A method and system for analyzing dynamic coal consumption cost of entering a furnace

Abstract

The present application relates to coal-fired power generation technology field, a kind of into furnace dynamic coal consumption cost analysis method and system are proposed, including the real-time effective identification of mixed batch of into furnace coal, the proportion of each batch into furnace coal in mixed batch is calculated, the actual price of into furnace coal is obtained, according to the real-time power data and coal consumption data of unit collected, dynamic coal consumption characteristic curve is fitted, the dynamic relationship between real-time power and coal consumption is obtained;According to the cost of into furnace coal and dynamic coal consumption characteristic curve, the real-time into furnace dynamic coal consumption cost of thermal power plant is calculated.Through the above technical scheme, the problem of not accurate in dynamic coal consumption cost analysis in the related art is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of coal-fired power generation technology, in particular, to a method and system for analyzing dynamic coal consumption cost in a furnace. BACKGROUND

[0002] Currently, the actual coal fed into the furnace in most power plants in China is mixed coal, that is, coal formed by mixing and grinding all or part of the current batch of coal in the coal bunker. However, it is difficult to accurately identify the batch and mixing ratio of mixed coal, and many power plants have not established a real-time relationship between the quality of coal fed into the furnace and the consumption of coal for burning, so the true price of the coal fed into the furnace cannot be accurately calculated. At the same time, the combustion of the boiler is continuous and the load of the unit is time-varying, which makes it impossible to accurately analyze the dynamic coal consumption cost in the furnace. With the development of digital technology, introducing digital analysis methods into coal consumption analysis can help improve the accuracy of dynamic coal consumption cost analysis in the furnace. SUMMARY

[0003] The present application provides a method and system for analyzing dynamic coal consumption cost in a furnace, which solves the problem of inaccurate dynamic coal consumption cost analysis in the related art.

[0004] The technical solution of the present application is as follows:

[0005] In a first aspect, a method for analyzing dynamic coal consumption cost in a furnace is provided, which effectively identifies the mixed batch of coal fed into the furnace in real time, calculates the mixing ratio of each batch of coal fed into the furnace in the mixed batch, and obtains the actual price of the coal fed into the furnace, which specifically includes the following steps:

[0006] Collect the coal quality test results of all batches of incoming coal in the thermal power plant, form a database of the coal quality test results of all batches of incoming coal, and use m to represent the total number of batches of incoming coal currently stored in the coal bunker, and represent each batch of incoming coal as a set {a1, a2, … a m},

[0007] Wherein a i (i=1, 2, … m) represents the incoming coal marked as the i-th batch,

[0008] Real-time acquisition of the heat value Q and sulfur content S of the coal burned in the furnace, and the relationship between the heat value and sulfur content of the coal burned in the furnace and each batch of incoming coal is as follows:

[0009] x1Q1+x2Q2+...+x m Q m =Q,(i=1,2,...m,0≤x i ≤1)

[0010] x1S1+x2S2+...+x m S m =S,(i=1,2,...m,0≤x i≤1)

[0011] wherein, Qi (i = 1, 2, …m) represents the heat value of the i-th batch of incoming coal, S i (i = 1, 2, …m) represents the sulfur content of the i-th batch of incoming coal, x i represents the blending ratio of the i-th batch of incoming coal in the current calculated furnace coal;

[0012] The blending ratio of each batch of incoming coal in the furnace coal is calculated by an evolutionary algorithm;

[0013] The furnace coal cost is obtained according to the blending ratio of each batch of incoming coal in the furnace coal, and is specifically represented as:

[0014]

[0015]

[0016] wherein, C i (i = 1, 2, …m) represents the unit price of the i-th batch of incoming coal, C represents the actual unit cost of the furnace coal, C SC represents the standard coal price of the furnace coal;

[0017] According to the collected real-time power data and coal consumption data of the unit, a dynamic coal consumption characteristic curve is fitted to obtain the dynamic relationship between real-time power and coal consumption;

[0018] The real-time dynamic coal consumption cost of the thermal power plant is calculated according to the furnace coal cost and the dynamic coal consumption characteristic curve.

