A method and system for analyzing frequency modulation potential of a polysilicon plant

Abstract

The present application relates to the technical field of industrial load frequency modulation, and particularly relates to a polysilicon plant frequency modulation potential analysis method and system, which comprises the following steps: constructing a mathematical model of a polysilicon production process; establishing a benchmark operation curve optimization model, solving the optimal production operation curve and the corresponding load baseline of the enterprise under normal production state; using an inverse optimization method to deduce the decision preference of the enterprise under different operation conditions, generating operation curves corresponding to multiple operation modes; progressively analyzing the frequency modulation capacity of the polysilicon plant, analyzing the frequency modulation feasibility and economy under different operation conditions; determining the frequency modulation potential interval of the polysilicon plant under different operation conditions, and outputting an optimized frequency modulation scheme. Through the present application, the problem that the prior art cannot accurately evaluate the frequency modulation potential under the condition of meeting the complex production process and operation constraint conditions of polysilicon is effectively solved, and the real potential of the polysilicon plant participating in power frequency modulation under different operation conditions is systematically analyzed.

Description

Technical Field

[0001] This invention relates to the field of industrial load frequency regulation technology, and in particular to a method and system for analyzing the frequency regulation potential of polysilicon plants. Background Technology

[0002] With the large-scale grid connection of new energy sources, the demand for flexible resources in the power system is increasing. As an important demand response resource, the frequency regulation potential assessment of industrial loads is of great significance for grid dispatching and market transactions. Polysilicon production enterprises, as typical high-energy-consuming industrial loads, have characteristics such as continuous production processes, high energy consumption, and adjustable loads, giving them a natural advantage in participating in grid frequency regulation.

[0003] Existing technologies, when assessing the ability of polysilicon manufacturers to participate in power system frequency regulation services, struggle to systematically and accurately analyze the true frequency regulation potential of enterprises under different operating conditions while meeting complex production processes and operational constraints. Because the polysilicon production process involves multiple continuous steps, strict timing and inventory constraints, and various operating states of electrical equipment, traditional frequency regulation assessment methods based on empirical rules or simplified models often fail to simultaneously consider both frequency regulation feasibility and economics. This leads to discrepancies between the assessment results and the actual achievable capabilities of the enterprise, making it difficult to provide an effective decision-making basis for polysilicon plants to participate in frequency regulation. Summary of the Invention

[0004] This invention provides a method for analyzing the frequency modulation potential of polysilicon plants, which can effectively solve the problems in the background art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for analyzing the frequency modulation potential of a polysilicon plant, the method comprising: A mathematical model for polysilicon manufacturing process is constructed based on state-task network theory; Under the constraints of the mathematical model, a baseline operating curve optimization model is established to solve for the optimal production operating curve and the corresponding load baseline of the enterprise under normal production conditions. Based on the baseline operating curve, the inverse optimization method is used to back-calculate the enterprise's decision preferences under different operating conditions, and generate operating curves corresponding to various operating modes. Based on the aforementioned multiple operating curves, a progressive analysis of the frequency modulation capability of polysilicon plants is conducted to analyze the feasibility and economy of frequency modulation under different operating conditions. Based on the frequency modulation capability analysis results, the frequency modulation potential range of the polysilicon plant under different operating conditions is determined, and the final frequency modulation scheme is output.

[0006] Furthermore, the mathematical model for constructing the polysilicon production process based on the state-task network theory includes representing the raw materials, intermediate products, and final products in the polysilicon production process as state nodes, representing the corresponding production processes as task nodes, describing the material transformation relationship through the connection relationship between the state nodes and task nodes, and setting constraints in the mathematical model to characterize equipment operating capacity, inventory capacity, and process sequence relationships.

[0007] Furthermore, under the constraints of the mathematical model, a baseline operating curve optimization model is established to solve for the optimal production operating curve and corresponding load baseline of the enterprise under normal production conditions, including: An optimization model covering a preset scheduling cycle is constructed with the goal of achieving the best economic performance for polysilicon production enterprises. In the optimization model, decision variables reflecting the operating status of production processes, equipment start-up and shutdown status, and power consumption mode are introduced, and production process constraints, equipment operation constraints, material balance constraints, and power balance constraints are used as model constraints. By solving the optimization model, the optimal production operation scheme that satisfies all constraints is obtained, and the power consumption curve corresponding to the optimal production operation scheme is determined as the load baseline.

[0008] Furthermore, based on the baseline operating curve, an inverse optimization method is used to deduce the enterprise's decision preferences under different operating conditions, generating operating curves corresponding to various operating modes, including: A target operating curve is constructed based on the benchmark operating curve, and different operating perturbations are applied to the benchmark operating curve to characterize the enterprise's operating behavior under different operating preferences; Using the target running curve as a reference, an inverse optimization problem is constructed. By inversely calculating the weight parameters of the objective function in the optimization model, the optimization result obtained under the weight parameters matches the target running curve. Based on the inverse optimization results, weight combinations reflecting different enterprise decision-making preferences are obtained, and operating curves under various corresponding operating modes are generated based on the weight combinations.

