A method for optimizing decision of coal blending and burning proportion in thermal power plant based on feasible region model

Abstract

The present application relates to the technical field of thermal power plant operation optimization, and discloses a method for optimizing the blending ratio of coal blending in a thermal power plant based on a feasible region model, comprising the following steps: Step 1, constructing a basic model of the feasible region of the blending ratio of coal blending; Step 2, determining a first type of feasible region boundary function based on the constraint conditions of coal blending, and determining a second type of feasible region boundary function based on the constraint condition of the total blending ratio of coal; the constraint conditions of coal blending include the calorific value, volatile matter, sulfur content, moisture content, ash content and single coal ratio of coal blending; Step 3, determining the feasible region of the blending ratio of coal blending under the condition of multiple coal types by using a space projection algorithm based on vertex search; Step 4, setting a target function, and outputting the optimal blending ratio based on the feasible region of the blending ratio of coal blending. The present application can realize the rapid calculation of the optimal blending ratio and the rapid adjustment of the blending ratio in the face of sudden changes in working conditions, and can assist the operation personnel of the thermal power plant in making the optimal decision on the blending ratio in real time and accurately.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of thermal power plant operation optimization, and in particular to a method for optimizing the blending ratio of coal blending in a thermal power plant based on a feasible region model. BACKGROUND

[0002] Traditional coal-fired power enterprises are facing unprecedented challenges. On the supply side, the installed capacity and on-grid power of clean energy such as wind power and photovoltaic power continues to grow, squeezing the market share of coal-fired power enterprises. On the cost side, the tight balance situation in the coal market leads to long-term fluctuations in coal prices, and the electricity consumption of coal-fired power units under low load conditions increases significantly, further compressing the profit space of coal-fired power enterprises and exacerbating the potential loss risk. The above factors have continuously put pressure on the overall profitability of coal-fired power enterprises, and in this context, effective control of power generation costs has become the key to improving the market competitiveness of coal-fired power enterprises.

[0003] Statistical data shows that the coal cost of a thermal power plant usually accounts for more than 70% of the entire power generation cost, so controlling the cost of coal-fired power is the most effective way to reduce the power generation cost of coal-fired power enterprises. Common coal cost control techniques include coal blending, coal washing and selection, etc. In view of the current situation of coal resource distribution and market situation in China, coal blending has become the main means for most domestic thermal power plants to reduce coal costs. As a flexible fuel management technique, coal blending is a technique of blending different types and qualities of coal in a certain proportion to form mixed coal, so as to optimize the chemical composition, physical properties and combustion characteristics of the coal, achieve coal quality complementation, and thus improve power generation efficiency and reduce coal consumption cost of thermal power plants.

[0004] The core of the coal blending technique is to achieve rapid and accurate decision-making of the blending ratio, while ensuring the safe operation of the boiler and auxiliary system, and taking into account the goals of improving combustion efficiency and controlling pollutant emissions. However, although existing research on the optimization of coal blending ratio has achieved certain results, there are still the following two shortcomings: 1) Existing methods mostly focus on obtaining a single optimal solution, only output the optimal blending ratio under a specific operating condition, and fail to provide complete distribution information of the adjustable blending ratio under this operating condition, making it difficult to provide sufficient operating optimization and emergency decision-making basis for operating personnel. 2) When the operating condition (such as environmental protection requirements) changes suddenly, the existing method needs to rebuild the optimization model and solve it by intelligent algorithm, which takes a long time to calculate, making it difficult to generate a feasible coal blending scheme to guide the switching of coal types in time, and unable to meet the rapid response demand in actual production when the operating condition changes suddenly. SUMMARY

[0005] The present application aims to provide a method for optimizing the blending ratio of coal blending in a thermal power plant based on a feasible region model, which can realize rapid calculation of the optimal blending ratio and rapid adjustment of the blending ratio for sudden changes in operating conditions, and assist the operating personnel of the thermal power plant in making optimization decisions on the blending ratio in real time and accurately.

