Polycrystalline silicon reduction production scheduling method, device, equipment and storage medium

Abstract

The invention relates to a polycrystalline silicon reduction production scheduling method and device, equipment and a storage medium. The method comprises the following steps: obtaining an effective production time period of each production shift in a single day and a reduction furnace to be started; determining the possible furnace starting time of each to-be-started reduction furnace of each production shift; generating a plurality of furnace starting moment combinations of each production shift; determining a plurality of furnace starting plans of a single day; determining the planned consumption of each production process element in each furnace starting plan; and comparing each planned consumption in each furnace starting plan with the reference consumption of the corresponding production process element, and screening out a single-day scheduling scheme for polycrystalline silicon reduction production from the plurality of furnace starting plans according to a comparison result. By adopting the method, the single-day scheduling scheme of polycrystalline silicon reduction production can be automatically formulated, the energy consumption in the polycrystalline silicon reduction production process is effectively reduced, and the high efficiency and safety of production and the maximum benefit are ensured.

Description

Technical Field

[0001] The present application relates to the technical field of polysilicon, and in particular to a polysilicon reduction production scheduling method, a polysilicon reduction production scheduling device, a computer device, a computer-readable storage medium, and a computer program product. Background Art

[0002] The modified Siemens process is one of the main methods for producing high-purity polysilicon in industry. This method generates high-purity polysilicon by reacting trichlorosilane (TCS, SiHCl3) or silicon tetrachloride (STC, SiCl4) with hydrogen (H2) at high temperature, but its energy consumption is high, especially in terms of electricity consumption, which accounts for most of the total production cost. With the continuous rise in energy prices and increasingly stringent environmental protection requirements, reducing energy consumption and improving energy efficiency have become urgent issues for polysilicon production companies.

[0003] In the polysilicon production process, reasonable production scheduling can significantly improve production efficiency and reduce energy consumption. However, traditional production scheduling mainly relies on manual experience, which makes it difficult to fully consider all influencing factors, resulting in subjectivity and uncertainty in the decision-making process, and lack of systematicity and scientificity. Summary of the invention

[0004] Based on this, it is necessary to provide a polysilicon reduction production scheduling method, device, computer equipment, computer readable storage medium and computer program product to address the above technical problems.

[0005] In a first aspect, the present application provides a polysilicon reduction production scheduling method, the method comprising:

[0006] Obtain the effective production time period and reduction furnaces to be started for each production shift on a single day;

[0007] Determine a possible start-up time of each of the reduction furnaces to be started in each production shift; the possible start-up time is within the effective production time period of the corresponding production shift;

[0008] Generate multiple furnace start time combinations for each production shift; each of the furnace start time combinations is composed of possible furnace start times of each of the reduction furnaces to be started in the production shift;

[0009] Determine multiple furnace start-up plans for a single day; the multiple furnace start-up plans are obtained by performing a Cartesian product operation on a set consisting of multiple furnace start-up time combinations for each production shift;

[0010] Determine the planned consumption of each production process element in each furnace start-up plan;

[0011] Each of the planned consumptions in each furnace start-up plan is compared with the benchmark consumption of the corresponding production process element, and a single-day scheduling plan for polysilicon reduction production is screened out from the multiple furnace start-up plans based on the comparison results.

[0012] In one embodiment, the production process elements include at least materials, tail gas recovery and steam balance.

[0013] In one embodiment, the step of comparing each of the planned consumption in each furnace start-up plan with the benchmark consumption of the corresponding production process element, and selecting a single-day scheduling plan for polysilicon reduction production from the multiple furnace start-up plans according to the comparison result, includes:

[0014] Obtain the maximum benchmark consumption and minimum benchmark consumption of each production process element in a single day respectively;

[0015] Compare each of the planned consumption in each furnace start-up plan with the maximum baseline consumption and the minimum baseline consumption of the corresponding production process element;

[0016] If each of the planned consumptions in the furnace startup plan is between the corresponding maximum baseline consumption and minimum baseline consumption, the furnace startup plan will be used as a single-day scheduling plan for polysilicon reduction production.

[0017] In one embodiment, when there are multiple single-day scheduling plans, the method further includes:

[0018] Calculate the electricity cost of each of the reduction furnaces to be started at different times for a corresponding production cycle; the different times are all within a single day;

[0019] According to the electricity cost of each reduction furnace to be started, the optimal scheduling plan is selected from the single-day scheduling plans.

[0020] In one embodiment, the step of selecting the optimal scheduling plan from the single-day scheduling plans according to the electricity cost of each reduction furnace to be started includes:

[0021] Calculate the electricity cost of each of the single-day scheduling plans according to the electricity cost of each of the reduction furnaces to be started;

[0022] The single-day dispatch plan with the lowest electricity cost is taken as the optimal dispatch plan.

[0023] In one embodiment, the method further comprises:

[0024] The optimal scheduling scheme is fine-tuned with the goal of maximizing the proximity between the electricity cost of the optimal scheduling scheme and the target cost.

[0025] In a second aspect, the present application further provides a polysilicon reduction production scheduling device, the device comprising:

[0026] An acquisition module is used to obtain the effective production time period and the reduction furnace to be started for each production shift on a single day;

[0027] A generating module, used to determine the possible start-up time of each of the reduction furnaces to be started in each production shift; the possible start-up time is within the effective production time period of the corresponding production shift; generate multiple start-up time combinations for each production shift; each start-up time combination is composed of the possible start-up time of each of the reduction furnaces to be started in the production shift;

[0028] A calculation module is used to determine multiple furnace start-up plans for a single day; the multiple furnace start-up plans are obtained by performing a Cartesian product operation on a set consisting of multiple furnace start-up time combinations of each production shift;

[0029] A determination module is used to determine the planned consumption of each production process element in each furnace start-up plan;

[0030] The screening module is used to compare each of the planned consumptions in each furnace start-up plan with the benchmark consumptions of the corresponding production process elements, and screen out a single-day scheduling plan for polysilicon reduction production from the multiple furnace start-up plans based on the comparison results.