[0009] Furthermore, based on the aforementioned various operating curves, a progressive analysis of the frequency regulation capability of the polysilicon plant is conducted, analyzing the feasibility and economic efficiency of frequency regulation under different operating conditions, including: The frequency regulation behavior under various operating conditions in a single time period is evaluated. Under the condition of meeting production process constraints and operating constraints, the frequency regulation potential and its economic efficiency corresponding to different frequency regulation time periods and frequency regulation amplitudes are analyzed. Based on the single-period frequency regulation evaluation, considering the temporal coupling relationship between multi-period frequency regulation behaviors, a joint optimization analysis is performed on the frequency regulation response process to evaluate the feasibility of the frequency regulation scheme under multi-period conditions and its overall operational impact. Based on the time-series response optimization results, a multi-objective comparative analysis is conducted on different frequency regulation schemes to construct a trade-off between frequency regulation benefits and costs, and the optimal frequency regulation scheme under different operating conditions is determined through Pareto front analysis.

[0010] Furthermore, the economics of frequency modulation are determined through a net frequency modulation revenue function, which is expressed as: in, This is the net revenue function for frequency modulation; For a period of time; For time period Within, the amount of power change caused by frequency modulation; For revenue from the FM market; The total daily operating cost after frequency adjustment and re-optimization; The benchmark operating cost.

[0011] Furthermore, based on the frequency modulation capability analysis results, the frequency modulation potential range of the polysilicon plant under different operating conditions is determined, and an optimal frequency modulation scheme is output, including: Based on the frequency modulation capability analysis results, the range of frequency modulation power variation that meets production process constraints and operational constraints under different operating conditions and different time periods is determined, forming the corresponding frequency modulation potential range. Based on the frequency modulation potential range and combined with the frequency modulation economic analysis results, frequency modulation schemes with positive frequency modulation benefits within the frequency modulation potential range are selected to form a set of candidate frequency modulation schemes. The candidate frequency modulation schemes are comprehensively compared, and the preferred frequency modulation scheme is determined according to the preset evaluation criteria. The preferred frequency modulation scheme is then used as the frequency modulation decision output of the polysilicon plant under the corresponding operating conditions.

[0012] Further, the candidate frequency modulation schemes are screened to determine the preferred frequency modulation scheme, including: Based on the candidate frequency modulation scheme set, a scheme set containing multiple feasible frequency modulation schemes is constructed; A multi-objective comparison is performed on the frequency modulation schemes in the scheme set. Based on the dominance relationship between frequency modulation benefits and frequency modulation costs, non-dominant frequency modulation schemes that do not simultaneously increase frequency modulation benefits and reduce frequency modulation costs are identified. The non-dominated frequency modulation scheme is used to form a Pareto front, and the preferred frequency modulation scheme is determined from the Pareto front based on a preset decision preference.

[0013] A frequency modulation potential analysis system for a polysilicon plant, the system comprising: The mathematical model building module constructs a mathematical model of the polysilicon production process based on state-task network theory. The baseline curve solving module, under the constraints of the mathematical model, establishes a baseline operating curve optimization model and solves for the optimal production operating curve and corresponding load baseline of the enterprise under normal production conditions. The decision preference inverse optimization module, based on the baseline operating curve, uses inverse optimization methods to back-calculate the enterprise's decision preferences under different operating conditions, generating operating curves corresponding to various operating modes; The frequency modulation capability analysis module performs a progressive analysis of the frequency modulation capability of the polysilicon plant based on the various operating curves, and analyzes the feasibility and economy of frequency modulation under different operating conditions. The frequency modulation scheme output module determines the frequency modulation potential range of the polysilicon plant under different operating conditions based on the frequency modulation capability analysis results, and outputs the final frequency modulation scheme.

[0014] Furthermore, the benchmark curve solving module includes: The optimization model building unit aims to optimize the economic efficiency of polysilicon production enterprises and construct an optimization model covering a preset scheduling cycle. The model structure setting unit introduces decision variables that reflect the operating status of production processes, the start-up and shutdown status of equipment, and the power consumption mode in the optimization model, and uses production process constraints, equipment operation constraints, material balance constraints, and power balance constraints as model constraint conditions. The optimization model solving unit solves the optimization model to obtain the optimal production operation scheme that satisfies all constraints, and determines the power consumption curve corresponding to the optimal production operation scheme as the load baseline.

[0015] The technical solution of this invention can achieve the following technical effects: This invention constructs a mathematical model of the polysilicon production process and determines the enterprise's baseline operating curve and load baseline under the constraints of this model. This allows for an accurate depiction of the enterprise's normal production state while fully considering the complexity of the polysilicon production process and operational constraints, avoiding the overestimation or underestimation of frequency regulation potential caused by model simplification in existing methods. Furthermore, this invention introduces an inverse optimization method to deduce the enterprise's decision preferences under different operating conditions. This effectively depicts the comprehensive trade-off behavior of polysilicon manufacturers regarding cost, stability, and risk in actual operation, making the frequency regulation potential analysis results closer to the enterprise's actual feasible capabilities and improving the credibility and practicality of the frequency regulation assessment results. In addition, this invention performs a progressive analysis of the polysilicon plant's frequency regulation capability based on multiple operating curves, simultaneously considering the trade-offs between single-period frequency regulation feasibility, multi-period frequency regulation timing effects, and frequency regulation economics. This achieves a systematic assessment of frequency regulation potential, overcoming the shortcomings of existing technologies that only analyze frequency regulation capability from a partial or single dimension. By further determining the frequency regulation potential range and outputting the optimal frequency regulation scheme based on the above analysis results, this invention can provide decision-making basis with direct guidance for polysilicon production enterprises to participate in power system frequency regulation, and improve the operability and practical application value of industrial loads participating in frequency regulation while ensuring production continuity and economy.