[0006] The basic scheme provided by the application is a coal blending and burning proportion optimization decision method for a thermal power plant based on a feasible region model, comprising the following steps:

[0007] Step 1, constructing a basic model of a coal blending and burning proportion feasible region;

[0008] Step 2, determining a first type of feasible region boundary function based on a coal blending and burning constraint condition, and determining a second type of feasible region boundary function based on a total coal blending proportion constraint condition; the coal blending and burning constraint condition comprises a coal blending calorific value, a coal blending volatile matter, a coal blending sulfur content, a moisture content, an ash content and a single coal type proportion;

[0009] Step 3, determining a coal blending and burning proportion feasible region under a multi-coal type condition by using a space projection algorithm based on vertex search;

[0010] Step 4, setting a target function, and outputting an optimal blending and burning proportion based on the coal blending and burning proportion feasible region.

[0011] The working principle and advantages of the application are as follows:

[0012] The coal blending and burning proportion optimization decision method for a thermal power plant based on a feasible region model provided by the application, in view of the problems raised in the background art, innovatively proposes an optimization decision method based on a feasible region model by referring to the domain model analysis idea of an electric power system, can realize rapid calculation of an optimal blending and burning proportion and rapid adjustment of a blending and burning proportion facing a working condition mutation, can provide a new technical path for coal blending and burning optimization of a thermal power plant, and has high practical application benefits. The focus is on:

[0013] The present scheme refers to the analysis paradigm of a power system security domain, converts a complex multi-constraint optimization problem into a feasible solution domain search problem in a geometric space by constructing a coal blending and burning proportion feasible region model, makes the optimization process have a strict theoretical basis, and can accurately and efficiently find an optimal solution. Among them, the space projection algorithm based on vertex search can reduce a multi-dimensional constraint space to an analyzable dimension, and the extreme points of polyhedron vertices are used to represent the boundary of the feasible region, compared with traditional simplex method and other optimization algorithms, the calculation efficiency is improved.

[0014] And, the scheme realizes the systematic integration of the full-factor constraint of the coal blending and burning by constructing a double boundary function system. The first type of boundary function covers the core coal quality indexes such as the calorific value, volatile matter, sulfur content, moisture content and ash content, and the influence mechanism of the blending ratio of different coals on the burning efficiency, pollutant emission and equipment safety is accurately reflected by establishing a linear weighted model of each index. Taking the volatile matter constraint as an example, the quantitative relationship model between the volatile matter content and the combustion stability is established, which can effectively prevent the combustion oscillation problem caused by too low volatile matter. The second type of boundary function focuses on the blending ratio constraint of the coals, including the engineering limit of the upper limit of the single coal blending and burning and the normalization requirement of the total blending ratio. Through the integration of multi-dimensional constraints, the optimization results can meet multiple goals such as environmental emission, boiler efficiency and equipment life, and the comprehensiveness and accuracy of the optimization decision are improved.

[0015] In addition, in specific application, the scheme can update the constraint conditions and reconstruct the feasible region boundary online by collecting the coal quality data and related index requirements in real time. When the running condition changes suddenly (such as the interruption of the supply of a certain coal), the feasible region range can be determined again in a short time based on the vertex search algorithm, the blending ratio of the coal blending and burning is quickly adjusted, and the alternative coal blending scheme is quickly generated. The scheme has certain dynamic adjustment property, can break through the limitation of the traditional static optimization model, and makes the blending ratio decision have the adaptive feature. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 It is a method flowchart of an embodiment of the present application of a coal blending and burning ratio optimization decision method for a thermal power plant based on a feasible region model;

[0017] Figure 2 It is a coal blending and burning ratio feasible region schematic diagram of an embodiment of the present application of a coal blending and burning ratio optimization decision method for a thermal power plant based on a feasible region model;

[0018] Figure 3 It is a two-type blending ratio feasible region boundary modeling schematic diagram of an embodiment of the present application of a coal blending and burning ratio optimization decision method for a thermal power plant based on a feasible region model;

[0019] Figure 4 It is a blending ratio feasible region construction flowchart based on vertex search of an embodiment of the present application of a coal blending and burning ratio optimization decision method for a thermal power plant based on a feasible region model;

[0020] Figure 5 It is a rapid solution flowchart of the optimal blending ratio based on the coal blending and burning ratio feasible region of an embodiment of the present application of a coal blending and burning ratio optimization decision method for a thermal power plant based on a feasible region model;

[0021] Figure 6A working condition mutation-oriented coal blending and burning proportion rapid adjustment method schematic diagram of an embodiment of the coal blending and burning proportion optimization decision method for thermal power plants based on a feasible region model of the present application;