[0019] In a second aspect, a real-time dynamic coal consumption cost analysis system comprises

[0020] A first calculation module: real-time effective identification of the mixed batch of furnace coal, calculation of the blending ratio of each batch of furnace coal in the mixed batch, and obtaining of the actual price of the furnace coal; specifically comprising the following steps:

[0021] Collecting the coal quality test results of all batches of incoming coal of the thermal power plant, forming a database of the coal quality test results of all batches of incoming coal, and using m to represent the total number of batches of incoming coal present in the coal bunker, and the batches of incoming coal are represented as a set {a1, a2, …am}, m

[0022] wherein a i (i = 1, 2, …m) represents the incoming coal marked as the i-th batch,

[0023] Real-time acquisition of the heat value Q and sulfur content S of the furnace coal, and the relationship between the heat value and sulfur content of the furnace coal and each batch of incoming coal is as follows:

[0024] x1Q1+x2Q2+...+xmQm​m Q m = Q, (i = 1, 2,..., m, 0 ≤ x i ≤ 1)

[0025] x1S1+x2S2+...+x m S m = S, (i = 1, 2,..., m, 0 ≤ x i ≤ 1)

[0026] Wherein, Q (i = 1, 2,..., m) represents the i batch of incoming coal combustion heat value, S i (i = 1, 2,..., m) represents the i batch of incoming coal sulfur content, x i It represents the i batch of incoming coal in the mixed batch of coal in the calculation of the blending ratio of this time;

[0027] The blending ratio of each batch of incoming coal in the mixed batch of coal in the furnace is calculated by the evolutionary algorithm;

[0028] The blending ratio of each batch of incoming coal in the mixed batch of coal in the furnace is calculated by the evolutionary algorithm;

[0029]

[0030]

[0031] Wherein, C i (i = 1, 2,..., m) represents the i batch of incoming coal unit price, C represents the actual unit cost of coal in the furnace, C SC It represents the standard coal price of coal in the furnace;

[0032] Characteristic curve fitting module: according to the collected unit real-time power data and coal consumption data, the dynamic coal consumption characteristic curve is fitted, and the dynamic relationship between real-time power and coal consumption is obtained;

[0033] The second calculation module: according to the cost of coal in the furnace and the dynamic coal consumption characteristic curve, the real-time dynamic coal consumption cost of thermal power plant is calculated.

[0034] The working principle and beneficial effects of the application are as follows:

[0035] This invention first uses the calorific value and sulfur content of the coal fed into the furnace as important factors for coal identification. Then, an evolutionary algorithm is used to effectively identify the mixed batches of coal fed into the furnace, and then the blending ratio of coal supplied to the plant in each batch of coal fed into the furnace is calculated. Then, based on the price of each batch of coal supplied to the plant, the actual price of coal fed into the furnace is calculated, and finally the actual cost of coal fed into the furnace is obtained. At the same time, by collecting real-time power data and coal consumption data of the unit, the dynamic coal consumption characteristic curve of the unit is obtained by fitting, realizing the dynamic analysis of coal consumption cost of the furnace, thus effectively solving the key problem of dynamic calculation of coal cost and achieving accurate identification of dynamic coal consumption cost of coal fed into the furnace. Attached Figure Description

[0036] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0037] Figure 1 This is a flowchart of a dynamic coal consumption cost analysis method for furnace feeding according to the present invention;

[0038] Figure 2 This is a flowchart illustrating the calculation of the coal blending ratio for each batch in this invention.

[0039] Figure 3 This is a flowchart of fitting the dynamic coal consumption characteristic curve in this invention;

[0040] Figure 4 This is a structural diagram of the SCSO-L1-norm-LSTSVR model improved based on the greedy formula in this invention;

[0041] Figure 5 This is a flowchart of an embodiment of the dynamic coal consumption cost analysis method for furnace feeding according to the present invention;

[0042] Figure 6 This is a schematic diagram of the structure of a dynamic coal consumption cost analysis system for furnace feeding according to the present invention. Detailed Implementation

[0043] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0044] Example 1

[0045] like Figures 1-5 As shown, this embodiment proposes a method for dynamic coal consumption cost analysis of furnace feed, including...

[0046] S100: Real-time effective identification of mixed batches of coal into the furnace, calculation of the mixing ratio of each batch of coal into the furnace, and obtaining the actual price of the coal into the furnace,

[0047] First, the calorific value and sulfur content of the coal into the furnace are taken as important factors for identifying coal, then the evolutionary algorithm is used to effectively identify the mixed batches of coal into the furnace, and then the mixing ratio of each batch of coal into the furnace is calculated, and finally the actual price of the coal into the furnace is obtained.

[0048] Specifically, the following steps are included:

[0049] S110: Collect the coal quality test results of all batches of coal into the power plant, form a database of the coal quality test results of all batches of coal into the power plant, and use m to represent the total number of coal batches present in the coal bunker, and the set of each batch of coal into the power plant is represented as {a1, a2, … a m m}, wherein a i (i=1, 2, … m) represents the coal into the power plant marked as the i-th batch,

[0050] The coal quality test results data of all batches of coal into the power plant are obtained from the fuel information system.