[0031] In a third aspect, an embodiment of the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of a polysilicon reduction production scheduling method when executing the computer program.

[0032] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of a polysilicon reduction production scheduling method are implemented.

[0033] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of a polysilicon reduction production scheduling method.

[0034] One of the above technical solutions has the following advantages or beneficial effects: by comprehensively considering the workshop shift work system, the effective production time of each production shift, the planned demand for each production process element and the actual supply and consumption, a single-day scheduling plan for polysilicon reduction production is automatically formulated. The above solution can effectively reduce the energy consumption of the polysilicon reduction production process, ensure efficient and safe production and maximize the overall operational benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A schematic diagram of a process flow of a polysilicon reduction production scheduling method in one embodiment;

[0036] Figure 2 A schematic flow chart of a polysilicon reduction production scheduling method in another embodiment;

[0037] Figure 3 It is a structural block diagram of a polysilicon reduction production scheduling device in one embodiment;

[0038] Figure 4 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0040] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0041] In one embodiment, Figure 1 As shown, a polysilicon reduction production scheduling method is provided, comprising the following steps:

[0042] S102, obtaining the effective production time period and the reduction furnace to be started for each production shift on a single day.

[0043] Among them, the production shifts can be divided according to the workshop shift work system, for example, two shifts per day, three shifts per day (such as day shift, mid-shift, night shift), four shifts per day, etc. The workshop shift work system can be set according to actual conditions, for example, including but not limited to shift allocation, shift working hours, shift workload arrangement plan, shift handover work arrangement, etc.

[0044] The effective production time period can be understood as the effective time period that each production shift can use to restore production. For example, it is necessary to exclude the time period or time of shift change in the specified shift period to obtain the effective production time period for each production shift.

[0045] The reduction furnace to be started for each production shift can be determined based on planning documents such as the production process baseline or the overall production plan. The production process baseline (referred to as the process baseline in this article) can be understood as a method of determining the benchmark values ​​of various key parameters in the process flow to ensure product quality and production efficiency. By establishing a process baseline, the company can clarify the requirements of each process parameter, reduce product variation, improve consistency, and ensure that the product meets customer requirements. The specific implementation process of this step can be to determine the production line of each production shift through the process baseline, and then determine the reduction furnace that needs to be started and the specific number of reduction furnaces through each production line.

[0046] S104, determining a possible start-up time of each reduction furnace to be started in each production shift; the possible start-up time is within the effective production time period of the corresponding production shift.

[0047] The number of reduction furnaces to be started in each production shift can be set according to the actual production line and production plan, and each reduction furnace to be started in each production shift has a corresponding possible start-up time.

[0048] The possible furnace start time can be understood as any time within the effective production time period of the corresponding production shift, that is, each reduction furnace to be started in the production shift can theoretically be started at any time within the effective production time period of the shift. In order to improve the computer data processing rate, the time between two possible furnace start times can be set to a fixed unit time length, such as 1 minute, 3 minutes... 1 hour, etc.

[0049] S106, generating a plurality of furnace start time combinations for each production shift; each furnace start time combination is composed of possible furnace start times of each reduction furnace to be started in the production shift.

[0050] In this step, for example, the production shifts are morning, mid-day and evening, and the number of reduction furnaces to be started in the three shifts is 2 (furnace 1, furnace 2), 1 (furnace 3) and 3 (furnace 4, furnace 5, furnace 6) respectively. The possible start-up time of furnace 1 determined by S104 is t1, t2..., and the possible start-up time of furnace 2 is T1, T2..., and so on. The start-up time combination of the morning shift can be determined as t1+T1, or t1+T2..., or t2+T1, or t2+T2..., thereby forming multiple start-up time combinations of the morning shift, each of which is composed of the possible start-up times of furnace 1 and furnace 2. Multiple start-up time combinations of the middle shift and the evening shift are obtained in the same way.

[0051] S108, determining multiple furnace start-up plans for a single day; the multiple furnace start-up plans are obtained by performing a Cartesian product operation on a set consisting of multiple furnace start-up time combinations of each production shift.

[0052] The above example is continued here. After the multiple furnace start time combinations of the morning, afternoon and evening shifts are determined, the multiple furnace start time of the morning shift can be taken as a set, the multiple furnace start time of the afternoon shift can be taken as a set, and the multiple furnace start time of the evening shift can be taken as a set, for a total of three sets. Then, Cartesian product operation is performed on the three sets to obtain multiple furnace start plans for a single day.

[0053] An example of Cartesian product operation on three sets is as follows: Assume A = {1, 2}, B = {2, 3}, C = {5}, then: B×C = {(2, 5), (3, 5)}, A×(B×C) = {(1, (2, 5)), (1, (3, 5)), (2, (2, 5)), (2, (3, 5))}, so we can see the form of the result obtained after performing Cartesian product operation on the above three sets.

[0054] S110, determining the planned consumption of each production process element in each furnace start-up plan.

[0055] Production process elements can be understood as things that can be consumed in the production process, such as materials, production lines, equipment, etc. Specifically, production process elements can be, but are not limited to, total TCS amount, total H2 amount, TCS fluctuation, CDI tail gas recovery, production line equipment, equipment safety, production line stability, etc.

[0056] The production of polysilicon by the modified Siemens process is a highly systematic project involving multiple key links, from upstream material supply to midstream reduction reaction, and then to downstream tail gas recovery. Each process is closely linked to ensure the smooth operation of the entire production system. This complex production process not only requires precise control and coordination, but is also subject to strict constraints of multiple production processes. Due to the unique process design and operation mode of the reduction furnace, the entire reduction process is cyclical. This step can be based on the process baseline of the production line to obtain the periodic cumulative consumption of materials, tail gas pressure, steam balance, etc. in each unit time (such as per hour) after the furnace is started at different times, so as to determine the planned consumption of each production process element in each furnace start-up plan.

