Cement industry load production optimization scheduling method considering adjustability of rotary kiln

Abstract

The application provides a cement industry load production optimization scheduling method considering the adjustability of a rotary kiln, and belongs to the technical field of cement industry load production optimization scheduling; in order to solve the defects that the rotary kiln is generally regarded as a constant load and the adjustable potential is ignored in the existing cement industry load, the technical scheme is that: the rotary kiln is modeled by combining mechanism and data, the adjustability is verified, and a mechanism model of the coupling relationship between the rotary kiln temperature and the materials of the front and rear production links is constructed on the basis of energy conservation and mass conservation; then, the model parameters are identified by using the historical operation data of the rotary kiln, and the mechanism interpretability and data accuracy are considered; finally, the whole cement production process is summarized as multiple continuous links by taking a state task network as a framework, each link corresponds to a node, and the cement industry load production optimization scheduling is realized through material balance, start-stop and storage constraints; the application is applied to cement production control.

Description

Cement Industry Load Production Optimization Scheduling Method Considering Rotary Kiln Adjustability Technical Field

[0001] This invention provides a method for optimizing the production scheduling of cement industry load considering the adjustability of rotary kilns, belonging to the technical field of cement industry load optimization scheduling. Background Technology

[0002] Currently, cement producers generally use the new dry precalciner process to produce cement. Its main process is "two grindings and one calcination," where "one calcination" refers to high-temperature calcination in a rotary kiln. In the cement production process, the rotary kiln, as a core component, mainly functions to convert raw materials into clinker. It usually needs to be kept running continuously, only occasionally pausing to adjust inventory when cement demand decreases, or shutting down during planned maintenance. Moreover, in the cement production process, the rotary kiln is generally considered to be unadjustable, which limits the flexibility and adjustability potential of cement production. In order to improve production efficiency, it is necessary to start from the structure and process of the rotary kiln and study its optimization possibilities. One important method is to model the rotary kiln and optimize production based on the model.

[0003] Currently, modeling schemes for rotary kilns can be divided into mechanistic modeling and data modeling. Mechanistic modeling typically starts from the physicochemical reactions within the cement kiln, combining energy conservation, mass conservation, and kinetic laws to establish a mathematical model, which has good interpretability. However, the rotary kiln system has a complex structure, numerous parameters, and strong coupling. The modeling process often relies on empirical assumptions and simplifications, resulting in limited model accuracy and difficulty in accurately describing the relationship between energy consumption parameters and the actual production process, thus limiting its applicability. In contrast, data modeling utilizes historical data containing information reflecting the characteristics of the rotary kiln for data analysis and processing. It allows for modeling research without a complete understanding of the mechanism or any assumptions. The rapid development of statistical methods such as partial least squares, and artificial intelligence technologies such as support vector machines and artificial neural networks provides solid theoretical support for data modeling research.

[0004] Current research on production optimization for cement industry load has analyzed the energy consumption ratio of each stage of cement production from the perspective of load power consumption. However, the constructed models only consider the cement grinding stage and fail to reflect the overall load characteristics.

[0005] Some studies have divided the electrical equipment used in cement plants into two categories: adjustable loads and non-adjustable loads, and constructed an optimization model for adjustable load electricity consumption in cement plants based on time-of-use pricing. However, this model does not fully consider the coupling relationship between materials and processes in the cement production process, and it is difficult to accurately reflect the load response characteristics in actual production.

[0006] Another study proposed an economic scheduling model for cement plant production processes under a demand response mechanism. It uses a state-task network method to model the cement production process and clarifies the material coupling relationship and start-up / shutdown constraints between each link. However, this model treats the rotary kiln as an unadjustable load and fails to fully explore the adjustment potential of cement load.

[0007] Research has also been conducted on the load regulation characteristics of the cement industry, and a cement load model based on a three-level load classification (Level I, II, and III) has been proposed. Although the model divides the rotary kiln into three adjustable load levels, it assumes that the rotary kiln does not participate in any load regulation during daily operation, which makes it difficult to reflect the adjustment potential and response capability of the rotary kiln in actual operation.

[0008] To address the problems existing in the above research, there is an urgent need to propose a cement industry load production optimization scheduling method that can fully consider the adjustability of rotary kilns, so as to more accurately tap the adjustment potential of cement load and thus improve the feasibility and accuracy of cement enterprises participating in optimization scheduling. Summary of the Invention

[0009] To address the technical problems existing in the background art, the present invention provides a method for optimizing production scheduling of cement industry loads, considering the adjustability of rotary kilns, comprising the following production optimization scheduling steps:

[0010] Step 1: Model the rotary kiln using a combination of mechanistic and data analysis:

[0011] Using the calcination zone temperature, pulverized coal injection rate, raw meal rate, and clinker rate as coupled variables, a mass-energy conservation mechanism equation is established under the constraint of a set temperature.

[0012] Verifying the adjustability of the rotary kiln:

[0013] Based on the principles of energy conservation and mass conservation, a mechanistic model of the coupling relationship between rotary kiln temperature and materials in the preceding and following production stages is constructed.