[0051] S120: Real-time acquisition of the coal combustion calorific value Q and sulfur content S of the coal into the furnace, and the relationship between the calorific value and sulfur content of the coal into the furnace and each batch of coal into the power plant is as follows:

[0052] x1Q1+x2Q2+...+x m Q m =Q,(i=1,2,...m,0≤x i ≤1)

[0053] x1S1+x2S2+...+x m S m =S,(i=1,2,...m,0≤x i ≤1)

[0054] Wherein, Q i (i=1, 2, … m) represents the coal combustion calorific value of the i-th batch of coal into the power plant, S i (i=1, 2, … m) represents the sulfur content of the i-th batch of coal into the power plant, and x i represents the mixing ratio of the i-th batch of coal into the power plant in the coal into the furnace in this calculation;

[0055] The mixing ratio of each batch of coal into the power plant in the coal into the furnace needs to satisfy the following formula:

[0056] x1+x2+...+x m =1,(i=1,2,...m,0≤x i ≤1)

[0057] S130: Calculate the blending ratio of each batch of incoming coal in the furnace coal in the mixed batch by the evolutionary algorithm;

[0058] The coal combustion heat value and sulfur content are taken as initial coefficients to solve, the deviation degree is calculated by taking the arithmetic mean value as the center, so as to adjust the coefficient to solve more quickly, and the evolutionary algorithm (in this embodiment, the genetic algorithm in the evolutionary algorithm) is used to solve.

[0059] Considering that there are many residual coal layers at the bottom of the coal bunker, a complete accurate solution cannot be obtained, so the problem is converted into a multi-objective optimization problem with the minimum deviation of the furnace coal combustion heat value Q and sulfur content S as the objective function. The genetic algorithm is used to solve the multi-objective optimization problem in this embodiment, which effectively avoids the subjective process of selecting weights for multiple objectives in the solving process of converting the multi-objective problem into a single objective problem, and can better depict and solve practical problems. The specific steps are as follows:

[0060] Step 1: Parameter initialization setting. Read the relevant economic parameters of the fuel management system (including the coal combustion heat value Q i , sulfur content S i and the coal combustion heat value Q and sulfur content S of the furnace coal of each batch of incoming coal);

[0061] Step 2: Initialize the population of the genetic algorithm. Randomly generate K possible genetic individuals, which are used as the initial population;

[0062] Step 3: Set the fitness function of the genetic algorithm to evaluate the fitness of the initial population U;

[0063] Step 4: Sort the initial population U. Calculate the fitness function between each individual and calculate the population crowding degree;

[0064] Step 5: Update the initial population U of the genetic algorithm. Get the new offspring population U2, U3 through selection, crossover and mutation operations.

[0065] Step 6: Update the fitness of the population U3;

[0066] Step 7: Sort the initial population {U, U3};

[0067] Step 8: Select the individual with high ranking corresponding to the minimum target function from the initial population {U, U3} to form a new population U z ;

[0068] Step 9: If the maximum number of iterations is reached, stop running, otherwise return to step 5;

[0069] Step 10: Take the obtained solution as the optimal solution of the multi-objective programming problem, that is, the blending ratio of each batch of incoming coal in the furnace coal.

[0070] S140: The blending ratio of each batch of incoming coal in the furnace coal is obtained, and the cost of the furnace coal is calculated, which is specifically expressed as:

[0071]

[0072]

[0073] Wherein, C i (i = 1, 2, … m) represents the unit price of the i-th batch of incoming coal, C represents the actual unit cost of the furnace coal, and C SC represents the standard coal price of the furnace coal;

[0074] S200: According to the collected real-time power data and coal consumption data of the unit, a dynamic coal consumption characteristic curve is fitted to obtain the dynamic relationship between real-time power and coal consumption;

[0075] During the construction stage of the power plant, the coal consumption test curve is obtained by fitting the unit design specification and coal consumption data. However, considering the influence of the actual running state (such as aging problems, seasonal characteristics, and coal quality) during the actual operation process, the actual running state inevitably deviates from the design running state, thereby affecting the coal consumption. Therefore, it is necessary to recalculate the standard coal consumption of the unit under different loads based on the test data. Through simple mechanism model analysis, the energy consumption characteristics of the unit cannot be fully and accurately reflected. Therefore, a new modeling method is needed, which not only conforms to the actual operation of the unit, but also reflects the energy consumption characteristics of the unit.