[0057] S112, comparing each planned consumption in each furnace start-up plan with the benchmark consumption of the corresponding production process element, and selecting a single-day scheduling plan for polysilicon reduction production from multiple furnace start-up plans based on the comparison results.

[0058] According to the supply and consumption of each production process element in actual production, such as the supply of upstream materials in the reduction workshop, the load of the downstream tail gas recovery system, and the balance of auxiliary systems such as steam, etc., the benchmark consumption of each production process element can be generated. The benchmark consumption can be set to multiple benchmark levels as needed, and of course, upper and lower limits can also be set, that is, the maximum benchmark consumption and the minimum benchmark consumption. By setting the benchmark consumption of each production process element, each start-up plan can be constrained to obtain a start-up plan that better meets the actual production needs.

[0059] In the above embodiments of the present application, the execution subject can be a terminal or a server, and of course, it can also be selected and changed according to the actual situation. In the polysilicon reduction production scheduling method of the above embodiments, by comprehensively considering the shift work system of the workshop, the effective production time of each production shift, the planned demand of each production process element and the actual supply and consumption, a single-day scheduling plan for polysilicon reduction production is automatically formulated. The above plan can effectively reduce the energy consumption of the polysilicon reduction production process, ensure the efficient and safe production, and maximize the overall operational benefits.

[0060] In at least one embodiment, the production process elements include at least materials, tail gas recovery and steam balance.

[0061] In at least one embodiment, S112 may specifically include: obtaining the maximum baseline consumption and the minimum baseline consumption of each production process element for a single day; comparing each planned consumption in each furnace start-up plan with the maximum baseline consumption and the minimum baseline consumption of the corresponding production process element; if each planned consumption in the furnace start-up plan is between the corresponding maximum baseline consumption and the minimum baseline consumption, then the furnace start-up plan is used as a daily scheduling plan for polysilicon reduction production.

[0062] Each planned consumption, maximum benchmark consumption and minimum benchmark consumption in each of the above-mentioned furnace start-up plans can be the quantities at different times during a single day. The planned consumption, maximum benchmark consumption and minimum benchmark consumption at the same time will be compared. If the planned consumption is between the maximum benchmark consumption and the minimum benchmark consumption, it will be retained, otherwise it will be eliminated. Finally, the furnace start-up plan in which each planned consumption is between the corresponding maximum benchmark consumption and minimum benchmark consumption is statistically calculated as a single-day scheduling plan for polysilicon reduction production.

[0063] In at least one embodiment, when there are multiple daily scheduling plans, the multiple daily scheduling plans will be further optimized; specifically, the polysilicon reduction production scheduling method may also include: separately calculating the electricity cost of each reduction furnace to be started and running a corresponding production cycle at different times; the different times are all within a single day time period; according to the electricity cost of each reduction furnace to be started at different times, the optimal scheduling plan is selected from each daily scheduling plan.

[0064] Among them, the production cycle of each reduction furnace to be started can be determined based on the optimal theoretical operating time of each reduction furnace to be started in the process baseline; further, the optimal theoretical operating time of the reduction furnace to be started can be used as its production cycle.

[0065] The amount of electricity used by each reduction furnace to be started in a corresponding production cycle is a fixed value. When the furnace is started at different unit times, the electricity charges (i.e., electricity costs) are different due to different peak, valley, and flat electricity prices at different times of the day. Based on the theoretical electricity consumption, real-time peak, valley, and flat electricity prices, and the division of daily peak, valley, and flat periods, the electricity cost of each reduction furnace to be started at different times in a corresponding production cycle can be calculated (i.e., the product of the sum of the peak price*peak duration ratio, valley price*valley duration ratio, and flat price*flat duration ratio and the total electricity consumption, where the sum of the peak duration ratio, valley duration ratio, and flat duration ratio is 1, and the peak duration ratio is the ratio of the peak duration to the corresponding production cycle duration, and the same applies to the valley duration ratio and flat duration ratio). Of course, the electricity consumption stage can also be further subdivided into peak, valley, flat, peak, and deep valley according to actual conditions, and the electricity price for each stage is different. In the polysilicon reduction production scheduling method of the above-mentioned embodiment, by comprehensively considering the national peak-valley and flat time-sharing electricity price policy and the refined analysis of the power consumption of the single furnace start-up and shutdown plan, under the premise that the total electricity consumption remains unchanged, the peak-valley and flat structure ratio of the total electricity consumption is optimized to significantly reduce the electricity cost.

[0066] According to the electricity cost of each reduction furnace to be started at different times, the optimal scheduling plan can be selected from various daily scheduling plans based on relevant cost constraints.

[0067] In at least one embodiment, the above-mentioned step of selecting the optimal scheduling plan from each daily scheduling plan based on the electricity cost of each reduction furnace to be started at different times may specifically include: calculating the electricity cost of each daily scheduling plan based on the electricity cost of each reduction furnace to be started at different times; and taking the daily scheduling plan with the lowest electricity cost as the optimal scheduling plan.

[0068] Each daily scheduling plan includes the start-up time of each furnace in each production shift. Knowing the electricity cost of each reduction furnace to be started at different times, the electricity cost of each furnace in each production shift of each daily scheduling plan can be calculated. The theoretical electricity cost of each daily scheduling plan can be generated by adding up the electricity costs.

[0069] In at least one embodiment, the polysilicon reduction production scheduling method further includes: fine-tuning the optimal scheduling scheme with the goal of maximizing the proximity between the electricity cost of the optimal scheduling scheme and the target cost.

[0070] The target cost can be understood as a cost KPI. By specifying the target cost, the goal and direction of scheduling optimization can be clarified. The electricity cost of the optimal scheduling solution is closest to the target cost, which can be understood as the absolute value of the difference between the current electricity cost of the optimal scheduling solution and the target cost is the smallest.