[0022] Figure 7 A feasible region model-based thermal power plant coal blending and burning proportion optimization decision system module relationship schematic diagram of an embodiment of the coal blending and burning proportion optimization decision method for thermal power plants based on a feasible region model of the present application;

[0023] Figure 8 A three-type coal blending proportion feasible region schematic diagram corresponding to three types of coal in an example of an embodiment of the coal blending and burning proportion optimization decision method for thermal power plants based on a feasible region model of the present application;

[0024] Figure 9 An optimal blending proportion schematic diagram with the lowest unit heat quantity cost as the target in an example of an embodiment of the coal blending and burning proportion optimization decision method for thermal power plants based on a feasible region model of the present application;

[0025] Figure 10 A calculation time comparison schematic diagram between the present application and a traditional particle swarm algorithm in an example of an embodiment of the coal blending and burning proportion optimization decision method for thermal power plants based on a feasible region model of the present application;

[0026] Figure 11 A blending proportion adjustment result schematic diagram in an example of an embodiment of the coal blending and burning proportion optimization decision method for thermal power plants based on a feasible region model of the present application. DETAILED DESCRIPTION

[0027] The following is further described in detail through specific embodiments:

[0028] The embodiment is basically as shown in the accompanying drawings: a coal blending and burning proportion optimization decision method for thermal power plants based on a feasible region model, comprising the following steps: Figure 1 Step 1, constructing a basic model of the coal blending and burning proportion feasible region.

[0029] Determine the basic model of the coal blending and burning proportion feasible region in combination with the actual number of coal types of each power plant.

[0030] Specifically, the feasibility of a coal blending and burning scheme depends on whether its key parameters (including sulfur content, moisture content, ash content, and calorific value, etc.) meet the technical requirements of unit operation. To this end, the feasible region boundary g(M) of the blending and burning proportion under this working condition can be determined by establishing the equality and inequality constraint conditions of the coal blending parameters. Based on g(M), the entire blending and burning proportion space can be divided into two regions: the feasible region and the infeasible region. Any blending and burning proportion within the feasible region meets the technical requirements.

[0031]

[0032] ​Here, taking the blending of two types of coal as an example, the following is shown to explain the blending of two types of coal Figure 2 .

[0033] By Figure 2 It can be seen that in the case of blending of two types of coal, the inequality constraint of the blending parameters will limit the blending ratio to a two-dimensional planar region surrounded by multiple feasible region boundaries, and the equality constraint will further limit the blending ratio to a straight line. Therefore, after considering various constraint conditions, the blending ratio feasible region in this case is the intersection region of a two-dimensional plane and a one-dimensional straight line. Further, for the blending of three or more types of coal, the corresponding feasible region will change to the intersection region of a super polyhedron and a super plane.

[0034] Through the construction of the blending ratio feasible region, the operator can more effectively make optimization decisions on the blending ratio based on different optimization objectives.

[0035] Step 2, determine a first type of feasible region boundary function based on the blending constraint conditions, and a second type of feasible region boundary function based on the total blending ratio constraint condition, as shown in Figure 3 ; the blending constraint conditions include the blending calorific value, blending volatile matter, blending sulfur content, moisture content, ash content, and single coal type ratio.

[0036] The calorific value is a key indicator of the quality of blended coal. A higher calorific value of blended coal is beneficial to reducing power generation costs, while a calorific value that is too low will result in insufficient combustion temperature of the boiler furnace, as well as increased exhaust loss and coal consumption per kilowatt-hour. Correspondingly, the first type of feasible region boundary function corresponding to the blending calorific value is g1(M):

[0037] g1(M) = M1Q coal,1 + M2Q coal,2 + … + M k Q coal,k -Q mix,min ;

[0038] In the formula, M1, M2, …, M k are the proportions of the 1st, 2nd, …, kth types of coal in the blended coal; Q coal,1 , Q coal,2 , …, Q coal,k are the calorific values of the 1st, 2nd, …, kth types of coal, respectively, with units of kJ / kg; Q mix,min is the predetermined lower limit value of the blending calorific value, with units of kJ / kg.