[0076] The influencing factors between the coal-fired units have the characteristics of nonlinearity and strong coupling. In view of this characteristic, a high-precision data regression modeling method can be used for analysis. Therefore, the present application uses historical data of the unit operation to establish a least squares twin support vector regression model (Least Squares Twin Support Vector Regression (LSTSVR)) based on L1-norm distance, and improves the sand cat swarm optimization algorithm (Sand Cat Swarm Optimization, SCSO) through the greedy formula to find the best parameters of the support vector machine, analyzes the energy consumption characteristics of the unit, and finally uses the K-means clustering method to fit the dynamic coal consumption characteristic curve.

[0077] The SCSO-L1-norm-LSTSVR model improved based on the greedy formula is used as the coal consumption characteristic curve fitting model, and the specific steps are as follows:

[0078] S210: Obtain sample data of factors affecting the power supply coal consumption of the unit;

[0079] Considering that there are many factors affecting the power supply coal consumption of the unit, the coal quality, flue gas oxygen content, primary air volume, secondary air volume, feed water flow, feed water temperature, flue gas temperature, superheating desuperheating water flow, reheating desuperheating water flow and the like are analyzed as the input parameters of the coal-fired side, and the output parameter is the standard power supply coal consumption of the unit.

[0080] S220: The collected samples are pretreated and divided into training samples and test samples;

[0081] The selected training samples are imported into the SCSO-L1-norm-LSTSVR model based on the improved greedy formula for calculating the coal consumption characteristic curve of the unit, the parameters in the model are set, the coal consumption value of the unit is obtained through model calculation.

[0082] S230: The dynamic coal consumption characteristic curve is fitted by the SCSO-L1-norm-LSTSVR model based on the improved greedy formula and the K-means clustering-least squares method.

[0083] Based on the SCSO-L1-norm-LSTSVR model based on the improved greedy formula, the collected unit operation parameters are classified to obtain the cluster center point according to the data characteristics. Then the distance is used as the classification division standard of the K-means algorithm to realize efficient clustering. The sample data used for the coal consumption characteristic curve of the unit is divided into groups according to the load, and the K-means clustering algorithm is used for clustering analysis for each group of samples, and the steps are as follows:

[0084] Step 1: Randomly select g sample data of input parameters as the starting cluster center point.

[0085] Step 2: Calculate the distance between each sample and the starting cluster center point, and divide all sample points according to the distance through comparison and analysis.

[0086] Step 3: After classification, the average value of each type of sample data is calculated by using the improved SCSO-L1-norm-LSTSVR model, that is, the average value of coal consumption, that is, the coal consumption reference value corresponding to the load, and it is used as the center point of the new round of clustering. Repeat the clustering process, and when and only when the following criterion function is satisfied, it can be determined that convergence has been reached:

[0087]

[0088] Wherein, E represents the sum of squared errors of sample data, x j represents the i-th sample point, and m i is the center point of the i-th sample set after clustering.

[0089] Step 4: Determine the number of clusters g using the Silhouette clustering efficiency evaluation function, as expressed by the following formula:

[0090]

[0091] Where S(i) represents the contour value that measures the reasonableness of clustering, and the larger the S(i) value, the more reasonable the classification. x represents the average distance between the i-th point and other points in the same class, and y represents the vector of distances between the i-th point in the same class and all sample points in other classes.

[0092] Finally, the load and its corresponding coal consumption data for each group are plotted as a scatter plot, and the relationship between the load and the standard coal consumption for power supply is fitted using the least squares method in MATLAB software to obtain the dynamic coal consumption characteristic curve.

[0093] S300: Calculates the real-time dynamic coal consumption cost of a thermal power plant based on the cost of coal fed into the furnace and the dynamic coal consumption characteristic curve.

[0094] Real-time costs for thermal power plants include variable costs and fixed costs. Variable costs include fuel costs, auxiliary equipment power consumption costs, pollution discharge costs, and maintenance costs. Fixed costs include capacity costs, financial costs, material costs, wages and benefits, and other expenses (these can be obtained directly from the finance department). To describe the dynamic coal consumption cost at the furnace more specifically, three indicators are used: marginal cost, variable cost, and monthly average power generation cost.