[0014] Step 2: Based on the historical operating data of the rotary kiln, the particle swarm optimization algorithm is used to identify the parameters of the comprehensive coefficients in the mechanism model, and the adjustable model of the rotary kiln is obtained.

[0015] Step 3: Construct a load state task network model for the cement industry, summarize the entire cement production process into multiple continuous links, and analyze the characteristics of the main equipment in each production link and the material coupling relationship between the links before and after them.

[0016] Step 4: Under the framework of the cement industry load state task network model, based on the material balance constraints, production process constraints, running time constraints, equipment start-up and shutdown constraints, storage constraints, output constraints, and rotary kiln model constraints of each stage of cement industry load production, the obtained adjustable rotary kiln model is embedded to couple the rotary kiln temperature fluctuation with the materials and power consumption of the preceding and following stages. With minimizing the power consumption cost as the objective function, a complete constraint set for the cement industry load production optimization scheduling method is constructed.

[0017] The specific method for establishing the mass-energy conservation mechanism equation in step 1 is as follows:

[0018] The established mass conservation mechanism equation is as follows:

[0019] ;

[0020] In the formula, M H M1 is the mass of the material inside the rotary kiln; M2 is the unit flow rate of raw meal flowing into the rotary kiln; M3 is the unit flow rate of pulverized coal flowing into the kiln through the pulverized coal injection pipe at the kiln head; M4 is the unit flow rate of high-temperature clinker flowing from the kiln head into the grate cooler; M5 is the mass of the material inside the rotary kiln. S The unit flow rate lost due to a series of chemical reactions occurring within the rotary kiln;

[0021] The method for establishing the energy conservation mechanism equation is as follows:

[0022] Calculation formula based on heat:

[0023] ;

[0024] In the formula, Q represents heat; C represents specific heat capacity; m represents mass; and T represents temperature.

[0025] Based on the energy conservation equation:

[0026] ;

[0027] In the formula, Q H Q1 is the unit heat of the material inside the rotary kiln; Q2 is the unit heat of the injected pulverized coal; Q3 is the unit heat of the raw meal fed into the rotary kiln; Q4 is the unit heat of the secondary air flowing into the rotary kiln from the grate cooler; Q5 is the heat released by the combustion of pulverized coal in the rotary kiln; Q6 is the 2CaO·S generated per unit time. i The heat released by O2; Q6 is the amount of tricalcium silicate (3CaO·S) generated per unit time in the firing zone. i The heat absorbed by O2; Q7 is the heat absorbed and consumed by CaCO3 during decomposition per unit time; Q8 is the unit heat of the high-temperature gas flowing into the decomposition furnace from the kiln tail flue; Q9 is the heat of the high-temperature clinker flowing into the grate cooler from the kiln head.

[0028] Furthermore, due to:

[0029] ;

[0030] Combining the above formulas, and simplifying and rearranging them, we get:

[0031] ;

[0032] In the formula, s is the Lagrange operator, and the parameters A, B, C, D and a are comprehensive coefficients derived from the mechanism of many variables that affect the heat-material coupling relationship of the rotary kiln.

[0033] The specific method for parameter identification of the comprehensive coefficients in the mechanistic model using the particle swarm optimization algorithm in step 2 is as follows:

[0034] By performing an inverse Lassian transformation on the mechanistic model formula, the original differential equation is rewritten as an explicit algebraic relation, which can then be used in optimized scheduling. The calculation formula for the transformation is as follows:

[0035] ;

[0036] In the formula, A, B, C, and a are the unknown coefficients to be identified; E ss This is a compensation error term used to compensate for deviations caused by model simplification;

[0037] Define the time interval for monitoring historical operating data of the rotary kiln as Δt = 1h. Divide the above equation backwards using difference equations, and rearrange to obtain the difference equation that can be directly used for parameter identification:

[0038] ;

[0039] In an N-dimensional search space, there is a population of m particles, where the position of the i-th particle is represented as a vector:

[0040] ;

[0041] Velocity is an N-dimensional vector:

[0042] ;

[0043] The optimal position experienced by the i-th particle at the current moment can be expressed as:

[0044] ;

[0045] The formulas for calculating particle velocity and position updates are as follows:

[0046] ;

[0047] ;

[0048] In the formula, m and N are learning factors; c1 and c2 are positive constants; and r1 and r2 are random numbers that follow a uniform distribution on [0,1].

[0049] v id (t) represents the current velocity of the i-th particle, x id (t) represents the current position of the i-th particle, p id (t) represents the optimal position of the particle at the current moment, p gd (t) represents the optimal position found by the entire particle swarm at the current moment;

[0050] The particle then updates and determines its optimal position using the following formula:

[0051] ;

[0052] ;

[0053] The particle updates and determines its global optimal position using the following formula:

[0054] .

[0055] The material balance constraints used in step 4 to construct the complete constraint set are as follows:

[0056] The available quantity of material in cement production at time t is calculated from the quantity of the material generated at time t, the quantity consumed at time t, and the available quantity at time t-1. The calculation formula is as follows:

[0057] ;

[0058] In the formula, M(j,t) is the quantity of material stored in the j-th stage at time t, and P(j,t) is the power consumption of the equipment in the j-th stage at time t; λ j This represents the operating condition coefficient of the j-th stage, which can characterize the mathematical relationship between electricity consumption and material quantity.