[0071] Specifically, based on big data algorithms, etc., with the goal of maximizing the proximity between the electricity cost of the optimal scheduling plan and the target cost, the start-up time of each furnace in each production shift in the optimal scheduling plan can be fine-tuned. For example, the original time step of each start-up time in the optimal scheduling plan can be shortened to make the cost calculation of the optimal scheduling plan more refined, and then a scheduling plan with lower cost and closer to the target cost can be determined, thereby achieving fine-tuning and iterative optimization of the scheduling plan.

[0072] Of course, fine-tuning can also be done through other data algorithms, model building methods and existing technical means. As long as the goal is to maximize the proximity between the electricity cost of the optimal scheduling plan and the target cost, it should be deemed to fall within the scope of protection of this application.

[0073] In other embodiments, Figure 2 As shown, the polysilicon reduction production scheduling method proposed in this application may also include:

[0074] S202. Detailed analysis of electricity consumption for single furnace start and shutdown plans.

[0075] Specifically, based on the combination of process baseline data and the national peak, valley and flat time-of-day electricity price policy, the two can be correlated and analyzed through the time dimension, and the proportion of electricity prices in the three time periods (peak, valley and flat) of the electricity consumed by starting the furnace at different times and running a complete batch can be calculated and evaluated in detail to generate a set of results of the optimization interval information of the single furnace production plan.

[0076] (1) First, detailed process baseline data can be collected, which may include the optimal theoretical operating time, the power consumption of a single furnace start-up and shutdown, and the power curve data of the equipment. The collected data is analyzed in detail to determine the power consumption requirements of different production stages; secondly, according to the specific content of the national peak-valley-average time-sharing electricity price policy, the latest peak-valley-average electricity price data is obtained to ensure the timeliness and accuracy of the analysis results.

[0077] (2) Divide a day T into multiple time slices t (e.g., every hour), each time slice corresponds to a specific electricity price period (peak P peak (t), flat P valley (t), Valley P flat (t)), combined with the process baseline data, calculate the electricity cost C(t) of starting and stopping the furnace once in each time slice.

[0078] Based on the optimal theoretical operating time L, calculate the peak-valley-average period proportion P(t) of a single furnace after the furnace is started at each time t and a complete batch is run. The specific formula is as follows:

[0079] P(t)=[P peak (t),P valley (t),P flat (t)];

[0080] Among them, P peak (t)+P valley (t)+P flat (t) = 1;

[0081] According to the peak-valley-flat electricity price ratio P(t) and the unit electricity price C peak , C valley , C flat , calculate the total electricity cost C(t) of starting the furnace at each time t:

[0082] C(t)=E(P peak (t)·C peak +P valley (t)·C valley +P flat (t)·C flat );

[0083] Where E is the electricity required to run a complete batch.

[0084] Through the above software algorithm, under the optimal theoretical operating time, the electricity cost set corresponding to starting the furnace at 0-23 moments and completing a complete production batch is calculated.

[0085] In addition, you can also select the first n moments {t1, t2, t3, …t n}, generate the single furnace production plan optimization interval information result set T opt :

[0086] T opt ={t1,t2,t2,…t n}

[0087] {t1,t2,t3…t n}∈arg min{C(t)};

[0088] Based on the above calculation results, the furnace start-up period with the lowest electricity cost is selected and defined as the optimal furnace start-up interval. This optimal furnace start-up interval can provide a scientific basis for production scheduling and fine-tuning of subsequent optimal scheduling plans.

[0089] S204. Optimization analysis of production plan under shift and multiple constraints.

[0090] Specifically, this step obtains an analysis and calculation time set that fits the application object by analyzing and restoring the production line or workshop and a series of algorithm operations. The actual production situation of the production line or workshop is restored, including but not limited to shift allocation, shift working hours, shift workload arrangement plan, shift handover work arrangement, etc.; on the basis of obtaining an analysis and calculation time set that fits the actual application, the algorithm of this application further combines the upstream and downstream production system constraints in actual production (including but not limited to total TCS, total H2, TCS fluctuations, CDI tail gas recovery system constraints, production line equipment constraints, equipment safety constraints, production line stability constraints, etc.), and through a series of algorithm operations, obtains a preliminary production overall plan (the number of furnace starts per shift and per production line) that meets the production arrangement of the application site (the plan is different for different application sites).

[0091] A specific implementation is as follows:

[0092] (1) According to the four-shift three-shift work system adopted by the reduction workshop, the production time is divided into three shifts: day shift, mid-shift and night shift. According to the actual work shift division, the production shift work set includes: day shift work time set DayWorkUn, mid-shift work time set MidWorkUn, and night shift work time set Nig tWorkUn. Taking the day shift as an example: DayWorkUn={DayWorkUn1, DayWorkUn2, DayWorkUn3,…, DayWorkUn n};

[0093] (2) According to the actual work requirements of the workshop team, at the end of each shift, a certain handover time is set for information transmission and equipment inspection between shifts. Some time periods cannot be executed according to the production plan, that is, a shift work plan restriction time set is generated, including the day shift work plan restriction time set DayWorkLimitUn, the mid-shift work plan restriction time set MidWorkLimitUn, and the night shift work restriction set NightWorkLimitUn. Take the day shift as an example:

[0094]

[0095] Through set operations, we can obtain the effective execution time set of the work plan for each shift throughout the day that meets the actual production on site, namely:

[0096]

[0097] According to the above step (1), the effective time set WorkTimUn for executing the work plan of each shift throughout the day that meets the actual production on site is obtained:

[0098]

[0099] WorkTimUn=[TimUn1,TimUn2,TimUn3];

[0100] Among them, TimUn1 is the actual effective working time set of the day shift, TimUn2 is the actual effective working time set of the mid-shift, and TimUn3 is the actual effective working time set of the night shift.

[0101] According to the actual daily work plan determined by the workshop production (the number of furnace starts per day) LoadAll, set the basic matrix LoadBase for the planned workload of each shift:

[0102] LoadBase = [A, A, A];

[0103] A = LoadAll / BatchNum;

[0104] Where A is a positive integer, BatchNum is the actual number of production shifts, usually 2 or 3.