[0095] 1. The marginal cost of power generation mainly consists of three parts: coal consumption cost, water cost, and pollutant treatment cost. The calculation formula is as follows:

[0096]

[0097] in, The slope of the dynamic coal consumption characteristic curve represents the marginal coal consumption for power generation. This refers to coal consumption costs; F f G represents the amount of pollutant f generated per unit of electricity generated; f Here, f represents the pollutant discharge fee standard; u represents the total number of items requiring fee-based pollutant treatment. That is, the cost of treating pollutants per unit of electricity generated; W is the water consumption per unit of electricity generated; C W The unit water price is W×C. W This refers to the unit water consumption cost for generating electricity; E e c represents the auxiliary machine operating power per unit of electricity generated. e For the price of electricity used by the power plant itself, E e ×c e This refers to the auxiliary equipment power consumption cost; g is the operation and maintenance cost per unit of power generation.

[0098] 2. Variable costs are expressed as the sum of the cost of coal fed into the furnace, the cost of auxiliary equipment electricity consumption, the cost of sewage discharge, and the cost of operation and maintenance. The calculation formula is as follows:

[0099]

[0100]

[0101] Among them, R T and Q T These represent the coal consumption and power generation in hour T, respectively. t Let be the output power of the generator set at time t.

[0102] 3. The average monthly power generation cost is measured by the sum of the total monthly variable cost and the monthly fixed cost. The calculation formula is as follows:

[0103]

[0104] Where d represents day, D represents the number of days in a month of the statistical period, and FC represents monthly fixed cost.

[0105] Example 2

[0106] like Figure 6 As shown, based on the same concept as Embodiment 1 above, this embodiment also proposes a dynamic coal consumption cost analysis system for furnace feed, including...

[0107] The first calculation module: This module effectively identifies mixed batches of coal fed into the furnace in real time, calculates the proportion of coal from each batch within the mixed batch, and obtains the actual price of the coal fed into the furnace; specifically, it includes the following steps:

[0108] Collect coal quality test results for all batches of coal entering the thermal power plant, and compile these results into a database. Let m represent the total number of coal batches currently in the coal bunker, and let each batch of coal be represented by a set {a1, a2, ... a...}. m},

[0109] Where a i (i = 1, 2, ..., m) represents the coal entering the plant marked as the i-th batch.

[0110] The calorific value Q and sulfur content S of the coal fed into the furnace are obtained in real time. The relationship between the calorific value and sulfur content of the coal fed into the furnace and each batch of coal entering the plant is as follows:

[0111] x1Q1+x2Q2+...+x m Q m =Q,(i=1,2,...m,0≤x) i ≤1)

[0112] x1S1+x2S2+...+xm S m =S,(i=1,2,...m,0≤x) i ≤1)

[0113] Where Qi (i = 1, 2, ..., m) represents the calorific value of the i-th batch of coal delivered to the plant, and S i (i = 1, 2, ..., m) represents the sulfur content of the i-th batch of coal delivered to the plant, x i This indicates the blending ratio of the i-th batch of coal delivered to the plant in the coal fed into the furnace in this calculation;

[0114] The blending ratio of each batch of coal entering the plant in the coal entering the furnace is calculated using an evolutionary algorithm.

[0115] The cost of coal fed into the furnace is obtained based on the blending ratio of each batch of coal delivered to the plant and the coal fed into the furnace, specifically expressed as follows:

[0116]

[0117]

[0118] Among them, C i (i = 1, 2, ..., m) represents the unit price of the i-th batch of coal delivered to the plant, and C represents the actual unit cost of the coal delivered to the furnace. SC This indicates the standard coal price of the coal fed into the furnace;

[0119] Characteristic curve fitting module: Based on the collected real-time power data and coal consumption data of the unit, fit the dynamic coal consumption characteristic curve to obtain the dynamic relationship between real-time power and coal consumption;

[0120] The second calculation module calculates the real-time dynamic coal consumption cost of the power plant based on the cost of coal fed into the furnace and the dynamic coal consumption characteristic curve.

[0121] Furthermore, the blending ratio of each batch of coal entering the plant in the furnace is calculated using an evolutionary algorithm, specifically including:

[0122] Step 1: Parameter initialization settings, read relevant economic parameters from the fuel management system, including the calorific value Qi and sulfur content Si of each batch of coal delivered to the plant, and the calorific value Q and sulfur content S of the coal delivered to the furnace;

[0123] Step 2: Initialize the population for the genetic algorithm by randomly generating K possible genetic individuals and using them as the initial population;

[0124] Step 3: Set the fitness function of the genetic algorithm to evaluate the fitness of the initial population U;

[0125] Step 4: Organize the initial population U, calculate the fitness function between individuals, and calculate the population crowding.