[0016] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the frequency modulation potential analysis method for polysilicon plants; Figure 2 A schematic diagram for finding the optimal production operation curve and corresponding load baseline for a company under normal production conditions; Figure 3 A flowchart illustrating the process of generating operating curves for various operating modes; Figure 4 A flowchart illustrating the feasibility and economics of frequency regulation under different operating conditions; Figure 5A flowchart illustrating the process of determining the frequency modulation potential range of a polysilicon plant under different operating conditions and outputting the final frequency modulation scheme. Figure 6 A flowchart illustrating the process of selecting the optimal frequency modulation scheme from candidate schemes. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0021] Example 1: like Figure 1 As shown, this application provides a method for analyzing the frequency modulation potential of a polysilicon plant, the method comprising: S1: Constructing a mathematical model for polysilicon manufacturing process based on state-task network theory; Specifically, considering the characteristics of polysilicon production processes—numerous steps, long processes, and complex coupling relationships—state-task network theory is introduced as a unified framework for production process modeling, providing a structured description of the polysilicon production process. During modeling, the polysilicon production process is treated as a whole, and production activities are abstractly expressed from the perspectives of material flow and process execution. This allows the constructed mathematical model to simultaneously reflect the technological structural characteristics of the production process and the basic constraints that need to be satisfied during production operation. The mathematical model established in this way does not rely on the local description of specific equipment or single processes, but rather characterizes the technological correlations and overall operational characteristics of the polysilicon production process in a network form, thus providing a fundamental support for subsequent production operation analysis within the unified model framework.

[0022] S2: Under the constraints of the mathematical model, establish a baseline operating curve optimization model to solve for the optimal production operating curve and corresponding load baseline of the enterprise under normal production conditions. Specifically, after completing the mathematical modeling of the polysilicon production process, the analysis further focuses on the normal production operation state of the enterprise. Within the production structure and operational boundaries described by the aforementioned mathematical model, optimization analysis methods are introduced to plan the overall operation of the polysilicon production process. This step aims to reflect the enterprise's routine production behavior without participating in power frequency regulation. By optimizing the overall operation state of the production process within a preset operating cycle, production operation results matching the normal production rhythm are obtained. In this preferred embodiment, the benchmark operating curve is used to depict the changes in power consumption corresponding to production activities in each time period while maintaining the established production targets and operational stability, thus forming a time series characteristic reflecting the enterprise's normal power consumption behavior. Simultaneously, this benchmark operating curve serves as a load baseline to characterize the reference power consumption level of the enterprise when no frequency regulation is implemented. The benchmark operating curve and load baseline determined in the above manner can truly reflect the overall power consumption characteristics of the polysilicon production enterprise under normal operating conditions, providing a unified reference basis for subsequent analysis of production operation adjustments and the impact of frequency regulation based on this benchmark state.

[0023] S3: Based on the baseline operating curve, the inverse optimization method is used to back-calculate the enterprise's decision preferences under different operating conditions and generate operating curves corresponding to various operating modes; Specifically, after obtaining the baseline operating curve reflecting the normal production status of the enterprise, this baseline operating curve is used as important reference information to characterize the enterprise's existing operating behavior. A reverse analysis approach is introduced to characterize the enterprise's production decision-making behavior. By analyzing the electricity consumption variation characteristics, operational stability, and production rhythm reflected in the baseline operating curve, the implicit decision-making tendencies of the enterprise during actual operation are regarded as decision preferences not explicitly given in the optimization model. These decision preferences are then inferred backwards using the inverse optimization method. In this preferred embodiment, the inverse optimization process does not directly set the enterprise's operating goals, but rather identifies combinations of decision preference parameters that reflect the enterprise's operating characteristics by ensuring that the model output is consistent with the reference operating behavior. Based on different combinations of decision preference parameters, multiple sets of production operation results that conform to production process constraints and have different operating characteristics can be obtained, thus forming multiple operating curves corresponding to different operating modes. Through this method, without changing the production process structure, the diverse operating strategies that polysilicon production enterprises may adopt under different operating conditions can be characterized, providing more realistic operating scenario inputs for subsequent frequency regulation capability analysis based on multiple operating modes.

[0024] S4: Based on multiple operating curves, a progressive analysis of the frequency modulation capability of polysilicon plants is conducted to analyze the feasibility and economy of frequency modulation under different operating conditions. Specifically, in a preferred embodiment, after obtaining multiple production operation curves reflecting different operating modes, these curves are used as a set of inputs representing the possible operating states of the polysilicon plant, and frequency modulation behavior is analyzed hierarchically and progressively. This progressive analysis method, under the premise of not affecting normal production operations, first assesses the responsiveness of production to frequency modulation commands under localized adjustment conditions from a single operating state. Then, it considers the continuity and cumulative impact of operating states over time, comprehensively analyzing the production changes that frequency modulation behavior may cause under multi-period conditions. Through this progressive analysis method, from local to overall, from single to multi-period, it is possible to systematically identify whether frequency modulation behavior is feasible within the constraints of the production process under different operating conditions, and simultaneously assess the impact of frequency modulation behavior on production operating costs and revenue structure. Using this progressive analysis method avoids the one-sidedness of evaluating frequency modulation capability solely from a single period or operating state, making the frequency modulation capability analysis results more comprehensive and robust, and providing a reliable basis for subsequent frequency modulation potential assessment and scheme selection based on the analysis results.

[0025] S5: Based on the frequency modulation capability analysis results, determine the frequency modulation potential range of the polysilicon plant under different operating conditions, and output the final frequency modulation scheme.