[0059] The specific production process constraints used in step 4 to construct the complete constraint set are as follows:

[0060] The operation of the next stage of cement production will be affected by P(j,t) and M(j,t), meaning that P(j+1,t) needs to satisfy the following constraints:

[0061] ;

[0062] The above formula represents the constraint on whether the equipment corresponding to the P(j+1,t) link is working:

[0063] If the output of the j-th stage in time period t is not less than the output of the (j+1)-th stage in time period t, or if the former is less than the latter but the storage in the previous time period is sufficient for production in this time period, then the device corresponding to P(j+1,t) can be operated; otherwise, the device corresponding to P(j+1,t) will stop working in this time period.

[0064] The runtime constraints used in step 4 to construct the complete constraint set are as follows:

[0065] Define the maximum daily operating hours of each piece of equipment in the cement production process as t. ohd In other words, the operating time of each device in the study cannot exceed the maximum daily operating hours t corresponding to the device. ohd The expression is:

[0066] ;

[0067] In the formula, t(j) represents the daily operating time of device j; t ohd (j) represents the maximum daily operating time of device j;

[0068] The specific start / stop constraints adopted are as follows:

[0069] Considering equipment start-up and shutdown times, equipment in different stages needs to maintain a minimum downtime / run time of 30 minutes, expressed as:

[0070] ;

[0071] In the formula, i represents the cumulative time for device j to start / stop; x _j,t This indicates the start / stop status of device j at time t, where 1 indicates start and 0 indicates stop.

[0072] The storage constraints used in step 4 to construct the complete constraint set are as follows:

[0073] The product of each stage of a cement plant is stored in warehouses, as expressed in the following expression:

[0074] ;

[0075] In the formula, η lb and η ub These are the lower limit coefficient and upper limit coefficient for storage capacity, respectively.

[0076] The specific production constraints used in step 4 to construct the complete constraint set are as follows:

[0077] The daily cement production must meet the production requirements, expressed as:

[0078] M F,P -M I,P≥M D,P,index ;

[0079] In the formula, M F,P and M I,P These represent the quantities of cement produced at the end of the scheduling period and the beginning of the scheduling period, respectively; M D,P,index This refers to the daily production requirements for cement.

[0080] The rotary kiln model constraints used in step 4 to construct the complete constraint set are as follows:

[0081] Based on the obtained adjustable rotary kiln model, the unit consumption M1, M2, and M3 of raw materials, coal, and clinker correspond to M(2,t), M(4,t), and M(5,t) in the material balance, respectively. Simultaneously, the materials are replaced with the corresponding equipment power consumption relationships, expressed as:

[0082] ;

[0083] The above formula reflects the adjustable load range of the rotary kiln itself, and couples the entire cement production process together in load optimization scheduling, verifying the adjustability of the rotary kiln.

[0084] The calculation method for step 4, which uses minimizing electricity costs as the objective function, is as follows:

[0085] With the objective function of minimizing electricity costs, the expression is:

[0086] ;

[0087] In the formula, F is the electricity purchase cost of the cement plant; T is the total number of time periods in a 24-hour day; P(D,t) is the power consumption of the cement production load in time period t; δ t Let t be the electricity price for time period t; Δt be the length of time period t.

[0088] The advantages of this invention compared to existing technologies are as follows: This invention provides a cement industry load production optimization scheduling method that considers the adjustability of rotary kilns. It mainly uses a combination of mechanism and data to model the rotary kiln: first, the energy-mass conservation mechanism is used to characterize the kiln reaction; then, the particle swarm optimization algorithm combined with historical operating data is used to identify the model parameters. The model output provides a quantitative coupling range between raw material, pulverized coal, clinker, and kiln temperature using explicit algebraic relationships, thus confirming the adjustability of the rotary kiln. This invention constructs the rotary kiln as an adjustable model and embeds it into the cement load framework, thereby overcoming the shortcomings of traditional cement load models that treat the rotary kiln as a constant load, ignore its adjustability potential, and fail to reflect the material-electricity coupling relationship between its upstream and downstream processes. This invention fully characterizes the adjustability characteristics of the load throughout the entire cement production process, enabling cement enterprises to participate more accurately in actual production optimization scheduling, thereby significantly improving the flexibility and adaptability of cement production. Attached Figure Description

[0089] The present invention will be further described below with reference to the accompanying drawings:

[0090] Figure 1 is a schematic diagram of the distribution of the reaction zone in the rotary kiln of the present invention;

[0091] Figure 2 is a schematic diagram of the STN model of the cement production process of the present invention;

[0092] Figure 3 is a flowchart of the steps of the cement industry load production optimization scheduling method of the present invention;

[0093] Figure 4 is a comparison chart of the historical values ​​of rotary kiln temperature and the calculated values ​​of the model in an embodiment of the present invention;

[0094] Figure 5 is a comparison of the load curves before and after the model established in this embodiment of the invention participates in demand response. Detailed Implementation

[0095] As shown in Figures 1 to 3, this invention addresses the shortcomings of the existing cement industry load, which generally treats rotary kilns as constant loads and ignores their adjustable potential. It proposes a cement industry load production optimization scheduling method that considers the adjustability of rotary kilns, helping cement production enterprises to participate in production scheduling more accurately and thereby improving the flexibility of cement production.