[0126] Step 5: Update the initial population U of the genetic algorithm, and obtain new offspring populations U2 and U3 through selection, crossover and mutation operations;

[0127] Step 6: Update the fitness assessment of the U3 population;

[0128] Step 7: Organize the initial population {U, U3};

[0129] Step 8: Select individuals with higher ranking levels corresponding to the minimized objective function from the initial population {U, U3} to form a new population Uz;

[0130] Step 9: If the maximum number of iterations is reached, stop running; otherwise, return to step 5.

[0131] Step 10: Take the obtained solution as the optimal solution to the multi-objective programming problem, that is, the blending ratio of each batch of coal fed into the furnace.

[0132] Furthermore, the characteristic curve fitting module includes:

[0133] The SCSO-L1-norm-LSTSVR model, improved based on the greedy formula, is used as the coal consumption characteristic curve fitting model. The specific steps are as follows:

[0134] Obtain sample data on factors affecting the coal consumption for power generation of the generating unit;

[0135] The collected samples are preprocessed and divided into training samples and test samples;

[0136] The dynamic coal consumption characteristic curve was fitted by the SCSO-L1-norm-LSTSVR model based on the greedy formula and the K-means clustering-least squares method.

[0137] Furthermore, sample data on factors affecting the unit's coal consumption for power generation were obtained, specifically including: coal quality, flue gas oxygen content, primary air volume, secondary air volume, feedwater flow rate, feedwater temperature, flue gas temperature, superheating desuperheating water flow rate, and reheating desuperheating water flow rate.

[0138] Furthermore, the second calculation module specifically includes: calculation of marginal generation cost, calculation of variable cost, and calculation of monthly average generation cost.

[0139] The calculation of marginal cost of power generation includes:

[0140]

[0141] in, The slope of the dynamic coal consumption characteristic curve represents the marginal coal consumption for power generation. This refers to coal consumption costs; F f G represents the amount of pollutant f generated per unit of electricity generated; fHere, f represents the pollutant discharge fee standard; u represents the total number of items requiring fee-based pollutant treatment. That is, the cost of treating pollutants per unit of electricity generated; W is the water consumption per unit of electricity generated; C W The unit water price is W×C. W This represents the water consumption cost per unit of electricity generated; g represents the auxiliary machine operating power corresponding to a unit of electricity generated; c e For the price of electricity used by the power plant itself, E e ×c e This refers to the auxiliary equipment power consumption cost; g is the maintenance cost per unit of power generation.

[0142] Variable cost calculation includes:

[0143]

[0144]

[0145] Among them, R T and Q T These represent the coal consumption and power generation in hour T, respectively. t Let t be the output power of the generator set;

[0146] The calculation of average monthly power generation cost includes:

[0147]

[0148] Where d represents day, D represents the number of days in a month of the statistical period, and FC represents monthly fixed cost.