[0105] Based on the above calculations, the effective working time and workload arrangement of the shifts in actual production are used to generate the production plan workload optimization matrix LoadAdj:

[0106] LoadAdj = [A1, A2, A3];

[0107] Among them, A1, A2, A3 are integers, including positive numbers, negative numbers or zero.

[0108] Finally, through matrix operations, the final workload scheduling matrix LoadRef for each shift is obtained:

[0109] LoadRef = LoadBase + LoadAdj;

[0110] That is: LoadRef=[A+A1,A+A2,A+A3];

[0111] (3) The above algorithm combines the actual production situation of the workshop, obtains the effective working time of the shift that meets the production execution plan through set operations, and obtains the production plan workload optimization matrix based on the difference in actual effective working time. Based on this matrix, the basic matrix that meets the shift production plan workload is optimized, so that the optimized shift production plan workload is more in line with the actual production situation. It not only meets the needs of the overall production plan, but also balances the workload problem caused by the difference in shift working time, thereby improving production efficiency and resource utilization.

[0112] S206. Calculation and analysis of the overall power consumption of multi-production line production combinations based on the optimization interval under the production plan.

[0113] Specifically, the single-furnace production plan optimization interval information result set generated by the algorithm of this application can be used to combine the determined number of furnace starts for each shift (day shift, mid-shift, night shift) and each production line and the preferred furnace start time after excluding the shift handover time, and calculate the furnace start time combination between each production line in each shift. Finally, the overall combination of multi-production line production of all shifts is comprehensively considered through the big data algorithm to generate a multi-production line production overall scheduling plan set that meets the production plan.

[0114] (1) Based on S204, the optimized workload scheduling matrix LoadRef of each shift and the effective working time WorkTimUn of the shift that meets the production execution plan are obtained, and the overall power consumption set of the multi-production line production combination based on the optimization interval is generated by using the big data parallel algorithm and calculated and analyzed. Specifically:

[0115] The obtained optimized workload scheduling matrix for each shift is LoadRef = [A+A1, A+A2, A+A3];

[0116] Available

[0117] Among them, A′1 is the day shift workload optimization arrangement plan, A′2 is the mid-shift workload optimization arrangement plan, and A′3 is the night shift workload optimization arrangement plan.

[0118] Further:

[0119] A. Based on the obtained set of actual effective working time of the day shift TimUn1, the set of all possible combinations of A′1 arbitrary optional furnace start-up times can be expressed as:

[0120] TimUn1={TimUn 11 ,TimUn 12 ,TimUn 13 ,…,TimUn 1n};

[0121]

[0122] in represents the set of all subsets of TimUn1 of size A′1, and |S| represents the number of elements in the set S.

[0123] B. For each A1' combination All its permutations can be expressed as:

[0124]

[0125] C. Combining the above two steps, it can be expressed as:

[0126]

[0127] in Represents all possible results of selecting A′1 elements from TimUn1 and permuting them.

[0128] (2) At this point, all possible permutations and combinations of the start-up time of each shift are obtained based on the optimal number of furnace starts for each shift and the actual effective working time of each shift. In order to ensure the smooth operation of the multi-line production system and take into account the impact of the differences in the optimal theoretical operating time between different production lines on the peak and valley electricity prices, a global optimization strategy is adopted to comprehensively consider the start-up time between the day shift, the middle shift and the night shift, and generate all possible permutations and combinations. This method aims to maximize the overall operational efficiency and effectively control costs.

[0129] Specifically, taking three production lines and four furnaces as an example, the possible combinations of furnace start times during the day, afternoon and night are as follows:

[0130]

[0131] in Arrange the possible furnace start time sets for day, afternoon and night, and obtain the set of overall furnace start time for multi-line production combination based on the optimization interval through Cartesian product operation. This process can be described as:

[0132]

[0133] Where M is the overall furnace start-up time set of the multi-production line production combination based on the optimization interval.

[0134] S208. Iterative optimization of production schedule under the constraints of multiple production processes (materials, exhaust gas, steam, etc.).

[0135] Specifically, based on the above-mentioned multi-line production overall scheduling plan set that meets the production plan, the actual scheduling plan represented by each combination of information in the set can be optimized at multiple levels (including but not limited to hierarchical iterative optimization based on material TCS, hydrogen, tail gas pressure, etc.), and finally a scheduling strategy that meets the production plan and all production constraints can be obtained. By optimizing material supply and consumption, hydrogen use balance, and tail gas pressure control multiple times, the efficiency and safety of the production process can be effectively ensured.

[0136] The production of polysilicon by the modified Siemens process is a highly systematic project involving multiple key links, from the upstream material supply to the midstream reduction reaction, and then to the downstream tail gas recovery. Each process is closely linked to ensure the smooth operation of the entire production system. This complex production process requires not only precise control and coordination, but also strict constraints of multiple production processes. In this step, the overall start-up time set of the multi-line production combination based on the optimization interval obtained by S206 is used to analyze the hourly material consumption, tail gas pressure, steam balance, etc. of each combination in the set, and the results obtained in S206 are hierarchically iteratively optimized using a big data parallel algorithm to obtain a multi-line production combination based on the optimization interval that meets the constraints of multiple production processes (materials, tail gas, steam, etc.). The overall start-up time set of the multi-line production combination is obtained.

[0137] Specific:

[0138] The improved Siemens process produces polysilicon. Due to the unique process design and operation mode of the reduction furnace, the entire reduction process is periodic. Based on the production line process baseline data, the periodic cumulative consumption of materials, tail gas pressure, steam balance, etc. at different furnace start-up times is obtained, and combined with the overall furnace start-up time set M of the multi-production line production combination based on the optimization interval in S206, the multi-production line production combination overall matrix DataMatrix containing multiple production process (materials, tail gas, steam, etc.) data is obtained. The specific process is as follows:

[0139]

[0140] where {X n ,Y n ,Z n}∈M,X n Arrange the possible start-up times for the day shift, Y n Arrange the possible start-up time for the mid-shift, Z nArrange the possible start times for the night shift;

[0141] A: According to the upstream material supply situation of the reduction workshop, the load of the downstream tail gas recovery system and the balance of auxiliary systems such as steam in actual production, the maximum floating value reference matrix MaxConstraint that meets the constraints of multiple production processes (materials, tail gas, steam, etc.) per hour is generated:

[0142]

[0143] Among them, maxflow0 is the maximum floating value of material consumption at time 0, maxgaspre0 is the maximum floating value of system exhaust gas pressure at time 0, maxsteampre0 is the maximum floating value of system steam at time 0, and so on for other times.