[0026] Specifically, after completing a progressive analysis of frequency regulation capabilities under various operating curves, the frequency regulation response characteristics obtained during the analysis are summarized and organized to reflect the frequency regulation adjustment range that polysilicon plants can achieve under different operating conditions and time periods. By summarizing the feasible range of frequency regulation behavior under each operating condition within production process constraints and operating boundaries, corresponding frequency regulation potential intervals are formed to characterize the capability boundary of polysilicon plants to participate in frequency regulation while ensuring normal production operation. Based on this, combined with the overall assessment results of the impact of frequency regulation behavior on production operation under different operating conditions, multiple feasible frequency regulation schemes located within the frequency regulation potential interval are comprehensively compared to screen out the frequency regulation scheme with better overall effect within the production tolerance range. The screened frequency regulation scheme is then used as the decision output for polysilicon plants to participate in power frequency regulation under the corresponding operating conditions. Through this step, the aforementioned analysis results can be transformed into frequency regulation decision information with practical application significance, realizing a complete closed loop from frequency regulation capability assessment to frequency regulation scheme determination.

[0027] This invention effectively solves the problem that existing technologies struggle to accurately assess frequency regulation potential under the complex production processes and operational constraints of polysilicon, and enables a systematic analysis of the true potential of polysilicon plants to participate in power frequency regulation under different operating conditions.

[0028] As a preferred embodiment of the above, a mathematical model of the polysilicon production process is constructed based on the state-task network theory. This includes representing the raw materials, intermediate products and final products in the polysilicon production process as state nodes, representing the corresponding production processes as task nodes, describing the material transformation relationship through the connection relationship between state nodes and task nodes, and setting constraints in the mathematical model to characterize equipment operating capacity, inventory capacity and process sequence relationship.

[0029] Specifically, firstly, based on the actual process flow of polysilicon production enterprises, the material forms involved in the production process are sorted out. Materials that can be stored, transferred, or used as process inputs and outputs are uniformly abstracted as state nodes. The material types corresponding to state nodes preferably include raw materials, intermediate products at each stage, and final products. Trackable inventory attributes are set for different materials to reflect the accumulation and consumption of materials during the production process. Subsequently, the production processes that realize material form changes or material transfers are abstracted as task nodes. Task nodes preferably cover the key process links in polysilicon production, and each task node is configured with operating attributes consistent with the actual equipment capabilities to characterize the processing capacity and operating boundaries of the process at different operating levels. On this basis, by establishing the connection relationship between state nodes and task nodes, the input and output states of each task node are clarified, so that the network structure can express the transformation path of materials from upstream to downstream and the material coupling relationship between different processes. This system forms a state-task network model that reflects the structural characteristics of polysilicon production processes. Furthermore, to ensure the model accurately reflects the operational boundaries, constraints related to equipment operating capacity are set to limit the processing capacity range achievable by task nodes. Constraints related to inventory capacity are also set to limit the upper and lower limits of inventory for each state node. Additionally, constraints related to process sequence relationships are set to reflect the necessary order, continuity, and logical start-up and shutdown procedures between processes. For example, in a typical production chain, if the output inventory of the upstream process does not reach the safety lower limit, the downstream process will not enter operation. Alternatively, during the operation of the downstream process, the upstream process must maintain material supply to prevent inventory from falling below the lower limit. These constraints simultaneously incorporate the process coupling relationship and operational boundaries into a unified networked mathematical description, enabling those skilled in the art to conduct subsequent production operation analysis and frequency regulation capability assessment based on this model.

[0030] As a preferred embodiment of the above, such as Figure 2 As shown, step S2 involves establishing a baseline operating curve optimization model under the constraints of the mathematical model, and solving for the optimal production operating curve and corresponding load baseline of the enterprise under normal production conditions, including: S21: To optimize the economic efficiency of polysilicon production enterprises, an optimization model covering the preset scheduling cycle is constructed. S22: In the optimization model, decision variables reflecting the operating status of production processes, equipment start-up and shutdown status, and power consumption mode are introduced, and production process constraints, equipment operation constraints, material balance constraints, and power balance constraints are used as model constraints. S23: By solving the optimization model, the optimal production operation scheme that satisfies all constraints is obtained, and the power consumption curve corresponding to the optimal production operation scheme is determined as the load baseline.

[0031] First, let's describe the above process using formulas: With optimal economic efficiency as the objective, we establish a mixed-integer linear programming (MILP) model with a 24-hour scheduling cycle, where the objective is a multi-objective function: Objective 1: Minimize operating costs: ; In the formula: Let t be the day-ahead market electricity price. Electricity purchased from the day-ahead market; For bilateral contract electricity prices, Purchase electricity under contract; The cost of self-generated electricity is 0.10 yuan / kWh; For equipment startup costs, For initiating action; υ k For downtime costs, This is a shutdown action.

[0032] Objective 2: Minimize power fluctuations: In the formula: Let be the total power consumption during time period t. The goal is to reduce load surges and improve power stability.

[0033] Objective 3: Maximize equipment utilization: In the formula: Let |Θ| represent the operating status of device θ in time period t (1 for running, 0 for stopped), T be the total number of time periods in the scheduling cycle, and |Θ| be the total number of devices.

[0034] Multi-objective integration: In the formula: w 1 w 2 w 3 Weighting coefficient (w) 1 +w 2 +w 3 =1).

[0035] Key constraints: Material balance constraints: In the formula: Let be the inventory level of the material corresponding to task k in time period t. For production, For conversion efficiency, for DC tasks that include byproduct recovery: In the formula: This refers to the output of the by-product processing stage. =0.88 is the by-product conversion rate, and σ=12% is the RF by-product output rate.

[0036] Inventory capacity constraints: .

