Method and device for cutting off coal-fired unit based on boiler combustion state soft measurement

Abstract

The invention discloses a method and device for cutting off a coal-fired unit based on boiler combustion state soft measurement, and the method comprises the steps: obtaining the structural data of a hearth, and building a grid model; according to the gridding model, establishing a reaction model in the hearth, and obtaining three-dimensional distribution data of a hearth temperature field, an O2 concentration field and a CO concentration field under various operation conditions through numerical simulation; training and verifying a BP neural network by taking the three-dimensional distribution data under the multiple operation conditions as output and taking the corresponding unit operation parameters under the multiple operation conditions as input to obtain a temperature component field prediction model; calculating the combustion stability coefficient of the hearth based on the predicted real-time three-dimensional distribution data of the temperature field, the O2 concentration field and the CO concentration field by using a secondary fuzzy comprehensive evaluation method; and after a thermal power generating unit cutting instruction is received, the real-time hearth combustion stability coefficients of all the units in the power station are sequenced, and the unit with the minimum combustion stability coefficient of the station is preferentially cut off. The method can effectively reduce the economic loss of coal-fired power plant generator tripping.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic control of power grid security and stability, and in particular to a method and device for removing a coal-fired unit based on soft measurement of boiler combustion state. Background Art

[0002] Despite the large-scale grid integration of renewable energy sources such as wind and solar, and a decline in the proportion of thermal power installed capacity, thermal power still accounts for the majority of my country's electricity generation, accounting for approximately 70% of annual power generation. Therefore, thermal power will remain the primary executor of the power supply side of the grid's safety and stability control system for some time to come.

[0003] Coal-fired units are the primary form of thermal power generation, and optimizing their stabilization and control tripping strategies is crucial for improving the efficiency of stabilization and control systems. Currently, coal-fired unit tripping decisions are primarily based on unit load, while other operating parameters are underutilized.

[0004] For example, the invention patent application with publication number CN118611078 A proposes a standardized regional safety and stability control method and system. This method includes obtaining information about target elements in a stability control system, matching this information with a constant value table corresponding to the control strategy, and determining the strategy type for the stability control system's actions; generating corresponding unit or load shedding control measures based on the unit or load shedding plan constant value table; and executing the unit or load shedding control measures. However, the unit or load shedding plan relies on a pre-designed control strategy constant value table rather than real-time operating parameters, making it difficult to achieve an economically optimal unit or load shedding measure.

[0005] Therefore, it is necessary to improve the existing stabilization and control system cutting measures. Summary of the Invention

[0006] The purpose of the present invention is to address the technical defects existing in the prior art and provide a method and device for cutting off coal-fired units based on soft measurement of boiler combustion status. The method aims to optimize the stable control cutting strategy by integrating soft measurement parameters of the operating status of the coal-fired units, thereby achieving the goal of reducing the cutting cost of coal-fired power plants.

[0007] The technical solution adopted to achieve the purpose of the present invention is:

[0008] According to a first aspect of the present invention, a method for removing a coal-fired unit based on soft measurement of boiler combustion status is provided. The method comprises the following steps:

[0009] Obtain the structural data of the furnace and establish a grid model;

[0010] Based on the grid model, a reaction model inside the furnace was established, and the three-dimensional distribution data of the furnace temperature field, O2 concentration field, and CO concentration field under various operating conditions were obtained through numerical simulation.

[0011] The BP neural network is trained and verified using the three-dimensional distribution data of the temperature field, O2 concentration field, and CO concentration field under the multiple operating conditions as output and the unit operating parameters corresponding to the multiple operating conditions as input to obtain a temperature component field prediction model for predicting the real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field, and CO concentration field based on the monitored real-time operating conditions of the thermal power unit;

[0012] A two-level fuzzy comprehensive evaluation method is used to calculate the real-time furnace combustion stability coefficient based on the predicted real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field, and CO concentration field.

[0013] After receiving the thermal power unit cut-off instruction, the real-time furnace combustion stability coefficients of each unit are sorted, and one or more units with the smallest combustion stability coefficients are preferentially cut off until the load cut-off demand is met.

[0014] In the above technical solution, the reaction model inside the furnace includes: a gas phase turbulence model inside the furnace, a gas-solid two-phase flow model, a particle pyrolysis model, a combustion model and a radiation heat transfer model; the numerical simulation adopts coal quality parameters, boiler coal feed rate, boiler primary and secondary air flow rate and boiler burnout air flow rate under multiple operating conditions, and through the established reaction model inside the furnace, obtains the three-dimensional distribution data of the furnace temperature field, O2 concentration field and CO concentration field under the corresponding operating conditions; the unit operating parameters include: boiler load, coal quality parameters, boiler coal feed rate, boiler secondary air door opening and boiler primary and secondary air flow rate.