[0144] A1: Based on the multi-line production combination matrix DataMatrix containing data of multiple production processes (materials, exhaust gas, steam, etc.) and the maximum floating value reference matrix MaxConstraint that satisfies the constraints of multiple production processes (materials, exhaust gas, steam, etc.) per hour, the deviation matrix Deviationmatrix is ​​obtained by using the "broadcast subtraction" in mathematics to reflect the difference between each record in the matrix Data_Matrix and the maximum floating value reference matrix MaxConstraint:

[0145] Deviationmatrix=DataMatrix-MaxConstraint;

[0146]

[0147] Among them, Δflow, Δgaspre, and Δsteampre are the deviations between the material, exhaust gas, and steam data per hour in the matrix DataMatrix and the benchmark matrix data.

[0148] A2: Based on the Deviationmatrix obtained in the above steps, set the elements greater than 0 in each record in the matrix to 1 and the rest to 0 to generate a new binary matrix BinaryMatrix:

[0149]

[0150] And generate a check matrix CheakMatrix (all-0 matrix) of the same dimension as the binary matrix BinaryMatrix:

[0151]

[0152] Where m and n are the dimensions of the deviation matrix Deviationmatrix.

[0153] A3: For each row in the binary matrix, use the test matrix CheakMatrix to perform a logical "OR" operation and generate the test result matrix CheakResMatrix:

[0154] CheakResMatrix=BinaryMatrix∨CheakMatrix;

[0155] Check whether the check result matrix CheakResMatrix has 1. If there is 1 in a row, the generated flag vector is 0, and the record is removed; if a row is all 0, the generated flag vector To 1, and keep the record:

[0156]

[0157] Among them, ∨ represents the logical "or" operation. If there is 1 in a row of records, then Otherwise, it is 1.

[0158] B: According to the upstream material supply situation of the reduction workshop, the load of the downstream tail gas recovery system and the balance of auxiliary systems such as steam in actual production, generate the minimum floating value benchmark matrix MinConstraint that meets the constraints of multiple production processes (materials, tail gas, steam, etc.) per hour:

[0159]

[0160] B1: Different from step 1, the deviation matrix Deviationmatrix2 here is obtained by running the "broadcast subtraction" in mathematics based on the multi-line production combination overall matrix DataMatrix containing multi-production process (material, exhaust gas, steam, etc.) data and the minimum floating value reference matrix MinConstraint that satisfies the multi-production process (material, exhaust gas, steam, etc.) constraints per hour:

[0161] Deviationmatrix2=DataMatrix-MinConstraint;

[0162] B2: Different from step 2, here, according to the deviation matrix Deviationmatrix2 obtained in the above steps, the elements less than 0 in each record in the matrix are set to 1, and the rest are set to 0, to generate a new binary matrix BinaryMatrix2:

[0163]

[0164] Except for the above two differences, the rest of the calculation process is the same as the above process A.

[0165] In summary, based on the actual production situation of the workshop, the floating range of hourly material consumption, tail gas pressure, steam balance and other process constraints were set, and the overall start-up time set of the multi-line production combination obtained in S206 was optimized in a hierarchical iterative manner. Not only was an optimized scheduling strategy generated that met the production plan and all production constraints, but also the stability and efficiency of the production system was ensured, while achieving the goals of effective resource utilization, improved production efficiency and environmental friendliness.

[0166] S210, optimization iteration driven by cost KPI factors.

[0167] Based on S202, the electricity cost set corresponding to the start-up of each reduction furnace to be started at 0-23 moments and the completion of a complete production batch under the optimal theoretical operating time is obtained; through this data, the electricity cost of each optimized production scheduling strategy finally obtained in S208 can be determined. The production scheduling strategy with the smallest electricity cost value is selected as the optimal scheduling solution.

[0168] Furthermore, the algorithm of this application also sets clear KPI goals to promote the iterative optimization of the production scheduling strategy that meets the production plan and all production constraints. Specifically, by optimizing and limiting the peak, valley and flat electricity cost revenue data of the optimal scheduling plan, the furnace start-up time and production batch arrangement are gradually adjusted to ensure that each step gradually approaches the cost KPI (i.e., target cost). Finally, the optimal production scheduling strategy that meets both the production plan and all production constraints is generated, thereby maximizing the overall operational benefits.

[0169] A specific implementation method is to calculate the peak, valley and flat electricity charges for each furnace start-up time in the optimal scheduling scheme generated in S208 that meets the production plan and satisfies all production constraints based on the optimization interval information of the single-furnace production plan at different furnace start-up times obtained in S202, and record the relevant results and compare them with the cost KPI, so as to iteratively optimize the optimal scheduling scheme generated in S208, gradually adjust the furnace start-up time and production batch arrangement, to ensure that each step gradually approaches the cost KPI.

[0170] According to the electricity cost data of the start-up time of the reduction furnaces to be started on different production lines obtained in S202 (shift change time can be excluded), the peak-valley and average electricity costs of the optimal scheduling scheme can be calculated. After completing the generation of the optimized production scheduling strategy for the production plan and ensuring that all combinations meet the established production constraints, for each combination of start-up time obtained in S208, the corresponding peak-valley and average electricity cost data are counted, and the peak-valley and average electricity costs of the specific combination are compared and analyzed with the cost KPI (corresponding to the KPI in the following formula). In order to further iteratively optimize each combination in the production scheduling set, a cost function CostFun is defined:

[0171] CostFun=∪ i∈I,j∈J,k∈K (DayPVF i +MidPVF j +Nig tPVF k -KPI);

[0172] Among them, I is the index set of the day shift start-up time, J is the index set of the middle shift start-up time, K is the index set of the night shift start-up time, DayPVF is the peak-valley average electricity price under the day shift combination, MidPVF is the peak-valley average electricity price under the middle shift combination, and NightPVF is the peak-valley average electricity price under the night shift combination.