[0037] Power balance constraints: In the formula: Let k be the energy consumption per unit output of task k at time t (kWh / ton).

[0038] Gear selection constraints: , In the formula: Select a variable (0-1 variable) for the gear position. The output level corresponding to gear g is determined, and gear switching is constrained by the principle of gradual shifting. .

[0039] Process coupling constraints: , ; Start / stop logic constraints: .

[0040] The baseline operating curve optimization model is used to determine the reasonable operating arrangements of polysilicon production enterprises under normal production conditions within a preset scheduling cycle. By setting an optimization objective with operating cost as the core, the model comprehensively considers the overall operating costs of the enterprise under different combinations of power consumption and equipment start-up and shutdown states, so that the resulting production operation plan meets the economic operation needs of the enterprise. At the same time, by introducing power change and equipment utilization-related objectives, the model constrains power consumption fluctuations and equipment operating status during production operation, so that the baseline operating state remains relatively stable in the time dimension and conforms to actual production organization habits.

[0041] Regarding constraints, the model ensures a reasonable output-consumption relationship for each production process within the scheduling cycle through material balance and inventory capacity constraints, avoiding raw material shortages or material backlogs. Power balance constraints ensure that the electricity demand for production activities in each time period can be met by both purchased and self-generated electricity, thus guaranteeing the feasibility of the production operation plan at the power level. Furthermore, constraints related to equipment operation, speed selection, and process coupling ensure that changes in equipment operating status and upstream / downstream process relationships conform to the actual process requirements of polysilicon production. By jointly solving the above objectives and constraints, the optimal production operation plan that meets normal production requirements can be obtained, and the corresponding power consumption changes are output as the load baseline.

[0042] As a preferred embodiment of the above, such as Figure 3 As shown, in step S3, based on the baseline operating curve, the inverse optimization method is used to back-calculate the enterprise's decision preferences under different operating conditions, generating operating curves corresponding to various operating modes, including: S31: Construct a target operating curve based on the baseline operating curve, and characterize the enterprise's operating behavior under different operating preferences by applying different operating perturbations to the baseline operating curve; S32: Using the target running curve as a reference, construct an inverse optimization problem, and inversely deduce the weight parameters of the objective function in the optimization model so that the optimization result obtained under the weight parameters matches the target running curve; S33: Based on the inverse optimization results, obtain weight combinations that reflect the decision-making preferences of different enterprises, and generate corresponding operating curves under various operating modes based on the weight combinations.

[0043] Specifically, after obtaining the baseline operating curve reflecting the normal production status of the enterprise, in order to characterize the diverse production decision-making behaviors that the enterprise may take under different operating conditions, multiple target operating curves are constructed based on the baseline operating curves. By making appropriate adjustments to the operating rhythm, load distribution, or equipment usage without changing the production process structure, the operating behavior characteristics that the enterprise may form under different operating preference conditions are simulated. Subsequently, using these target operating curves as reference operating results, the inverse optimization analysis method is introduced to inversely infer the decision preferences in the production operation optimization model, so that the operating results obtained by the model under given decision preference conditions can be consistent with the corresponding target operating curves, thereby identifying the combination of decision preference parameters that match different operating behaviors. On this basis, according to the different combination of decision preference parameters obtained from the inverse optimization analysis, the production operation model is solved again, generating multiple sets of production operation curves that meet the production process constraints and have different operating characteristics, so that each set of operating curves corresponds to a feasible operating mode. By using the above methods, while maintaining the consistency of production processes, it is possible to systematically characterize the various possible operating states of polysilicon production enterprises under different operating conditions and different decision-making preferences, providing a more realistic operating scenario basis for subsequent frequency modulation capability analysis based on multiple operating modes.

[0044] As a preferred embodiment of the above, such as Figure 4 As shown, step S4 involves a progressive analysis of the frequency regulation capability of the polysilicon plant based on various operating curves, analyzing the feasibility and economics of frequency regulation under different operating conditions, including: S41: Evaluate the frequency regulation behavior in a single time period under various operating conditions, and analyze the frequency regulation potential and its economic efficiency corresponding to different frequency regulation periods and amplitudes, while meeting production process constraints and operating constraints. S42: Based on the single-period frequency regulation evaluation, considering the temporal coupling relationship between multi-period frequency regulation behaviors, a joint optimization analysis of the frequency regulation response process is conducted to evaluate the feasibility of the frequency regulation scheme under multi-period conditions and its overall operational impact. S43: Based on the time-series response optimization results, conduct multi-objective comparative analysis on different frequency regulation schemes, construct the trade-off relationship between frequency regulation benefits and frequency regulation costs, and determine the optimal frequency regulation scheme under different operating conditions through Pareto front analysis.

[0045] Specifically, from a single-period perspective, without altering the production process structure and meeting equipment operation and material flow requirements, the ability of each operating condition to respond to frequency modulation commands within a single time period is evaluated. By analyzing the frequency adjustment range and its impact on production operations, the frequency modulation range that can be implemented within each time period is identified, and the impact of corresponding frequency modulation behavior on the enterprise's operating costs and revenue structure within that time period is simultaneously evaluated. Subsequently, based on the single-period evaluation results, the continuity and cumulative effect of frequency modulation behavior over time are further considered, incorporating frequency modulation behavior across multiple time periods into a unified analysis framework. The frequency modulation response process is then jointly evaluated to determine the comprehensive impact of continuous or multiple frequency modulations under multi-period conditions on production stability, output completion, and overall operating status. Finally, based on the frequency modulation response results obtained from the multi-period analysis, different frequency modulation schemes are comprehensively compared and analyzed. Considering both frequency modulation benefits and costs, a trade-off relationship between the two is constructed, and the Pareto front analysis method is used to screen out frequency modulation schemes that do not exhibit significant deterioration under different operating conditions, thereby determining the frequency modulation scheme with superior overall performance under each operating condition. Through the progressive analysis process described above, from single-period to multi-period and then to multi-objective trade-offs, we can systematically characterize the feasibility and economy of polysilicon plants participating in frequency regulation under different operating conditions, laying a reliable foundation for subsequent determination of frequency regulation potential range and output of frequency regulation schemes.