[0015] In the above technical solution, the BP neural network selects Logsig as the neuron activation function of the input layer, selects Tansig as the neuron activation function of the hidden layer, selects Purelin as the neuron activation function of the output layer, selects Trainlm function as the training function of the BP neural network, selects Learngdm function as the deviation learning function, and selects Mse function as the performance evaluation function.

[0016] In the above technical solution, the training and verification of the BP neural network also includes: installing multiple temperature, O2 measuring points and CO measuring points at the furnace fire viewing port, and verifying the accuracy of the real-time three-dimensional distribution data of the predicted temperature field, O2 concentration field and CO concentration field through field tests.

[0017] In the above technical solution, the two-level fuzzy comprehensive evaluation method is used to calculate the real-time furnace combustion stability coefficient based on the predicted real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field and CO concentration field, including:

[0018] Based on the predicted real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field, and CO concentration field, the following indicators at the current moment are calculated: the average temperature of the two-dimensional furnace temperature at the plane where the flame deflection angle is located, the temperature variance, the area ratio of the optimal temperature area, the average O2 concentration, the area ratio of the economic operating range of O2 concentration, the area ratio of the high O2 concentration range, the average CO concentration, the area ratio of the economic operating range of CO concentration, and the area ratio of the high CO concentration range;

[0019] According to the correlation between the various indicators and combustion stability, different membership functions are selected to normalize the various indicators at the current moment;

[0020] The temperature evaluation factor at the current moment is obtained by weighted summing the normalized temperature average, temperature variance, and optimal temperature area ratio at the current moment;

[0021] The normalized average O2 concentration, the area ratio of the economic operation range of O2 concentration, and the area ratio of the high O2 concentration range are weighted summed to obtain the O2 concentration evaluation factor at the current moment.

[0022] The CO concentration evaluation factor at the current moment is obtained by weighted summing the normalized average CO concentration, the CO concentration economic operation range area ratio, and the CO high concentration range area ratio.

[0023] The temperature evaluation factor, O2 concentration evaluation factor and CO concentration evaluation factor at the current moment are weighted and summed to obtain the furnace combustion stability coefficient at the current moment.

[0024] In the above technical solution, the average temperature x of the two-dimensional temperature of the furnace in the plane where the flame deflection angle is located is 11 It is obtained by the following formula:

[0025]

[0026] Where: t i is the temperature value at the center of the i-th grid, and n is the number of cross-sectional grids in the plane where the flame deflection angle is located;

[0027] Temperature variance x 12 It is obtained by the following formula:

[0028]

[0029] Optimal temperature area ratio x 13 It means that the temperature value in the furnace is in the optimal combustion temperature range of pulverized coal under the operating conditions [t a ,t b ] is the percentage of the number of grids in the plane to the total number of grids, and is obtained by the following formula:

[0030]

[0031] Among them, t a and t b are the lower and upper limits of the optimal combustion temperature range of pulverized coal, respectively. ε(·) is a step function. When the independent variable is x:

[0032]

[0033] Average O2 concentration x 21 It is obtained by the following formula:

[0034]

[0035] Where: is the O2 concentration value at the center of the i-th grid;

[0036] O2 concentration economic operation area ratio x 22 Refers to the O2 concentration value in the furnace Located in the economic operation range under operating conditions [C a ,C b The percentage of the number of grids in the plane to the total number of grids in the plane is calculated by the following formula:

[0037]

[0038] Among them, C a and C b They are the lower and upper limits of the economic operation range under operating conditions;

[0039] O2 high concentration range area ratio x 23 Refers to the O2 concentration value in the furnace Greater than the maximum threshold of CO concentration in the furnace The percentage of the number of grids in the total number of grids in the plane is calculated by the following formula:

[0040]

[0041] Average CO concentration x 31 By the following formula:

[0042]

[0043] Where: C CO,i is the CO concentration at the center of the i-th grid, and n is the number of horizontal cross-section grids;

[0044] CO concentration economic operation area ratio x 32 Refers to the CO concentration value C in the furnace CO,i Located in the economic operation range under operating conditions [C c ,Cd The percentage of the number of grids in the plane to the total number of grids in the plane is calculated by the following formula:

[0045]

[0046] Among them, C c and C d They are the lower and upper limits of the economic operation range under operating conditions;

[0047] CO high concentration range area ratio x 33 Refers to the CO concentration value C in the furnace CO,i Greater than the maximum threshold value of CO concentration in the furnace C CO,max The percentage of the number of grids in the total number of grids in the plane is calculated by the following formula:

[0048]

[0049] In the above technical solution, different membership functions are selected to normalize the various indicators at the current moment according to the correlation between the various indicators and combustion stability, including:

[0050] The trapezoidal membership function is selected to normalize the average temperature, average O2 concentration and average CO concentration of the two-dimensional furnace temperature in the plane where the flame fold angle is located;

[0051] A relatively large membership function was selected to normalize the area ratio of the optimal temperature region, the area ratio of the economic operation range of O2 concentration, and the area ratio of the economic operation range of CO concentration.