[0039] As a key parameter affecting combustion characteristics, the volatile matter content of the coal blend directly affects the stability of combustion. When the volatile matter content is too high, it will cause flame extension and coking problems in the furnace; while insufficient volatile matter content will lead to unstable combustion. Correspondingly, the first type of feasible region boundary function corresponding to the volatile matter content of the coal blend includes the upper limit function g2(M) of the volatile matter content of the coal blend and the lower limit function g3(M) of the volatile matter content of the coal blend:

[0040] g2(M)=M1H coal,1 +M2H coal,2 +…+M k H coal,k -H mix,min ;

[0041] g3(M)=H mix,max -(M1H coal,1 +M2H coal,2 +…+M k H coal,k );

[0042] In the formula, H coal,1 , H coal,2 ,…, H coal,k are the volatile matter values of the 1st, 2nd,…, kth coal types, with the unit of %; H mix,max and H mix,min are the predetermined upper limit value and lower limit value of the volatile matter content of the coal blend, with the unit of %.

[0043] Excessive sulfur content of the coal blend will exacerbate low-temperature corrosion in the tail flue and cause environmental pollution problems; excessive moisture content will reduce the proportion of combustible components of the coal blend, increasing the cost of coal consumption; and excessive ash content will weaken the bonding performance of the coal blend, easily causing slagging and damaging the structure of the furnace. Therefore, all of the three indexes have upper limit restrictions, and three feasible region boundary functions corresponding to the inequality constraints of the sulfur content, moisture content, and ash content of the coal blend can be established accordingly. Specifically, the first type of feasible region boundary functions corresponding to the sulfur content, moisture content, and ash content of the coal blend are g4(M), g5(M), and g6(M) respectively:

[0044] g4(M)=S mix,max -(M1S coal,1 +M2S coal,2 +…+M k S coal,k );

[0045] g5(M)=W mix,max -(M1W coal,1 +M2W coal,2 +…+M k W coal,k );

[0046] g6(M)=Dmix,max -(M1D coal,1 +M2D coal,2 +…+M k D coal,k );

[0047] In the formula, S coal,1 , S coal,2 , …, S coal,k are sulfur content values of the 1st, 2nd, …, kth coal types, with units of %; W coal,1 , W coal,2 , …, W coal,k are moisture content values of the 1st, 2nd, …, kth coal types, with units of %; D coal,1 , D coal,2 , …, D coal,k are ash content values of the 1st, 2nd, …, kth coal types, with units of %; S mix,max , W mix,max , and D mix,max are predetermined upper limit values of the sulfur content, moisture content, and ash content of the blended coal, with units of %.

[0048] The blending proportions of the coal types need to satisfy non-negative constraints and upper limit constraints, and correspondingly, the first type of feasible region boundary functions corresponding to the single coal type proportions include a function g7(M) based on the non-negative constraints and a function g8(M) based on the upper limit constraints:

[0049] g7(M)=1-M i i=1,2,…,k;

[0050] g8(M)=M i i=1,2,…,k。

[0051] The feasible region boundary functions g1(M) to g8(M) need to satisfy g l (M)≥0, where l=1,2, …, 8.

[0052] The sum of the blending proportions of all coal types should equal 1, and correspondingly, the second type of feasible region boundary function based on the total blending proportion constraint condition is g9(M):

[0053] g9(M)=M1+M2+…+M k -1;

[0054] g9(M) satisfies g9(M)=0.

[0055] Step 3: A space projection algorithm based on vertex search is used to determine the blending blending proportion feasible region under the condition of multiple coal types. The specific sub-steps include:

[0056] S3.1, when the coal blending coal type k≥3, the blending ratio feasible region is a complex set formed by the intersection of hyper-multiple polyhedron and hyperplane in high-dimensional space.

[0057] Initialization of hyper-multiple polyhedron Θ * whose vertices are composed of the coordinate origin and the optimal vertices of k-dimensional coordinate axes.

[0058] Where the optimal vertex on the i-th (i=1, 2, …, k) coordinate axis can be solved by the following linear programming model:

[0059] max

[0060] s.t.g l (M)≥0 l=1,2,…,8;

[0061] In the formula, V region is the optimal vertex on the k-dimensional coordinate axis; is the unit basis vector in the direction of the i-th coordinate axis.

[0062] In addition, for the blending of two types of coal (k=2), the blending ratio feasible region can be characterized as a straight line segment in a two-dimensional plane, which can be directly solved by existing technology.