[0046] As a preferred embodiment of the above, the economics of frequency modulation are determined by a net frequency modulation benefit function, which is expressed as follows: in, This is the net revenue function for frequency modulation; For a period of time; For time period Within, the amount of power change caused by frequency modulation; For revenue from the FM market; The total daily operating cost after frequency adjustment and re-optimization; The benchmark operating cost.

[0047] Specifically, the economics of frequency regulation is determined by comparing the changes in benefits and costs brought about by frequency regulation within a given time period. The net benefit function of frequency regulation measures the change in the overall operational effect of an enterprise after implementing frequency regulation. This determination method uses benchmark operating costs as a reference, taking the enterprise's operating costs under normal production conditions without participating in frequency regulation as the baseline. Based on this, it analyzes the changes in the enterprise's total daily operating costs due to adjustments in production operation plans after implementing frequency regulation for a specific time period and at a specific frequency regulation amplitude. Simultaneously, the frequency regulation revenue obtained by the enterprise from the electricity market due to participation in frequency regulation is included in a unified evaluation framework. By calculating the difference between the frequency regulation revenue and the increase in operating costs, it is determined whether the frequency regulation behavior is economically worthwhile. When the net benefit of frequency regulation is positive, it indicates that the frequency regulation behavior at the corresponding time period and frequency regulation amplitude can bring additional economic benefits to the enterprise while meeting production operation requirements; conversely, when the net benefit of frequency regulation is negative, it indicates that the frequency regulation behavior is not economically feasible under the current operating conditions. By adopting the above-mentioned method for determining the net benefit of frequency regulation, the economic efficiency of different frequency regulation periods and amplitudes can be evaluated in a unified and intuitive manner without deviating from the overall constraints of production and operation, providing clear economic criteria for subsequent frequency regulation scheme selection and frequency regulation potential analysis.

[0048] As a preferred embodiment of the above, such as Figure 5 As shown, in step S5, based on the frequency regulation capability analysis results, the frequency regulation potential range of the polysilicon plant under different operating conditions is determined, and the final frequency regulation scheme is output, including: S51: Based on the frequency modulation capability analysis results, determine the range of frequency modulation power variation that meets production process constraints and operational constraints under different operating conditions and time periods, and form the corresponding frequency modulation potential range. S52: Based on the frequency modulation potential range and combined with the frequency modulation economic analysis results, select frequency modulation schemes with positive frequency modulation benefits within the frequency modulation potential range to form a set of candidate frequency modulation schemes; S53: Conduct a comprehensive comparison of the candidate frequency modulation scheme set, determine the preferred frequency modulation scheme according to the preset evaluation criteria, and use the preferred frequency modulation scheme as the frequency modulation decision output of the polysilicon plant under the corresponding operating conditions.

[0049] Specifically, after completing the progressive analysis of frequency regulation capability under various operating conditions, the frequency regulation response results under different operating states and time periods are summarized and organized to identify the power variation range that allows for frequency regulation adjustment under the conditions of meeting production process requirements and operating boundaries, and this range is described as the frequency regulation potential range under the corresponding operating conditions. The frequency regulation potential range formed in this way can intuitively reflect the upper and lower limits of regulation that a polysilicon plant can achieve when participating in frequency regulation under different operating conditions, providing a clear capability boundary for frequency regulation decisions. Based on this, the frequency regulation potential range is combined with the aforementioned frequency regulation economic analysis results to screen various feasible frequency regulation schemes within the range. Schemes that can generate positive frequency regulation benefits under corresponding operating conditions without adversely affecting normal production operations are prioritized, thus forming a candidate frequency regulation scheme set. Subsequently, a comprehensive comparative analysis of the candidate scheme set is conducted. Based on pre-set evaluation criteria, different schemes are weighed in terms of benefit level, operational impact, and implementation stability to determine the frequency regulation scheme with the best overall effect under the current operating conditions. This scheme is then output as the decision result for the polysilicon plant's participation in power frequency regulation. Through these steps, the frequency regulation capability analysis results can be effectively transformed into executable frequency regulation decisions, achieving a complete application closed loop from capability assessment to scheme determination.

[0050] As a preferred embodiment of the above, such as Figure 6 As shown, the candidate frequency modulation schemes are screened to determine the preferred frequency modulation scheme, including: A10: Based on the candidate frequency modulation scheme set, construct a scheme set containing multiple feasible frequency modulation schemes; A20: Perform multi-objective comparison of frequency modulation schemes in the scheme set, and based on the dominance relationship between frequency modulation benefits and frequency modulation costs, identify non-dominant frequency modulation schemes that do not simultaneously increase frequency modulation benefits and reduce frequency modulation costs. A30: Construct the Pareto front from the non-dominated frequency modulation scheme, and determine the preferred frequency modulation scheme from the Pareto front based on the preset decision preference.