[0173] In each iteration, by gradually adjusting (X i , Y j , Z k ) to find new combinations of furnace start-up times to ensure that each step gradually approaches the cost KPI.

[0174] This application generates a set of results of optimization interval information for a single furnace production plan, and analyzes in detail the impact of different furnace start-up times on electricity costs. Taking into account the shift work system of the workshop, upstream material supply, and the actual production plan of each production line, combined with the national peak-valley-flat electricity price policy, advanced scheduling algorithms and big data technologies are introduced to perform hierarchical iterative optimization based on process constraints such as material consumption, hydrogen flow, and tail gas pressure. Finally, through iterative optimization based on cost KPI, the globally optimal combination of furnace start-up times is found to achieve energy consumption reduction and maximize overall operational benefits.

[0175] S212. Output the optimal production scheduling strategy that complies with the production plan and satisfies the production constraints.

[0176] This application is based on restoring the production line process baseline data, and combined with the national time-of-use electricity price policy, comprehensively considering the differences of multiple production lines, and obtaining a set of overall scheduling solutions for multi-production line production combinations based on the optimization interval under the production plan. Through the hierarchical iterative optimization of process constraints and cost KPIs, the global optimal scheduling strategy is found, thereby significantly reducing the overall electricity cost. Specifically, this application has the following characteristics: First, a comprehensive model is systematically constructed to comprehensively consider all relevant factors, including furnace type, production equipment, process control baseline, etc.; secondly, global hierarchical iterative optimization is carried out, not only for a single line, but multiple production lines are planned as a whole system to achieve the optimal allocation of resources; thirdly, a data-driven approach is adopted, and big data analysis is used to provide a scientific basis for scheduling decisions; finally, economic benefits are emphasized, and the low electricity price period is maximized through optimized scheduling to effectively reduce electricity costs.

[0177] Specifically, this application generates the optimization interval information of the single-furnace production plan, combines the workshop shift work system (four shifts and three shifts), eliminates the handover time, and ensures the smooth implementation of the production plan. At the same time, based on the upstream material supply situation and the actual consumption of each production line, the algorithm determines the number of furnace starts for each shift and each production line to ensure the safe operation of the equipment and improve production efficiency. In addition, by determining the number of furnace starts for each shift (day shift, middle shift, night shift) and each production line, and combining the preferred furnace start time after eliminating the handover time, the furnace start time permutation and combination between each production line in each shift is calculated. Finally, the big data algorithm is used to comprehensively consider the overall permutation and combination of multi-production line production of all shifts, generate the optimal production scheduling plan set that meets the production plan, and achieve the optimal allocation of resources.

[0178] On this basis, the present application further ensures the efficiency and safety of the production process and enhances the flexibility and adaptability of production by performing hierarchical iterative optimization of materials, tail gas pressure, and steam balance for the overall production scheduling set of multiple production lines that meet the production plan. Through the optimization of cost KPI, the peak-valley-flat electricity cost revenue data corresponding to each combination is optimized and limited, and the furnace start-up time and production batch arrangement are gradually adjusted to ensure that each step gradually approaches the cost KPI. Finally, the optimal production scheduling strategy that meets both the production plan and all production constraints is generated to maximize the overall operational benefits. In summary, the method of the present application not only effectively reduces production costs and improves production efficiency, but also enhances the flexibility and sustainability of production, bringing significant economic and social benefits to the enterprise.

[0179] It should be understood that, for the aforementioned various method embodiments, although the various steps in the flow chart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flow chart of the method embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0180] Based on the same inventive concept, the embodiment of the present application also provides a polysilicon reduction production scheduling device for implementing the polysilicon reduction production scheduling method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more polysilicon reduction production scheduling device embodiments provided below can refer to the limitations of the polysilicon reduction production scheduling method above, and will not be repeated here.

[0181] In one embodiment, Figure 3 As shown, a polysilicon reduction production scheduling device 300 is provided, comprising: an acquisition module 301, a generation module 302, a calculation module 303, a determination module 304 and a screening module 305, wherein:

[0182] An acquisition module 301 is used to acquire the effective production time period and the reduction furnace to be started for each production shift in a single day;

[0183] A generation module 302 is used to determine the possible start-up time of each reduction furnace to be started in each production shift; the possible start-up time is within the effective production time period of the corresponding production shift; generate multiple start-up time combinations for each production shift; each start-up time combination is composed of the possible start-up times of each reduction furnace to be started in the production shift;

[0184] The operation module 303 is used to determine multiple furnace start-up plans for a single day; the multiple furnace start-up plans are obtained by performing a Cartesian product operation on a set consisting of multiple furnace start-up time combinations of each production shift;

[0185] A determination module 304 is used to determine the planned consumption of each production process element in each furnace start-up plan;

[0186] The screening module 305 is used to compare each planned consumption in each furnace start-up plan with the benchmark consumption of the corresponding production process element, and screen out a single-day scheduling plan for polysilicon reduction production from multiple furnace start-up plans based on the comparison results.

[0187] In one embodiment, the production process elements include at least materials, tail gas recovery and steam balance.

[0188] In one embodiment, the screening module 305 can be specifically used to: obtain the maximum baseline consumption and the minimum baseline consumption of each production process element for a single day; compare each planned consumption in each furnace start-up plan with the maximum baseline consumption and the minimum baseline consumption of the corresponding production process element; if each planned consumption in the furnace start-up plan is between the corresponding maximum baseline consumption and the minimum baseline consumption, then the furnace start-up plan is used as a daily scheduling plan for polysilicon reduction production.