[0063] S3.2, search for new vertices on each face of Θ * to expand the feasible region.

[0064] New vertices can be obtained by solving the following linear programming model, which aims to find the vertex farthest from the interior point V * of Θ ip :

[0065]

[0066] s.t.

[0067] g l (M)≥0 l=1,2,…,8;

[0068] In the formula, H j is the height of the j-th face of Θ * ; H max is the maximum height of all faces of Θ * ; G j is the normal vector of the j-th face; ||G j || is the L2 norm of G j ; is the optimal solution of the linear programming model, i.e. the newly determined vertex.

[0069] S3.3, use the new vertex and the original vertex to construct a new sub-region together, and update Θ* ;

[0070] S3.4, repeat S3.2 and S3.3 until the volume of the newly generated hyperpolyhedron changes less than a threshold value compared to the previous round change value, and terminate the iteration.

[0071] Specifically,

[0072] In the formula, U bf and U af are the volumes of the hyperpolyhedron of the previous round and this round respectively; ò is the convergence coefficient (i.e. the threshold value).

[0073] S3.5, calculate the hyperpolyhedron Θ * and the intersection with the hyperplane g9(M) = 0, thereby determining the blending ratio feasible region As Figure 4 shown.

[0074] Step 4, set the objective function, based on the blending ratio feasible region, output the optimal blending ratio.

[0075] The objective function includes an optimization function F ue with the minimum use of the first and second types of coal as the target, and an optimization function F ef with the lowest cost per unit of heat as the target:

[0076] minF ue = M1+ M2;

[0077]

[0078] In the formula, R coal,1 , R coal,2 , …, R coal,k are the cost values of the first, second, …, k types of coal, with the unit being kg / yuan.

[0079] After determining the blending ratio feasible region of the coal blending, the range of the optimal blending ratio solution has been determined, and the adaptive grid search method can be directly used for iterative optimization until the termination condition is met, and the optimal blending ratio is output.

[0080] Specifically, as Figure 5 shown, first, Latin hypercube sampling is used to uniformly sample E grid points {M op,1 , M op,2 , …, M op,E} from the previously established blending ratio feasible region of the coal blending. Calculate the objective function F ue or F ef corresponding to each grid point, select the top C optimal grid points as the candidate grid points, and determine the optimal objective function value and the corresponding blending ratio of this round.

[0081] Then, the step-by-step grid contraction optimization is performed. The C optimal grid points determined in the foregoing are processed one by one, and C sub-regions are constructed with each candidate grid point as a center and a contraction radius of o. In each sub-region, E new grid points are generated by reusing the Latin hypercube sampling. The objective function value of each grid point in the sub-region is calculated, and the optimal solution of each sub-region is selected. Further, the optimal objective function value under the current grid contraction round is determined.

[0082] It is determined whether any of the following three termination conditions is met. The termination conditions include whether the current grid contraction round reaches a predetermined maximum iteration round Iter, whether the sub-region grid width is less than a preset threshold, and whether the number of candidate points is less than 1. If any of the three conditions is yes, it is determined that the termination condition is met, and the optimization result is output. If none of the three conditions is yes, it is determined that the termination condition is not met, and the step-by-step grid contraction optimization is performed again. In the next iteration, C-2 new candidate grid points are selected from the C sub-regions, and the region contraction and optimization are continued until the termination condition is met.

[0083] In addition, when the operating condition changes suddenly, the blending ratio of coal blending is quickly adjusted to respond to the real-time operating condition demand, so that the unit operation meets the new constraint condition; as shown in FIG. 2, the blending ratio adjustment process is equivalent to a geometric process of moving the current blending ratio to the boundary of the new feasible region, and the mathematical model is represented as follows: Figure 6

[0084] min d(M af ,M bf );

[0085] s.t.

[0086]

[0087] In the formula, M af and M bf are the blending ratios before and after adjustment, respectively; is a blending feasible region constructed based on the new constraint condition, i.e., the new feasible boundary; by solving the minimum d(M af ,M bf ), the adjusted blending ratio of coal blending is obtained.

[0088] In the solving process, the adaptive grid search method can also be used for solving.

