by the energy contained in the consumed steam (expressed as steam enthalpy, reflecting the steam energy level), the steam consumption, and the turbine's own energy conversion efficiency. The more energy in the steam and the higher the conversion efficiency, the more the power generation conforms to the logic of energy conservation; there is no arbitrary increase or decrease in energy.

[0030] Operating condition constraints refer to setting "safe operating boundaries" for two key types of equipment: gas holders and boilers. These boundaries prevent equipment failures caused by operating beyond its limits, ensuring system stability. For gas holders, the constraints limit the liquid level to a maximum value (to prevent overflow and overpressure) or a minimum value (to prevent cavitation and inability to meet emergency supply). They also limit the rate of liquid level change between adjacent time periods to prevent sudden rises and falls in liquid level that could lead to structural instability or fluctuations in gas supply. For boilers, operating boundaries are defined in three dimensions: first, the heat load (i.e., the boiler's total heat production capacity) must not exceed the maximum limit (to prevent overheating) or fall below the minimum limit (to prevent flameout and unstable operation); second, the consumption of a single type of by-product gas must not exceed the upper limit (to prevent excessive combustion) or fall below the lower limit (to prevent incomplete combustion); and third, the rate of change in gas consumption between adjacent time periods must not be too rapid to prevent sudden load changes from disrupting combustion conditions and affecting steam production stability.

[0031] S3. Penalty Factor Setting: Based on the peak-valley electricity pricing mechanism, the penalty factor is set to a higher level during non-peak-shaving periods to allow the model to better control the power supply position at the middle level during these times. During other periods, the penalty factor is set to a lower level. Based on field operational experience and gas regulation rate limitations, peak-load power generation is generally performed in the two hours before the electricity price transitions from low to high (this time can be appropriately extended or shortened depending on the local conditions of the steel company). During non-peak-shaving periods, the focus is on maintaining stable supply and demand. This ensures that the model is as unaffected as possible by power supply fluctuations during peak-shaving, thus releasing greater peak-shaving potential.

[0032] Example 2 The method described in Example 1 was applied to a steel plant in a certain region of a certain province.

[0033] like Figure 2As shown, given the time-of-use electricity prices in this region and the optimal control level for gas holders based on these prices, the best control strategy should be to adjust the power level of generator sets to save on electricity costs before and after sudden price fluctuations (green background). During this time, to accommodate the power adjustments of the generator sets, the gas holder position will fluctuate significantly within the permissible operating range. Conversely, during the middle of a long period of stable electricity prices (red background), the gas holder position should be controlled near the reference level to better cope with potential emergencies.

[0034] Based on the peak-valley electricity pricing mechanism, the penalty factor is set to a higher level for periods 1-52, 126-145, 173-198, and 236-240, so that the model can better control the grid position at the middle grid position during these periods. In other periods, the penalty factor is set to a lower level so that the model is not limited by grid position fluctuations as much as possible during peak shaving, thereby releasing greater peak shaving potential.

[0035] The steel plant has a total of 5 power generation units and one blast furnace gas holder, one converter gas holder, and one coke oven gas holder. Detailed equipment parameters are shown in Table 1. The penalty factor settings and results are shown in Table 2.

[0036] Table 1 Key parameters of relevant equipment in steel plants

[0037] Table 2. Penalty Factor Setting Levels and Results

[0038] The calculation results of the single penalty factor show that when the penalty factor is set to 0.002, the model can take into account both peak shaving capacity and maintain good cabinet position control results. When the penalty factor is set to 0.01 during non-peak shaving periods and 0.00005 during peak periods, the model can significantly improve power generation efficiency while maintaining a similar level of SSDV during non-peak shaving periods as when the single penalty factor is set to 0.002.

[0039] The total power generation and load of each boiler in each time period before and after optimization are as follows: Figure 3 As shown in the diagram, the red line represents the result of manual operation, while the green and blue lines represent the optimization calculation results under two different penalty factor settings, respectively. It can be seen that, compared to the fixed penalty factor model, the time-varying penalty factor model responds more actively to the peak-valley electricity pricing mechanism during peak shaving periods, resulting in a higher proportion of power generation during peak hours and thus generating more revenue.

[0040] like Figure 4 As shown, Figure 4 In the figure, 'a' represents the ratio of power generation during peak and off-peak hours under different penalty factor settings. Figure 4Figure 'b' illustrates the total power generation and total revenue of the generating units under different penalty factor settings. Under all penalty factor settings, compared to manual operation, the model significantly increased total power generation and the proportion of power generation during peak hours, resulting in a substantial increase in revenue. With a single penalty factor setting, increasing the penalty factor significantly reduced the proportion of power generation during PPP periods, leading to a decrease in revenue. However, with a variable penalty factor setting, the peak power generation ratio was at the same level as with the minimum single penalty factor setting, and the revenue was also at a comparable level.

[0041] like Figure 5 The figure illustrates the changes in grid positions under different penalty factor settings. The graph shows that regardless of the penalty factor setting, the model effectively controls the grid positions closer to the intermediate level. With a single penalty factor, reducing the penalty factor amplifies grid position fluctuations during peak-shaving periods, resulting in higher power generation revenue. However, this also significantly increases the sum of standard deviations (SSDV) of the grid positions across all periods, including off-peak periods. Conversely, reducing the penalty factor during peak-shaving periods and increasing it during off-peak periods allows for more precise grid position control while releasing greater peak-shaving potential. When the penalty factor for off-peak periods is increased to 0.01 and the penalty factor for peak-shaving periods is decreased to 0.00005, after one peak-shaving cycle, the grid positions recover to the reference level more quickly and are maintained at that level more effectively. This makes the SSDV during off-peak periods comparable to that with a single penalty factor of 0.002, while only amplifying the SSDV during peak-shaving periods.