[0096] This invention first models the rotary kiln using a combination of mechanism and data to verify its adjustability. Based on energy and mass conservation, a mechanistic model of the coupling relationship between rotary kiln temperature and materials in the preceding and following production stages is constructed. Then, historical operating data of the rotary kiln is used to identify the model parameters, taking into account both the interpretability of the mechanism and the accuracy of the data. Finally, using a State Task Network (STN) framework, the entire cement production process is summarized into six continuous stages, each corresponding to a "state-task" node. Through constraints such as material balance, start-up, shutdown, and storage, the optimal scheduling of cement industry load production considering the adjustability of the rotary kiln is achieved.

[0097] Furthermore, to achieve the above objectives, the present invention specifically adopts the following technical solution:

[0098] First, the parameters of the rotary kiln and the mechanism modeling were selected. The rotary kiln is a continuous rotating long cylindrical equipment with an inclination angle of about 3% to 6%. The raw material completes carbonate decomposition, solid-phase reaction and liquid-phase reaction in the decomposition zone, transition zone, calcination zone and cooling zone of the rotary kiln in sequence. The area and temperature of different reaction zones are shown in Figure 1.

[0099] This step requires verifying the adjustability of the rotary kiln. Through mechanism and data modeling, specifically using the firing zone temperature, pulverized coal injection rate, raw material quantity, and clinker quantity as coupling variables, a mass-energy conservation mechanism equation is established under the constraint of a set temperature. The parameters are then identified using the particle swarm optimization algorithm to obtain an adjustable model of the rotary kiln.

[0100] Current modeling of rotary kilns focuses on studying the mapping relationship between the temperature of the firing zone and its production process parameters, lacking a systematic characterization of the material coupling between its upstream and downstream stages. This invention improves upon this by selecting key parameters that can simultaneously reflect the rotary kiln temperature and the material flow coupling between its upstream and downstream stages, thus connecting the three stages of raw material grinding, cement grinding, and fuel grinding related to the rotary kiln in cement production.

[0101] The key parameters for selecting the rotary kiln model include:

[0102] Firing zone temperature: The firing zone is the core area of ​​the rotary kiln. The clinker content produced therein determines the quality and pass rate of the final clinker product. The clinker content is also significantly related to the firing zone temperature. In order to ensure the quality of the produced clinker, the firing zone temperature must be controlled within the specified range. Generally, the firing zone temperature reflects the rotary kiln temperature.

[0103] Coal injection rate: The coal injection rate directly determines the heat supply of the rotary kiln. The heat energy required for the chemical reaction of raw materials in the rotary kiln and the high temperature required for the raw materials to burn into clinker both come from the heat released by the combustion of pulverized coal injected into the kiln from the coal injection pipe at the kiln head. This is a direct factor affecting the temperature of the burning zone.

[0104] Raw meal feed rate: When raw meal enters the kiln, it encounters the counter-current high-temperature hot air and undergoes convective heat transfer, absorbing heat from the kiln. Increasing the amount of raw meal will cause the kiln temperature to decrease. If the amount of raw meal continues to increase, the temperature will continue to decrease, affecting the quality of clinker firing.

[0105] Clinker output: Clinker is both the final product of the chemical reaction in the firing zone and the "exit" for heat balance within the kiln. Its instantaneous output directly reflects the firing efficiency and heat utilization level. A steady increase in output with a constant kiln temperature indicates high system output; a sudden decrease in output and a surge in kiln temperature can easily lead to overfiring; excessive output and insufficient heat compensation can result in underfiring. Therefore, clinker output is not only a direct indicator of firing quality but also a key feedback variable for closed-loop regulation of kiln temperature, coal quantity, and raw material feed rate.

[0106] The established mass conservation mechanism equation is as follows:

[0107] ;

[0108] In the formula, M H M1(kg / s) represents the mass of the material inside the rotary kiln; M1(kg / s) represents the unit flow rate of the raw material flowing into the rotary kiln; M2( M3 (kg / s) is the unit flow rate of pulverized coal flowing into the kiln through the pulverized coal injection pipe at the kiln head; M3 (kg / s) is the unit flow rate of high-temperature clinker flowing from the kiln head into the grate cooler; M S (kg / s) represents the unit flow rate lost due to a series of chemical reactions occurring within the rotary kiln.

[0109] The method for establishing the energy conservation mechanism equation is as follows:

[0110] The commonly used formula for heat is:

[0111] ;

[0112] In the formula, Q represents heat, in J; C represents specific heat capacity, in J·kg. -1 ·K -1 m represents mass, with the unit being kg; T represents temperature, with the unit being Kelvin (K).