[0189] In one embodiment, when there are multiple daily scheduling plans, the polysilicon reduction production scheduling device also includes: a optimization module, which is used to calculate the electricity cost of each reduction furnace to be started at different times to run a corresponding production cycle; the different times are all within a single day period; according to the electricity cost of each reduction furnace to be started at different times, the optimal scheduling plan is selected from each daily scheduling plan.

[0190] In one embodiment, the preferred module can be specifically used to calculate the electricity cost of each daily scheduling plan based on the electricity cost of each reduction furnace to be started at different times; and the daily scheduling plan with the lowest electricity cost is taken as the optimal scheduling plan.

[0191] In one embodiment, the polysilicon reduction production scheduling device further includes: a fine-tuning module for fine-tuning the optimal scheduling scheme with the goal of maximizing the proximity between the electricity cost of the optimal scheduling scheme and the target cost.

[0192] For the specific definition of the polysilicon reduction production scheduling device, please refer to the definition of the polysilicon reduction production scheduling method above, which will not be repeated here. Each module in the above-mentioned polysilicon reduction production scheduling device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0193] In addition, in the implementation of the polysilicon reduction production scheduling device in the above-mentioned example, the logical division of each program module is only an example. In actual applications, the above-mentioned functions can be assigned to different program modules as needed, for example, for the configuration requirements of the corresponding hardware or the convenience of software implementation. That is, the internal structure of the polysilicon reduction production scheduling device is divided into different program modules to complete all or part of the functions described above.

[0194] In one embodiment, a computer device is provided. The computer device may be a mobile terminal or a server terminal. The internal structure diagram of the computer device may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, an input / output interface and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a polysilicon reduction production scheduling method is implemented.

[0195] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0196] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0197] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0198] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0199] It should be noted that in the embodiments of the present application, some existing solutions in the industry such as certain software, components, and models may be mentioned, which should be considered as exemplary, and their purpose is only to illustrate the feasibility of the implementation of the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution. The methods described in this application are all fully independently developed executable software algorithms, and all software algorithms are implemented with the help of general high-level languages, such as C++ language, Python language, etc. The development environment Visual Studio Community Edition, Pycharm Community Edition, etc. are all free and open software, and do not involve software licensing and other intellectual property issues.

[0200] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0201] The terms "including" and "having" and any variations thereof in the embodiments herein are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or (module) units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices. "Multiple" mentioned herein refers to two or more.

[0202] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A polysilicon reduction production scheduling method, characterized in that: The method comprises: Obtain the effective production time period and reduction furnaces to be started for each production shift on a single day; Determine a possible start-up time of each of the reduction furnaces to be started in each production shift; the possible start-up time is within the effective production time period of the corresponding production shift; Generate multiple furnace start time combinations for each production shift; each of the furnace start time combinations is composed of possible furnace start times of each of the reduction furnaces to be started in the production shift; Determine multiple furnace start-up plans for a single day; the multiple furnace start-up plans are obtained by performing a Cartesian product operation on a set consisting of multiple furnace start-up time combinations for each production shift; Determine the planned consumption of each production process element in each furnace start-up plan; Each of the planned consumptions in each furnace start-up plan is compared with the benchmark consumption of the corresponding production process element, and a single-day scheduling plan for polysilicon reduction production is screened out from the multiple furnace start-up plans based on the comparison results.

2. The method according to claim 1, characterized in that The production process elements include at least materials, tail gas recovery and steam balance.

3. The method according to claim 1, characterized in that The step of comparing each of the planned consumption in each furnace start-up plan with the benchmark consumption of the corresponding production process element, and selecting a single-day scheduling plan for polysilicon reduction production from the multiple furnace start-up plans according to the comparison results, comprises: Obtain the maximum benchmark consumption and minimum benchmark consumption of each production process element in a single day respectively; Compare each of the planned consumption in each furnace start-up plan with the maximum baseline consumption and the minimum baseline consumption of the corresponding production process element; If each of the planned consumptions in the furnace startup plan is between the corresponding maximum baseline consumption and minimum baseline consumption, the furnace startup plan will be used as a single-day scheduling plan for polysilicon reduction production.

4. The method according to any one of claims 1 to 3, characterized in that In the case where there are multiple single-day scheduling plans, the method further includes: Calculate the electricity cost of each of the reduction furnaces to be started at different times for a corresponding production cycle; the different times are all within a single day; According to the electricity cost of each reduction furnace to be started, the optimal scheduling plan is selected from the single-day scheduling plans.

5. The method according to claim 4, characterized in that The step of selecting the optimal scheduling scheme from the single-day scheduling schemes according to the electricity cost of each reduction furnace to be started includes: Calculate the electricity cost of each of the single-day scheduling plans according to the electricity cost of each of the reduction furnaces to be started; The single-day dispatch plan with the lowest electricity cost is taken as the optimal dispatch plan.

6. The method according to claim 5, characterized in that The method further comprises: The optimal scheduling scheme is fine-tuned with the goal of maximizing the proximity between the electricity cost of the optimal scheduling scheme and the target cost.

7. A polysilicon reduction production scheduling device, characterized in that: The device comprises: An acquisition module is used to obtain the effective production time period and the reduction furnace to be started for each production shift on a single day; A generating module, used to determine the possible start-up time of each of the reduction furnaces to be started in each production shift; the possible start-up time is within the effective production time period of the corresponding production shift; generate multiple start-up time combinations for each production shift; each start-up time combination is composed of the possible start-up time of each of the reduction furnaces to be started in the production shift; A calculation module is used to determine multiple furnace start-up plans for a single day; the multiple furnace start-up plans are obtained by performing a Cartesian product operation on a set consisting of multiple furnace start-up time combinations of each production shift; A determination module is used to determine the planned consumption of each production process element in each furnace start-up plan; The screening module is used to compare each of the planned consumptions in each furnace start-up plan with the benchmark consumptions of the corresponding production process elements, and screen out a single-day scheduling plan for polysilicon reduction production from the multiple furnace start-up plans based on the comparison results.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images