[0051] Specifically, the candidate frequency regulation schemes are first uniformly organized, and schemes that are feasible under different operating conditions and time periods are gathered to form a comparable scheme set, allowing all schemes to be analyzed under the same evaluation framework. Subsequently, a multi-objective comparative analysis is conducted on each scheme in the scheme set from two dimensions: frequency regulation benefits and frequency regulation costs. Frequency regulation benefits reflect the market return level obtained by enterprises from participating in frequency regulation, while frequency regulation costs reflect the adverse effects of frequency regulation on production operations, such as increased costs, operational disturbances, or management burdens. By comparing the performance of different frequency regulation schemes in the above two dimensions, schemes that do not have an advantage over them while simultaneously increasing frequency regulation benefits and reducing frequency regulation costs are identified, and these schemes are determined as non-dominated frequency regulation schemes. Based on this, the non-dominated frequency regulation schemes are mapped into the target space consisting of frequency regulation benefits and costs, forming a Pareto front reflecting the trade-offs between different schemes. Finally, according to the enterprise's decision-making preferences regarding benefit levels, operational stability, and risk tolerance in actual operation, the frequency regulation scheme that best meets the current operational needs is selected from the non-dominated frequency regulation schemes corresponding to the Pareto front. This selected scheme is then output as the preferred frequency regulation scheme for polysilicon plants participating in power frequency regulation under the corresponding operating conditions. Through this screening process, a rational choice between multiple frequency regulation schemes can be made while ensuring the feasibility of the schemes, improving the scientific nature and feasibility of frequency regulation decisions.

[0052] Example 2: Based on the same inventive concept as the polysilicon plant frequency modulation potential analysis method in the foregoing embodiments, the present invention also provides a polysilicon plant frequency modulation potential analysis system, comprising: The mathematical model building module constructs a mathematical model of the polysilicon production process based on state-task network theory. The baseline curve solving module establishes a baseline operating curve optimization model under the constraints of the mathematical model, and solves for the optimal production operating curve and the corresponding load baseline of the enterprise under normal production conditions. The decision preference inverse optimization module, based on the baseline operating curve, uses inverse optimization methods to back-calculate the enterprise's decision preferences under different operating conditions, generating operating curves corresponding to various operating modes; The frequency modulation capability analysis module performs a progressive analysis of the frequency modulation capability of polysilicon plants based on various operating curves, and analyzes the feasibility and economy of frequency modulation under different operating conditions. The frequency modulation scheme output module determines the frequency modulation potential range of the polysilicon plant under different operating conditions based on the frequency modulation capability analysis results, and outputs the optimal frequency modulation scheme.

[0053] The analysis system described above in this invention can effectively realize the frequency modulation potential analysis method for polysilicon plants, and the technical effects it can achieve are as described in the above embodiments, which will not be repeated here.

[0054] Furthermore, the baseline curve solving module includes: The optimization model building unit aims to optimize the economic efficiency of polysilicon production enterprises and construct an optimization model covering a preset scheduling cycle. The model structure setting unit introduces decision variables that reflect the operating status of production processes, equipment start-up and shutdown status, and power consumption mode in the optimization model, and uses production process constraints, equipment operation constraints, material balance constraints, and power balance constraints as model constraints. The optimization model solving unit solves the optimization model to obtain the optimal production operation scheme that satisfies all constraints, and determines the power consumption curve corresponding to the optimal production operation scheme as the load baseline.

[0055] Similarly, the above-mentioned optimization schemes for the system can also achieve the optimization effects corresponding to the methods in Embodiment 1, which will not be repeated here.

[0056] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of the application as defined herein, and are to be considered as covering any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Thus, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for analyzing the frequency modulation potential of a polysilicon plant, characterized in that, The method includes: A mathematical model for polysilicon manufacturing process is constructed based on state-task network theory; Under the constraints of the mathematical model, a baseline operating curve optimization model is established to solve for the optimal production operating curve and the corresponding load baseline of the enterprise under normal production conditions. Based on the baseline operating curve, the inverse optimization method is used to back-calculate the enterprise's decision preferences under different operating conditions, and generate operating curves corresponding to various operating modes. Based on the aforementioned multiple operating curves, a progressive analysis of the frequency modulation capability of polysilicon plants is conducted to analyze the feasibility and economy of frequency modulation under different operating conditions. Based on the frequency modulation capability analysis results, the frequency modulation potential range of the polysilicon plant under different operating conditions is determined, and the final frequency modulation scheme is output.

2. The method for analyzing the frequency modulation potential of a polysilicon plant according to claim 1, characterized in that, The mathematical model for polysilicon production based on state-task network theory includes representing raw materials, intermediate products and final products in the polysilicon production process as state nodes, representing the corresponding production processes as task nodes, describing the material transformation relationship through the connection relationship between the state nodes and task nodes, and setting constraints in the mathematical model to characterize equipment operating capacity, inventory capacity and process sequence relationship.

3. The method for analyzing the frequency modulation potential of a polysilicon plant according to claim 1, characterized in that, Under the constraints of the mathematical model, a baseline operating curve optimization model is established to solve for the optimal production operating curve and corresponding load baseline of the enterprise under normal production conditions, including: An optimization model covering a preset scheduling cycle is constructed with the goal of achieving the best economic performance for polysilicon production enterprises. In the optimization model, decision variables reflecting the operating status of production processes, equipment start-up and shutdown status, and power consumption mode are introduced, and production process constraints, equipment operation constraints, material balance constraints, and power balance constraints are used as model constraints. By solving the optimization model, the optimal production operation scheme that satisfies all constraints is obtained, and the power consumption curve corresponding to the optimal production operation scheme is determined as the load baseline.