[0113] Energy conservation equation:

[0114] ;

[0115] In the formula, Q H Q1(J / s) represents the unit heat of the material inside the rotary kiln; Q2(J / s) represents the unit heat of the injected pulverized coal; Q3(J / s) represents the unit heat of the raw meal fed into the rotary kiln; Q4(J / s) represents the unit heat of the secondary air flowing into the rotary kiln from the grate cooler; Q5(J / s) represents the heat released by the combustion of pulverized coal in the rotary kiln; Q6(J / s) represents the 2CaO·S generated per unit time. i The heat released by O2; Q6 (J / s) is the amount of tricalcium silicate (3CaO·S) generated per unit time in the firing zone. i The heat absorbed by O2; Q7 (J / s) is the heat absorbed and consumed by CaCO3 in the decomposition zone per unit time; Q8 (J / s) is the unit heat of the high-temperature gas flowing into the decomposition furnace from the kiln tail flue; Q9 (J / s) is the heat of the high-temperature clinker flowing into the grate cooler from the kiln head.

[0116] And because:

[0117] ;

[0118] Combining the above formulas and undergoing a series of simplifications and rearrangements, we can obtain:

[0119] ;

[0120] In the formula, s is the Laplace operator, and parameters A, B, C, D, and a are comprehensive coefficients derived from the mechanism of numerous variables affecting the heat-material coupling relationship of the rotary kiln (such as the specific heat capacity of each substance, the dynamic combustion rate of pulverized coal, and molar mass). Except for the specific heat capacity, all coefficients are constants. Although the specific heat capacity is temperature-dependent, its variation is very small within the temperature constraints of the rotary kiln. Furthermore, in general engineering and industrial calculations such as cement rotary kilns, the specific heat capacity can be considered a constant. Therefore, within the allowable error range, parameters A, B, C, D, and a can all be regarded as constants, thus enabling parameter identification using historical operating data of the rotary kiln.

[0121] The adjustability of the rotary kiln was verified:

[0122] Research based on existing rotary kiln modeling results shows that the rotary kiln temperature (calcination zone temperature) can fluctuate steadily within the range of 1300℃–1450℃ without affecting normal production. Under the constraint of this temperature range, the model quantifies the dynamic coupling relationship between pulverized coal injection rate, raw meal feed rate, and clinker output rate, allowing these three to be adjusted within a certain range. This characteristic enables the rotary kiln not only to actively increase or decrease its own load within a certain range, but also to achieve load transfer by changing the power demand of upstream and downstream stages (raw meal mill, coal mill, cement mill). Compared with the traditional constant load assumption, this significantly expands the adjustment potential of cement production load in optimal scheduling, proving that the rotary kiln possesses adjustability.

[0123] Then, based on the historical operating data of the rotary kiln, the particle swarm optimization algorithm was used to identify the parameters of the comprehensive coefficients in the mechanism model.

[0124] To apply the adjustability of the rotary kiln to the scheduling model, the mechanistic model formula is transformed using an inverse Lassian transformation, rewriting the original differential equation as an explicit algebraic relation, which can then be used in optimized scheduling. The calculation formula for the transformation is as follows:

[0125] ;

[0126] In the formula, A, B, C, and a are the unknown coefficients to be identified; E ss To compensate for the error term, which is used to offset the bias caused by model simplification, and since the monitoring time interval for the historical operating data of the rotary kiln in this invention is Δt = 1 hour, the above formula is discretized by backward difference, and the resulting difference equation that can be directly used for parameter identification is:

[0127] ;

[0128] The basic principle of particle swarm optimization algorithm: In an N-dimensional search space, each particle is a potential solution to the optimization problem. Each particle has a velocity vector to change the direction and step size of each movement. Particles update their position and velocity by tracking two extreme values: one is the optimal value found by the particle itself at the current moment, i.e., the individual extreme value, and the other is the optimal value found by the population at the current moment, i.e., the global extreme value.

[0129] Suppose there is a population of m particles in an N-dimensional search space, where the position of the i-th particle is represented as a vector:

[0130] ;

[0131] Velocity is an N-dimensional vector:

[0132] ;

[0133] The optimal position that the i-th particle has reached at the current moment:

[0134] ;

[0135] The formulas for updating the particle's velocity and position are as follows:

[0136] ;

[0137] ;

[0138] In the formula, m and N are learning factors, c1 and c2 are positive constants, and r1 and r2 are random numbers uniformly distributed on [0,1]. id (t) represents the current velocity of the i-th particle, x id (t) represents the current position of the i-th particle, p id (t) represents the optimal position of the particle at the current moment, p gd (t) represents the optimal position found by the entire particle swarm at the current moment.

[0139] The particle updates and determines its optimal position using the following formula:

[0140] ;

[0141] ;

[0142] The particle's global optimal position is determined by updating the following formula:

[0143] ;

[0144] The particle swarm optimization algorithm is used to identify the parameters of the mechanism model. It can optimize the coupling coefficient in the mechanism equation within the kiln temperature range of 1300-1450℃ without making linearization assumptions for the multivariable, nonlinear, and highly hysteretic kiln reaction process. The algorithm has strong global search capabilities and fast convergence speed, and can identify model parameters based solely on historical operation monitoring data such as kiln temperature, raw material, pulverized coal, and clinker that can be measured on-site.