[0089] The embodiment also provides a modular architecture of a coal blending ratio optimization decision system of a thermal power plant based on the method. Figure 7 ​​As shown, the method can provide technical support for a thermal power enterprise to make a blending ratio optimization decision. The modular architecture includes: an input module for collecting coal type characteristic data and index requirement data of a thermal power plant; a boundary function modeling and feasible region solving module for performing steps 1-3; a feasible region output display module for performing step 4; a target function modification module for modifying a target function; and an output module for summarizing a coal blending and combustion implementation scheme.

[0090] The method for making a blending ratio optimization decision for a thermal power plant provided by the embodiment is based on a feasible region model. First, a blending ratio feasible region modeling method is proposed based on a linear boundary function and a vertex search algorithm, and a blending ratio feasible region mathematical model of the thermal power plant is established. Second, a fast calculation method for an optimal blending ratio based on a feasible region is proposed in combination with a grid search method. In addition, a fast adjustment method for the blending ratio under a working condition mutation condition is proposed by equating the adjustment process of the blending ratio to a geometric process of moving the current blending ratio to a new boundary of the feasible region. The scheme can assist a thermal power plant operator to make an optimization decision on the blending ratio in real time and accurately.

[0091] Further, the method for making a blending ratio optimization decision for a thermal power plant is verified for feasibility and effectiveness by taking a blending of three types of coal in a certain power plant as an example. The characteristic parameters of the coal types and the characteristic parameter requirements of the coal blending are shown in Table 1.

[0092] Table 1: Introduction of coal type characteristic parameters

[0093]

[0094] By computer program calculation and visual processing, the coal blending and combustion feasible region satisfying the above constraint conditions can be obtained, and the result is as follows Figure 8 As shown, the I region is a super polyhedron, and the II region is a superplane.

[0095] Based on the established coal blending and combustion feasible region, the method proposed in the present application can quickly solve the optimal blending ratio satisfying the given coal blending parameter constraints with the optimization target of minimizing the unit heat quantity cost, and the optimization result is as shown in Figure 9

[0096] When the coal blending and combustion is performed according to the above ratio, the minimum unit heat quantity cost is 1.09x10-5 yuan / kJ. It should be noted that the calculation time of the fast calculation method proposed in the present application and the traditional particle swarm algorithm is compared as shown in Figure 10 Figure 10 As can be seen from the above table, the calculation time of the optimal blending ratio decision made by the method proposed in the present application is less than 1 minute, which fully meets the real-time calculation requirements of the site, and verifies the effectiveness and engineering practicability of the method.​​

[0097] On this basis, it is assumed that the coal blending heat value needs to be increased to 22500kg / kJ, at this time, the adjusted blending ratio of each coal is obtained by computer programming solution, and the specific is as shown in Figure 11

[0098] When the coal blending is carried out according to the above-mentioned proportion, the volatile matter, sulfur content, moisture content and ash content of the coal blending are 22.01%, 1.13%, 10.34% and 17.33% respectively, which meet the given characteristic parameter requirements, and the effectiveness of the blending ratio rapid adjustment method is proved.

[0099] The above example analysis proves that the method can effectively assist the operation personnel of the thermal power plant to make optimization decision of the blending ratio, and has good engineering practical value.

[0100] The above-mentioned is only an embodiment of the present application, and the well-known specific structure and characteristics of the scheme are not described in detail, the ordinary technical personnel in the art know all the ordinary technical knowledge in the technical field of the present application before the application date or the priority date, can know all the prior art in the field, and has the ability to apply the conventional experimental means before the date, the ordinary technical personnel in the art can improve and implement the present scheme under the guidance of the present application, some typical well-known structure or well-known method should not be an obstacle for the ordinary technical personnel in the art to implement the present application. It should be pointed out that for the technical personnel in the art, without departing from the structure of the present application, a number of modifications and improvements can be made, which should also be considered as the protection scope of the present application, which will not affect the effect and practicality of the present application.​