4. The method for analyzing the frequency modulation potential of a polysilicon plant according to claim 1, characterized in that, Based on the baseline operating curve, an inverse optimization method is used to deduce the enterprise's decision preferences under different operating conditions, generating operating curves corresponding to various operating modes, including: A target operating curve is constructed based on the benchmark operating curve, and different operating perturbations are applied to the benchmark operating curve to characterize the enterprise's operating behavior under different operating preferences; Using the target running curve as a reference, an inverse optimization problem is constructed. By inversely calculating the weight parameters of the objective function in the optimization model, the optimization result obtained under the weight parameters matches the target running curve. Based on the inverse optimization results, weight combinations reflecting different enterprise decision-making preferences are obtained, and operating curves under various corresponding operating modes are generated based on the weight combinations.

5. The method for analyzing the frequency modulation potential of a polysilicon plant according to claim 1, characterized in that, Based on the aforementioned various operating curves, a progressive analysis of the frequency regulation capability of polysilicon plants is conducted, analyzing the feasibility and economics of frequency regulation under different operating conditions, including: The frequency regulation behavior under various operating conditions in a single time period is evaluated. Under the condition of meeting production process constraints and operating constraints, the frequency regulation potential and its economic efficiency corresponding to different frequency regulation time periods and frequency regulation amplitudes are analyzed. Based on the single-period frequency regulation evaluation, considering the temporal coupling relationship between multi-period frequency regulation behaviors, a joint optimization analysis is performed on the frequency regulation response process to evaluate the feasibility of the frequency regulation scheme under multi-period conditions and its overall operational impact. Based on the time-series response optimization results, a multi-objective comparative analysis is conducted on different frequency regulation schemes to construct a trade-off between frequency regulation benefits and costs, and the optimal frequency regulation scheme under different operating conditions is determined through Pareto front analysis.

6. The method for analyzing the frequency modulation potential of a polysilicon plant according to claim 5, characterized in that, The economics of frequency modulation are determined by the net frequency modulation revenue function, which is expressed as follows: in, This is the net revenue function for frequency modulation; For a period of time; For time period Within, the amount of power change caused by frequency modulation; For revenue from the FM market; The total daily operating cost after frequency adjustment and re-optimization; The benchmark operating cost.

7. The method for analyzing the frequency modulation potential of a polysilicon plant according to claim 1, characterized in that, Based on the frequency modulation capability analysis results, the frequency modulation potential range of the polysilicon plant under different operating conditions is determined, and the optimal frequency modulation scheme is output, including: Based on the frequency modulation capability analysis results, the range of frequency modulation power variation that meets production process constraints and operational constraints under different operating conditions and different time periods is determined, forming the corresponding frequency modulation potential range. Based on the frequency modulation potential range and combined with the frequency modulation economic analysis results, frequency modulation schemes with positive frequency modulation benefits within the frequency modulation potential range are selected to form a set of candidate frequency modulation schemes. The candidate frequency modulation schemes are comprehensively compared, and the preferred frequency modulation scheme is determined according to the preset evaluation criteria. The preferred frequency modulation scheme is then used as the frequency modulation decision output of the polysilicon plant under the corresponding operating conditions.

8. The method for analyzing the frequency modulation potential of a polysilicon plant according to claim 1, characterized in that, The process of screening the candidate frequency modulation schemes to determine the preferred frequency modulation scheme includes: Based on the candidate frequency modulation scheme set, a scheme set containing multiple feasible frequency modulation schemes is constructed; A multi-objective comparison is performed on the frequency modulation schemes in the scheme set. Based on the dominance relationship between frequency modulation benefits and frequency modulation costs, non-dominant frequency modulation schemes that do not simultaneously increase frequency modulation benefits and reduce frequency modulation costs are identified. The non-dominated frequency modulation scheme is used to form a Pareto front, and a preferred frequency modulation scheme is determined from the Pareto front based on a preset decision preference.

9. A frequency modulation potential analysis system for a polysilicon plant, characterized in that, The system includes: The mathematical model building module constructs a mathematical model of the polysilicon production process based on state-task network theory. The baseline curve solving module, under the constraints of the mathematical model, establishes a baseline operating curve optimization model and solves for the optimal production operating curve and corresponding load baseline of the enterprise under normal production conditions. The decision preference inverse optimization module, based on the baseline operating curve, uses inverse optimization methods to back-calculate the enterprise's decision preferences under different operating conditions, generating operating curves corresponding to various operating modes; The frequency modulation capability analysis module performs a progressive analysis of the frequency modulation capability of the polysilicon plant based on the various operating curves, and analyzes the feasibility and economy of frequency modulation under different operating conditions. The frequency modulation scheme output module determines the frequency modulation potential range of the polysilicon plant under different operating conditions based on the frequency modulation capability analysis results, and outputs the final frequency modulation scheme.

10. The polysilicon plant frequency modulation potential analysis system according to claim 9, characterized in that, The baseline curve solving module includes: The optimization model building unit aims to optimize the economic efficiency of polysilicon production enterprises and construct an optimization model covering a preset scheduling cycle. The model structure setting unit introduces decision variables that reflect the operating status of production processes, the start-up and shutdown status of equipment, and the power consumption mode in the optimization model, and uses production process constraints, equipment operation constraints, material balance constraints, and power balance constraints as model constraint conditions. The optimization model solving unit solves the optimization model to obtain the optimal production operation scheme that satisfies all constraints, and determines the power consumption curve corresponding to the optimal production operation scheme as the load baseline.

Images