[0145] Finally, using State-Task Network (STN) as a framework, the constraints of each "state-task" in the load production optimization scheduling of the cement industry are analyzed. STN (State-Task Network) modeling method is an effective tool widely used in chemical and manufacturing fields to describe production processes, and can intuitively reflect complex production processes and resource utilization.

[0146] In the cement plant production process, the STN model is constructed based on its main production stages and material states, as shown in Figure 2. Cement production is divided into six stages according to the five storage facilities of the cement plant: raw material crushing, raw meal grinding, clinker calcination, cement grinding, fuel grinding, and packaging and transportation. Throughout the production process, materials such as raw materials, raw meal, clinker, pulverized coal, and cement undergo transformations from one state to another in different processes.

[0147] The main operating equipment of each process described above was analyzed to determine its operating characteristics and load adjustability, as shown in Table 1 below.

[0148] Table 1. Equipment and characteristics of each process in a cement plant

[0149]

[0150] It can be seen that among the main operating equipment in the six main links of the cement production process, the rotary kiln is a continuous load, while the raw material crusher, raw meal mill, cement mill, coal mill, and packaging and conveying machine are all equipment that can only be controlled to start and stop and cannot be flexibly adjusted in power, which are discrete loads.

[0151] This invention, based on various constraints at each stage of cement industry load production, embeds the obtained adjustable rotary kiln model, coupling rotary kiln temperature fluctuations with upstream and downstream material and power consumption. With minimizing electricity costs as the objective function, a complete constraint set for the cement industry load production optimization scheduling method is constructed. Specifically, each constraint includes:

[0152] Material balance constraints: According to the STN model of the cement production process in Figure 2, the materials in cement production are not only related to the stage in which they are produced, but will also be used as raw materials in the next stage. Therefore, the available quantity of materials in cement production at time t can be calculated from the quantity of materials produced at time t, the quantity consumed at time t, and the available quantity at time t-1, expressed as:

[0153] ;

[0154] In the formula, M(j,t) is the quantity of material stored in the j-th stage at time t, and P(j,t) is the power consumption of the equipment in the j-th stage at time t; λ j This represents the operating condition coefficient of the j-th stage, which can characterize the mathematical relationship between electricity consumption and material quantity.

[0155] Production process constraints: As can be seen from the STN model, the operation of the next stage of cement production will be affected by P(j,t) and M(j,t), that is, P(j+1,t) needs to satisfy the following constraints:

[0156] ;

[0157] The above formula represents the constraints on whether the equipment corresponding to stage P(j+1,t) works: if the output of stage j in time period t is not less than the output of stage j+1 in time period t, or if the former is less than the latter but the storage in the previous time period is sufficient for production in this time period, then the equipment corresponding to stage P(j+1,t) can run; otherwise, the equipment corresponding to stage P(j+1,t) will stop working in this time period.

[0158] Operating time constraints: The maximum daily operating hours specified for each major piece of equipment in the cement production process is t. ohd In other words, the operating time of each device in the study cannot exceed the maximum daily operating hours t corresponding to the device. ohd The expression is as follows:

[0159] ;

[0160] In the formula, t(j) represents the daily operating time of device j; t ohd (j) represents the maximum daily working time of device j.

[0161] Equipment start-up and shutdown constraints, considering equipment start-up and shutdown times, require equipment in different stages to maintain at least 30 minutes of downtime / running time, as expressed below:

[0162] ;

[0163] In the formula, i represents the cumulative time for device j to start / stop; x_j,t This indicates the start / stop status of device j at time t, where 1 indicates start and 0 indicates stop.

[0164] Storage constraints: In actual production, cement plants have storage facilities for the products of each stage of production. These facilities allow cement plants to freely control the activation sequence of various equipment in the production process, without having to strictly follow the production flow. This facilitates load transfer and improves the flexibility of the cement plant. The expression is:

[0165] ;

[0166] In the formula, η lb and η ub These are the lower limit coefficient and upper limit coefficient for storage capacity, respectively.

[0167] Production constraints: The premise for the cement industry load to participate in optimal scheduling is not to affect normal cement production, that is, the daily cement output needs to meet the production requirements, as expressed below:

[0168] M F,P -M I,P ≥M D,P,index ;

[0169] In the formula, M F,P and M I,P These represent the quantities of cement produced at the end of the scheduling period and the beginning of the scheduling period, respectively; M D,P,index This refers to the daily production requirements for cement.

[0170] The rotary kiln model constraints, based on the adjustable model obtained in the second step, assign unit consumption values ​​M1, M2, and M3 for raw materials, coal, and clinker, respectively, to M(2,t), M(4,t), and M(5,t) in the material balance. Simultaneously, the materials are replaced with their corresponding equipment power consumption relationships, expressed as follows:

[0171] ;

[0172] The above formula not only intuitively reflects the adjustable load range of the rotary kiln itself, but also for the first time incorporates the power demand of upstream and downstream processes such as raw material mill, coal mill, and cement mill into the same optimization framework, thereby coupling the entire cement production process in load optimization scheduling and directly verifying the adjustability of the rotary kiln.