[0149] The various variations and specific examples of the dynamic coal consumption cost analysis method for furnace feed in the aforementioned Embodiment 1 are also applicable to the dynamic coal consumption cost analysis system for furnace feed in this embodiment. Through the detailed description of the aforementioned dynamic coal consumption cost analysis method for furnace feed, those skilled in the art can clearly understand the implementation method of the dynamic coal consumption cost analysis system for furnace feed in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0150] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method of in-furnace dynamic coal consumption cost analysis, characterized by, Comprising Real-time effective identification of mixed batches of coal into the furnace, calculating the mixing ratio of each batch of coal into the furnace in the mixed batch, and obtaining the actual price of the coal into the furnace; specifically comprising the following steps: Collecting the coal quality test results of all batches of incoming coal of the thermal power plant, forming a database of the coal quality test results of all batches of incoming coal, denoting the total number of batches of incoming coal currently stored in the coal bunker as m, and denoting the batches of incoming coal as a set represented as {a1, a2, … a m} where a i (i = 1, 2,... m) represents the incoming coal labeled as the ith batch, Real-time acquisition of the heat value Q and sulfur content S of the coal into the furnace, and the relationship between the heat value and sulfur content of the coal into the furnace and each batch of coal into the plant is as follows: Wherein, Qi (i=1, 2, …m) represents the heat value of the i-th batch of incoming coal, S i (i=1, 2, …m) represents the sulfur content of the i-th batch of incoming coal, x i represents the blending ratio of the i-th batch of incoming coal in the incoming coal for this calculation. Calculate the mixing ratio of each batch of coal into the plant in the coal into the furnace by an evolutionary algorithm; According to the mixing ratio of each batch of coal into the plant in the coal into the furnace, the cost of the coal into the furnace is obtained, which is specifically expressed as: Wherein, C i (i = 1, 2, … m) represents the unit price of the i batch of incoming coal, C represents the actual unit cost of the incoming coal, C SC represents the price of the incoming coal converted to standard coal. According to the collected real-time power data and coal consumption data of the unit, the dynamic coal consumption characteristic curve is fitted, and the dynamic relationship between real-time power and coal consumption is obtained; According to the cost of the coal into the furnace and the dynamic coal consumption characteristic curve, the real-time dynamic coal consumption cost of the thermal power plant is calculated; The calculation of the real-time dynamic coal consumption cost of the thermal power plant according to the cost of the coal into the furnace and the dynamic coal consumption characteristic curve specifically includes: marginal cost calculation, variable cost calculation and monthly average power generation cost calculation, The marginal cost calculation includes: wherein, is the slope of the dynamic coal consumption characteristic curve, representing the marginal coal consumption for power generation, is the coal consumption cost; is the amount of the fth pollutant generated per unit of power generation; is the fth pollutant emission charge standard; represents the total number of items that need to be charged for pollution treatment, is the pollution treatment cost per unit of power generation; W is the water consumption per unit of power generation; is the unit water price, is the water consumption cost per unit of power generation; is the auxiliary machine operating power corresponding to a unit of power generation, is the self-use electricity price of the power plant, is the auxiliary machine electricity consumption cost; is the maintenance cost per unit of power generation; The variable cost calculation includes: wherein, and are the first hourly coal consumption and power generation, is the output power of the generator set at time t. The monthly average power generation cost calculation includes: wherein, represents the day of the week, represents the number of days in the statistical period month, represents the monthly fixed cost.

2. The method of claim 1, wherein, The calculation of the mixing ratio of each batch of coal into the plant in the coal into the furnace by the evolutionary algorithm specifically includes: Step 1: parameter initialization setting, reading the relevant economic parameters of the fuel management system, including the heat value Qi and sulfur content Si of each batch of coal into the plant, and the heat value Q and sulfur content S of the coal into the furnace; Step 2: initialize the population of the genetic algorithm, randomly generate K possible genetic individuals, and take them as the initial population; Step 3: set the fitness function of the genetic algorithm, and evaluate the fitness of the initial population U; Step 4: sort the initial population U, calculate the fitness function between each individual, and calculate the population crowding degree; Step 5: update the initial population U of the genetic algorithm, and get the new offspring population U2, U3 through selection, crossover and mutation operations; Step 6: update the fitness of the population U3; Step 7: sort the initial population {U, U3}; Step 8: select the individual with high ranking corresponding to the minimum target function from the initial population {U, U3} to form a new population Uz; Step 9: if the maximum number of iterations is reached, stop running, otherwise return to step 5; Step 10: take the obtained solution as the optimal solution of the multi-objective programming problem, that is, the mixing ratio of each batch of coal into the furnace.

3. The method of claim 1, wherein, The fitting of the dynamic coal consumption characteristic curve to obtain the dynamic relationship between real-time power and coal consumption specifically includes: An SCSO-L1-norm-LSTSVR model improved based on the greedy formula is used as the fitting model of the coal consumption characteristic curve, and the specific steps are as follows: Obtain sample data of factors affecting the power supply coal consumption of the unit; Preprocess the collected samples and divide them into training samples and test samples; Fit the dynamic coal consumption characteristic curve by the SCSO-L1-norm-LSTSVR model improved based on the greedy formula and the K-means clustering-least squares method.

4. The method of claim 3, wherein, The sample data of the factors affecting the power supply coal consumption of the unit are acquired, and specifically include coal quality, flue gas oxygen content, primary air volume, secondary air volume, feed water flow, feed water temperature, flue gas temperature, superheating temperature reducing water flow, reheating temperature reducing water flow.