[0042] In summary, based on the model's optimal balance between peak shaving and grid control when the fixed penalty factor is set to 0.002, this study investigates the impact of varying penalty factor settings. It concludes that setting the penalty factor to 0.00005 for peak shaving periods and 0.01 for non-peak shaving periods allows the model to fully unleash peak shaving potential while also ensuring grid control during non-peak periods. Compared to the scheduling model with a fixed penalty factor, this model better distinguishes between peak and non-peak periods, achieving better peak shaving potential release during peak periods while maintaining grid control levels close to the comparative model during non-peak periods. Due to the more efficient release of peak shaving potential, the model increases peak-period power generation from 40.34% in the comparative model to 42.59%, resulting in an additional 4290.86 CNY in power generation revenue over a 24-hour scheduling cycle, potentially saving enterprises over 1.2 million CNY annually in purchased electricity costs.

[0043] Therefore, this invention adopts the above-mentioned method, system, equipment and medium for optimizing the scheduling of by-product coal gas in steel enterprises, combined with the MINLP model of peak-valley electricity pricing mechanism and operation optimization, to achieve a balance between reducing energy costs and ensuring the stability of gas holder operation.

[0044] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for optimizing the scheduling of by-product coal gas in steel enterprises, characterized in that, Includes the following steps: S1. Establishment of the objective function: Minimize the gas storage penalty cost and electricity cost under the peak-valley electricity pricing mechanism; Input into the MINLP model; Penalty factor setting: Based on the peak-valley electricity pricing mechanism, the penalty factor during non-peak periods is set to a higher level than that during peak periods; S2. Constraints: These include mass balance constraints, energy balance constraints, and operating condition constraints. S1 specifically refers to: To minimize the gas storage penalty cost and electricity cost under the peak-valley electricity pricing mechanism, the following formula is used: ； in, GPC To incur penalties for gas holders, EPC For electricity costs; Gas holder penalty cost GPC It is the absolute value of the difference between the gas holder position offset and the reference position and the corresponding penalty factor. Pf The sum of the products over all time intervals, where the penalty factor is... Pf The gas holder penalty cost varies with time and the degree of deviation from the reference position. GPC The calculation method is shown in the following formula: ； in, t It is a time period. g It refers to a specific type of gas. G It is a collection of different types of gas; Electricity costs EPC It is the sum of the difference between total electricity demand and self-generated electricity, multiplied by the electricity price for the current period and the prices for all periods. EPC The calculation method is shown in the following formula: 。 2. The method for optimizing the scheduling of by-product coal gas in steel enterprises according to claim 1, characterized in that, In S2, the mass balance constraint is the flow and volume balance of by-product gas and steam in the steel plant system, and the change of gas holder liquid level matches the difference between the by-product gas production and the consumption of the steel system and boiler. The volume of by-product gas consumed by the boiler is determined by the number of burners turned on and the unit load. The steam produced by the boiler simultaneously meets the direct needs of the steel system and the input needs of the turbine power generation.

3. The method for optimizing the scheduling of by-product coal gas in steel enterprises according to claim 1, characterized in that, In S2, the energy balance constraint means that the energy in the system is converted only between the chemical energy of coal gas, the thermal energy of steam, and the electrical energy. The conversion process must be related to the equipment characteristics and the energy properties of the medium.

4. The method for optimizing the scheduling of by-product coal gas in steel enterprises according to claim 3, characterized in that: For boilers, the steam volume is matched with the gas energy input. The steam volume is determined by the total heat released from the combustion of by-product gas, while also taking into account the inherent characteristics formed by the boiler's historical operation. For steam turbines, the amount of electricity generated is determined by the energy contained in the steam consumed, the amount of steam used, and the steam turbine's own energy conversion efficiency.

5. The method for optimizing the scheduling of by-product coal gas in steel enterprises according to claim 1, characterized in that, In S2, operating condition constraints are used to set safe operating boundaries for gas holder equipment and boiler equipment; For gas holder equipment, the liquid level is limited to not be higher than the maximum value or lower than the minimum value, and the rate of change of liquid level between adjacent time periods is limited; For boiler equipment, the heat load should not exceed the maximum limit or fall below the minimum limit, the consumption of a single type of by-product gas should not exceed the upper limit or fall below the lower limit, and the rate of change of gas consumption in adjacent time periods should be limited.

6. The method for optimizing the scheduling of by-product coal gas in steel enterprises according to claim 1, characterized in that: In S1, peak power generation is carried out 1-3 hours before the electricity price turns from low to high, and supply and demand are kept stable during non-peak periods.

7. A system for optimizing the dispatch of by-product coal gas in steel enterprises, using the method described in any one of claims 1 to 6, characterized in that, include: A model module is constructed to construct an objective function based on minimizing the gas holder penalty cost and electricity cost under the peak-valley electricity price mechanism. Constraints are constructed, and a time-varying penalty factor strategy is set. Peak-load power generation is performed 1-3 hours before the electricity price change, and supply and demand are maintained in a stable manner during non-peak-shaving periods. The solution module constructs and solves the gas scheduling optimization model based on the objective function, constraints, and penalty factors.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the method for optimizing the scheduling of by-product coal gas in steel enterprises as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements a method for optimizing the scheduling of by-product coal gas in steel enterprises as described in any one of claims 1 to 6.

Images