[0173] The objective function of cement plant load production scheduling optimization is to transfer the cement production load as much as possible while ensuring normal plant production. This invention uses minimizing electricity costs as the objective function, and its expression is as follows:

[0174] ;

[0175] In the formula, F is the electricity purchase cost of the cement plant; T is the total number of time periods in a 24-hour day; P(D,t) is the power consumption of the cement production load in time period t; δ t Let t be the electricity price for time period t; Δt be the length of time period t.

[0176] In an embodiment of the present invention, parameter identification is performed on the rotary kiln model. Taking a new type of dry process cement production industrial user as the object, 30 sets of historical monitoring data of its rotary kiln load are selected, as shown in Table 2 below.

[0177] Table 2. Rotary kiln production data for each group

[0178]

[0179] Based on the 30 sets of historical operation data monitored in Table 2, the firing zone temperature TH was used as the verification target. The coupling coefficients A, B, C, a and the compensation error term Ess in the mechanism equation were identified in one go using the particle swarm optimization algorithm. The simulation results are shown in Figure 4.

[0180] The load production optimization scheduling method for the cement industry, which considers the adjustability of rotary kilns, provided by this invention was simulated and verified under a 24-hour time-of-use electricity price scenario. The load curves before and after the optimization scheduling are shown in Figure 5. In the figure, "traditional method" refers to the case where the rotary kiln is regarded as a constant load and does not participate in the regulation.

[0181] Compared with traditional methods, the production optimization scheduling method provided by this invention—which incorporates rotary kilns into adjustable loads—significantly expands the optimization scheduling space for cement industry load production and further enhances load transfer capabilities. Simulation results show that peak electricity consumption of cement enterprises can be reduced to about 24% of the original level, while off-peak electricity consumption increases accordingly; the electricity cost of cement plants is reduced by 27%.

[0182] For cement production enterprises, the cement industry load production optimization scheduling method provided by this invention can greatly reduce electricity costs without affecting cement production, and has extremely high implementation value. In addition, during the implementation process, it can also perform peak shaving and valley filling of the power grid load, greatly improving the stability of the power grid.

[0183] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing production scheduling in the cement industry considering the adjustability of rotary kilns, characterized by: The production optimization scheduling steps include the following: Step 1: Modeling the rotary kiln using a combination of mechanism and data: Using the firing zone temperature, pulverized coal injection rate, raw material quantity, and clinker quantity as coupled variables, a mass-energy conservation mechanism equation is established under the constraint of a set temperature. The specific method is as follows: The established mass conservation mechanism equation is: In the formula, M H M1 represents the mass of the material inside the rotary kiln; M2 represents the unit flow rate of the raw material flowing into the rotary kiln; M3 represents the unit flow rate of the pulverized coal flowing into the kiln through the pulverized coal injection pipe at the kiln head; and M4 represents the unit flow rate of the high-temperature clinker flowing from the kiln head into the grate cooler. M S The unit flow rate lost due to a series of chemical reactions occurring within the rotary kiln; The method for establishing the energy conservation mechanism equation is based on the calculation formula for heat: In the formula, Q represents heat; C represents specific heat capacity. m represents mass; T represents temperature; based on the energy conservation equation: In the formula, Q H Q1 is the unit heat of the material inside the rotary kiln; Q2 is the unit heat of the injected pulverized coal; Q3 is the unit heat of the raw meal fed into the rotary kiln; Q4 is the unit heat of the secondary air flowing into the rotary kiln from the grate cooler; Q5 is the heat released by the combustion of pulverized coal in the rotary kiln; Q6 is the 2CaO·S generated per unit time. i The heat released by O2; Q6 is the amount of tricalcium silicate (3CaO·S) generated per unit time in the firing zone. i The heat absorbed by O2; Q7 is the heat absorbed and consumed by CaCO3 during decomposition per unit time; Q8 is the unit heat of the high-temperature gas flowing into the decomposition furnace from the kiln tail flue; Q9 is the heat of the high-temperature clinker flowing into the grate cooler from the kiln head; and because: Combining the above formulas, simplifying and rearranging them, we get: In the formula, s is the Laplace operator, and parameters A, B, C, D, and a are comprehensive coefficients derived from the mechanism of numerous variables affecting the heat-material coupling relationship of the rotary kiln. Then, based on energy conservation and mass conservation, a mechanism model of the coupling relationship between the rotary kiln temperature and the materials in the preceding and following production stages is constructed. Step 2: Based on the historical operating data of the rotary kiln, the particle swarm optimization algorithm is used to identify the parameters of the comprehensive coefficients in the mechanism model, and an adjustable model of the rotary kiln is obtained. The specific method is as follows: by performing an inverse Laplace transformation on the mechanism model formula, the original differential equation is rewritten as an explicit algebraic relationship, which can then be used in the optimization scheduling. The calculation formula for the transformation is: In the formula, A, B, C, and a are the unknown coefficients to be identified; E ss To compensate for the error term and offset the bias caused by model simplification, the time interval for monitoring historical operating data of the rotary kiln is defined as Δt = 1 hour. The above equation is then discretized backwards, and the resulting difference equation, which can be directly used for parameter identification, is: Step 3: Construct a cement industry load state task network model, summarizing the entire cement production process into multiple continuous stages, and analyzing the main equipment characteristics of each production stage and the material coupling relationships between the stages before and after them; Step 4: Under the framework of the cement industry load state task network model, based on the material balance constraints, production stage constraints, running time constraints, equipment start-up and shutdown constraints, storage constraints, output constraints, and rotary kiln model constraints of each stage of cement industry load production, embed the obtained adjustable rotary kiln model, coupling the rotary kiln temperature fluctuation with the materials and power consumption of the stages before and after, and using minimizing power consumption as the objective function, construct a complete constraint set for the cement industry load production optimization scheduling method, where: the material balance constraint used to construct the complete constraint set is specifically: the available quantity of material in cement production at time t is calculated from the quantity generated at time t, the quantity consumed at time t, and the available quantity at time t-1, and the calculation formula is: In the formula, M(j,t) is the quantity of material stored in the j-th stage at time t, and P(j,t) is the power consumption of the equipment in the j-th stage at time t; λ j The condition coefficient of the j-th stage represents the mathematical relationship between electricity consumption and material quantity. The specific production stage constraints used to construct the complete constraint set are as follows: the working condition of the next stage of cement production will be affected by P(j,t) and M(j,t), that is, P(j+1,t) needs to satisfy the following constraints: The above formula represents the constraint on whether the equipment corresponding to stage P(j+1,t) is working: if the output of stage j in time period t is not less than the output of stage j+1 in time period t, or if the former is less than the latter but the storage in the previous time period is sufficient for production in this time period, then the equipment corresponding to P(j+1,t) can be operated; otherwise, the equipment corresponding to P(j+1,t) will stop working in this time period. The rotary kiln model constraints used to construct the complete constraint set are as follows: based on the obtained adjustable rotary kiln model, the unit consumption M1, M2, and M3 of raw material, coal, and clinker correspond to M(2,t), M(4,t), and M(5,t) in the material balance, respectively. At the same time, the materials are replaced with the corresponding equipment power consumption relationship, and the expression is: The above formula reflects the adjustable load range of the rotary kiln itself, and couples the entire cement production process together in load optimization scheduling, verifying the adjustability of the rotary kiln.