Claims

1. A method for optimizing the coal blending ratio in thermal power plants based on a feasible region model, characterized in that, Includes the following steps: Step 1: Construct a basic model of the feasible region for coal blending ratio; Step 2: Determine the boundary function of the first type of feasible region based on the coal blending and combustion constraints, and determine the boundary function of the second type of feasible region based on the total coal blending ratio constraints; the coal blending and combustion constraints include the calorific value of the blended coal, the volatile matter of the blended coal, the sulfur content of the blended coal, the moisture content, the ash content, and the proportion of single coal types; The second type of feasible region boundary function is : ； In the formula, The first The proportion of different types of coal in coal blending; satisfy =0; Step 3: Use a vertex search-based spatial projection algorithm to determine the feasible region for blending and combustion ratios of multiple coal types. Step 3 includes the following sub-steps: S3.1, when the type of coal used in the blend is k At that time, the feasible domain for coal blending ratio is a set formed by the intersection of hyperpolyhedra and hyperplanes in high-dimensional space; Initialize hyperpoly Its vertices are formed by the origin and the optimal vertices of the k-dimensional coordinate axes; S3.2, in Search for new vertices on each face to expand the feasible region; S3.3, construct a new sub-region by combining the new vertices with the existing vertices, and update... ; S3.4, Repeat S3.2 and S3.3 until the volume change of the newly generated hyperpoly is less than the threshold value compared with the previous round, then terminate the iteration; S3.5, Calculating Hyperpolyhedra With the hyperplane The intersection points are used to determine the feasible region for the blending ratio. ; Step 4: Define the objective function and output the optimal blending ratio based on the feasible region of the coal blending ratio. Step 4 also includes: When operating conditions change abruptly, the coal blending ratio is rapidly adjusted. The process of adjusting the blending ratio can be represented as a geometric process of moving the current blending ratio to the boundary of a new feasible region, and its mathematical model is as follows: ； ； ； In the formula, and These are the coal blending ratios before and after the adjustment; To construct the feasible region for coal blending and combustion based on the new constraints, i.e., the new feasible boundary; and to solve for the minimization The adjusted coal blending ratio was obtained.

2. The method for optimizing the coal blending ratio in thermal power plants based on a feasible region model according to claim 1, characterized in that, The first type of feasible region boundary function corresponding to the calorific value of the coal blend is: : ； In the formula, The first The proportion of different types of coal in coal blending; The first The calorific value of coal types, expressed in kJ / kg; This is the predetermined lower limit of the calorific value of the coal blend, expressed in kJ / kg.

3. The method for optimizing the coal blending ratio in thermal power plants based on a feasible region model according to claim 1, characterized in that, The first type of feasible region boundary function corresponding to the volatile matter in the coal blend includes: the upper bound function of the volatile matter in the coal blend. and the lower limit function of volatile matter in coal blending : ； ； In the formula, The first The proportion of different types of coal in coal blending; The first Volatile content of coal types, in % and These are the predetermined upper and lower limits of volatile matter content in blended coal, expressed in units of %.

4. The method for optimizing the coal blending ratio in thermal power plants based on a feasible region model according to claim 1, characterized in that, The first-type feasible region boundary functions corresponding to the sulfur content, moisture content, and ash content of the blended coal are respectively... , and : ； ； ； In the formula, The first The proportion of different types of coal in coal blending; The first Sulfur content of coal types, in % (%) The first Moisture content of coal types, in % % The first Ash content of coal types, in percentage (%) , and These are the predetermined upper limits for sulfur, moisture, and ash content in the blended coal, expressed in units of %.

5. The method for optimizing the coal blending ratio in thermal power plants based on a feasible region model according to claim 1, characterized in that, The first type of feasible region boundary function corresponding to the proportion of the single coal type includes: a function based on nonnegativity constraints. and functions based on upper limit constraints : ； ； In the formula, The first The proportion of different types of coal in coal blending.

6. The method for optimizing the coal blending ratio in thermal power plants based on a feasible region model according to claim 1, characterized in that, The objective function includes an optimization function that aims to minimize the usage of both Type 1 and Type 2 coal. And the optimization function with the objective of minimizing the cost per unit of heat output. : ； ； In the formula, The first Cost value of coal type, in kg / yuan.

7. The method for optimizing the coal blending ratio in thermal power plants based on a feasible region model according to claim 1, characterized in that, In step 4, after determining the feasible region of the coal blending ratio, the adaptive grid search method is used for iterative optimization until the termination condition is met, and the optimal blending ratio is output. The termination conditions include: whether the current grid shrinkage round has reached the predetermined maximum iteration round Iter, whether the sub-region grid width is less than a preset threshold, and whether the number of candidate points is less than 1. If any one of these conditions is met, the termination condition is deemed to be satisfied; if none of these conditions are met, the termination condition is deemed not to be satisfied.

Images