5. A system for analyzing the cost of dynamic coal consumption into the furnace, characterized by, Comprise The first calculation module: real-time effective identification of mixed batches of incoming coal, calculation of the mixing ratio of each batch of incoming coal in the mixed batch, and obtaining the actual price of the incoming coal; specifically comprising the following steps: Collecting the coal quality test results of all batches of incoming coal of the thermal power plant, forming a database of the coal quality test results of all batches of incoming coal, denoting the total number of batches of incoming coal currently in the coal bunker as m, and denoting the batches of incoming coal as a set represented as {a1, a2, … a m} where a i (i = 1, 2,... m) represents the incoming coal labeled as the ith batch, Real-time acquisition of the heat value Q and sulfur content S of the coal burned by the incoming coal, and the relationship between the heat value and sulfur content of the incoming coal and each batch of incoming coal is as follows: Wherein, Qi (i=1, 2, …m) represents the heat value of the i-th batch of incoming coal, S i (i=1, 2, …m) represents the sulfur content of the i-th batch of incoming coal, x i represents the blending ratio of the i-th batch of incoming coal in the incoming coal for this calculation. The mixing ratio of each batch of incoming coal in the incoming coal is calculated by an evolutionary algorithm; The cost of the incoming coal is obtained according to the mixing ratio of each batch of incoming coal in the incoming coal, and is specifically expressed as: Wherein, C i (i=1, 2, …m) represents the unit price of the i batch of incoming coal, C represents the actual unit cost of the incoming coal, C SC represents the price of the incoming coal converted to standard coal. The characteristic curve fitting module: fitting the dynamic coal consumption characteristic curve according to the collected real-time power data and coal consumption data of the unit, and obtaining the dynamic relationship between real-time power and coal consumption; The second calculation module: calculating the real-time incoming dynamic coal consumption cost of the thermal power plant according to the incoming coal cost and the dynamic coal consumption characteristic curve; The second calculation module specifically includes: marginal cost calculation, variable cost calculation, and monthly average power generation cost calculation, The marginal cost calculation includes: wherein, is the slope of the dynamic coal consumption characteristic curve, representing the marginal coal consumption for power generation, is the coal consumption cost; is the amount of the fth pollutant generated per unit of power generation; is the fth pollutant emission charge standard; represents the total number of items that need to be charged for pollution treatment, is the pollution treatment cost per unit of power generation; W is the water consumption per unit of power generation; is the unit water price, is the water consumption cost per unit of power generation; is the auxiliary machine operating power corresponding to a unit of power generation, is the self-use electricity price of the power plant, is the auxiliary machine electricity consumption cost; is the maintenance cost per unit of power generation; The variable cost calculation includes: wherein, and are the first hourly coal consumption and power generation, is the output power of the generator set at time t. The monthly average power generation cost calculation includes: wherein, represents the day of the month, represents the number of days in the statistical month, represents the monthly fixed cost.

6. The system for analyzing dynamic coal consumption cost upon entry into a furnace according to claim 5, wherein The mixing ratio of each batch of incoming coal in the incoming coal is calculated by an evolutionary algorithm, specifically including: Step 1: parameter initialization setting, reading relevant economic parameters of the fuel management system, including the heat value Qi and sulfur content Si of each batch of incoming coal, and the heat value Q and sulfur content S of the incoming coal; Step 2: initialize the population of the genetic algorithm, randomly generate K possible genetic individuals, and take them as the initial population; Step 3: set the fitness function of the genetic algorithm, and evaluate the fitness of the initial population U; Step 4: sort the initial population U, calculate the fitness function between each individual, and calculate the population crowding degree; Step 5: update the initial population U of the genetic algorithm, and obtain new offspring populations U2, U3 through selection, crossover and mutation operations; Step 6: update the fitness of the population U3; Step 7: sort the initial population {U, U3}; Step 8: select the individual with the highest ranking corresponding to the minimum target function from the initial population {U, U3}, and form a new population Uz; Step 9: if the maximum number of iterations is reached, stop running, otherwise return to step 5; Step 10: take the obtained solution as the optimal solution of the multi-objective programming problem, that is, the mixing ratio of each batch of incoming coal.

7. The system for analyzing dynamic coal consumption cost upon entering a furnace according to claim 5, wherein, The characteristic curve fitting module includes: An SCSO-L1-norm-LSTSVR model improved based on the greedy formula is used as the coal consumption characteristic curve fitting model, and the specific steps are as follows: Acquire sample data of factors affecting the power supply coal consumption of the unit; The collected samples are preprocessed and divided into training samples and test samples; The dynamic coal consumption characteristic curve is fitted by the SCSO-L1-norm-LSTSVR model improved based on the greedy formula and the K-means clustering-least squares method.

8. The system for analyzing dynamic coal consumption cost upon entering a furnace according to claim 7, wherein, The sample data of the factors affecting the power supply coal consumption of the unit are acquired, and specifically include coal quality, flue gas oxygen content, primary air volume, secondary air volume, feed water flow, feed water temperature, exhaust gas temperature, superheating desuperheating water flow, reheating desuperheating water flow.

Images