2. The cement industry load production optimization scheduling method considering the adjustability of rotary kilns according to claim 1, characterized in that: The specific method for parameter identification of the comprehensive coefficients in the mechanism model using the particle swarm optimization algorithm in step 2 is as follows: Define an N-dimensional search space with m particles forming a swarm, where the position of the i-th particle is represented as a vector: Velocity is an N-dimensional vector. The optimal position experienced by the i-th particle at the current moment is expressed as: The formulas for calculating the particle's velocity and position updates are as follows: ； In the formula, m and N are learning factors; c1 and c2 are positive constants; r1 and r2 are random numbers uniformly distributed on [0,1]; v id (t) represents the current velocity of the i-th particle, x id (t) represents the current position of the i-th particle, p id (t) represents the optimal position of the particle at the current moment, p gd (t) represents the optimal position found by the entire particle swarm at the current moment; then the particle updates and determines its own optimal position using the following formula: ； The particle updates and determines its global optimal position using the following formula: 。 3. The cement industry load production optimization scheduling method considering the adjustability of rotary kilns according to claim 2, characterized in that: The runtime constraint used in step 4 to construct the complete constraint set is specifically defined as follows: the maximum daily operating hours of each piece of equipment in the cement production process is defined as t. ohd In other words, the operating time of each device in the study cannot exceed the maximum daily operating hours t corresponding to the device. ohd The expression is: In the formula, t(j) represents the daily operating time of device j; t ohd (j) represents the maximum daily operating time of equipment j; the specific equipment start-up and shutdown constraints are as follows: considering equipment start-up and shutdown times, equipment in different stages needs to maintain at least 30 minutes of downtime / running time, expressed as: In the formula, i represents the cumulative time for device j to start / stop; x _j,t This indicates the start / stop status of device j at time t, where 1 indicates start and 0 indicates stop.

4. The cement industry load production optimization scheduling method considering the adjustability of rotary kilns according to claim 3, characterized in that: The storage constraint used in step 4 to construct the complete constraint set is as follows: It is defined that the products of each stage of the cement plant are stored in warehouses, expressed as: In the formula, η lb and η ub These are the lower limit coefficient and upper limit coefficient for storage capacity, respectively.

5. The cement industry load production optimization scheduling method considering the adjustability of rotary kilns according to claim 4, characterized in that: In step 4, the production constraint used to construct the complete constraint set is as follows: the daily cement production must meet the production requirements, expressed as: In the formula, M F,P and M I,P These represent the quantities of cement products at the end of the scheduling period and the beginning of the scheduling period, respectively; C D,P,index This refers to the daily production requirements for cement.

6. The cement industry load production optimization scheduling method considering the adjustability of rotary kilns according to claim 5, characterized in that: The calculation method for minimizing electricity cost as the objective function in step 4 is as follows: The expression for minimizing electricity cost as the objective function is: In the formula, F represents the electricity purchase cost of the cement plant; T represents the total number of time periods in a 24-hour day; P(D,t) represents the power consumption of the cement production load during time period t; δ t Let t be the electricity price for time period t; Δt be the length of time period t.

Images