[0052] The temperature variance, the area ratio of high O2 concentration range, and the area ratio of high CO concentration range were normalized using a small membership function.

[0053] In the above technical solution, the trapezoidal membership function μ1(x) is expressed as:

[0054]

[0055] Among them, x is the index to be normalized; x min 、x max are the minimum and maximum values of the corresponding indicators to be normalized; x a 、x b are the lower and upper limits of the normalized index in the reasonable operating range respectively; x min 、x max 、x a and x b All are determined according to the unit design parameters.

[0056] In the above technical solution, the large-scale membership function μ2(x) is expressed as:

[0057]

[0058] Among them, x is the index to be normalized; x min 、x max are the minimum and maximum values of the corresponding indicators to be normalized, respectively, and are determined according to the unit design parameters.

[0059] In the above technical solution, the small-scale membership function μ3(x) is expressed as:

[0060]

[0061] Among them, x is the index to be normalized; x min 、x max are the minimum and maximum values of the corresponding indicators to be normalized, respectively, and are determined according to the unit design parameters.

[0062] According to a second aspect of the present invention, a device for removing a coal-fired unit based on soft measurement of boiler combustion status is provided. The device utilizes the method described in the first aspect of the present invention. The device comprises:

[0063] The gridding module is used to obtain the structural data of the furnace and establish a gridding model;

[0064] The sample acquisition module is used to establish a reaction model inside the furnace based on the grid model, and obtain the three-dimensional distribution data of the furnace temperature field, O2 concentration field and CO concentration field under various operating conditions through numerical simulation;

[0065] a training module for training and validating a BP neural network using the three-dimensional distribution data of the temperature field, O2 concentration field, and CO concentration field under the multiple operating conditions as output and the unit operating parameters corresponding to the multiple operating conditions as input, to obtain a temperature component field prediction model, and for predicting the real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field, and CO concentration field based on the monitored real-time operating conditions of the thermal power unit;

[0066] The evaluation module is used to calculate the real-time furnace combustion stability coefficient based on the predicted real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field and CO concentration field using a two-level fuzzy comprehensive evaluation method;

[0067] The shedding module is used to sort the real-time furnace combustion stability coefficients of each unit after receiving the shedding instruction of the thermal power unit, and give priority to shedding one or more units with the smallest combustion stability coefficients until the load shedding demand is met.

[0068] According to a third aspect of the present invention, a terminal is provided. The terminal includes a processor and a storage medium;

[0069] The storage medium is used to store instructions;

[0070] The processor is configured to operate according to the instructions to execute the steps of the method according to the first aspect of the present invention.

[0071] Compared with the prior art, the present invention has the following beneficial effects:

[0072] 1. The distribution of internal furnace temperature, O2 concentration and CO concentration under all working conditions can be obtained through numerical simulation, which can reduce the number of temperature, O2 concentration and CO concentration detection equipment installed.

[0073] 2. Through the BP neural network prediction method, the amount of numerical simulation calculations can be reduced and the calculation costs can be saved.

[0074] 3. Through the two-level fuzzy comprehensive evaluation method, a variety of coal-fired unit operating parameter indicators are integrated to make the calculation of furnace combustion stability coefficient more accurate.

[0075] 4. The furnace combustion stability coefficient is used to determine the units that need to be cut off for stabilization and control, and the units with poor combustion conditions are cut off first, which can reduce the loss of coal-fired power plants.

[0076] 5. By selecting appropriate membership functions to normalize each indicator separately, it can accurately reflect the characteristics of the indicator, enhance the flexibility of the evaluation, and improve the robustness of the evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 The present invention is a flow chart of a method for removing a coal-fired unit based on soft measurement of boiler combustion state.

[0078] Figure 2 It is a schematic diagram of a device for removing a coal-fired unit based on soft measurement of boiler combustion state according to the present invention. DETAILED DESCRIPTION

[0079] The present invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0080] Example 1

[0081] This embodiment provides a method for removing coal-fired units based on soft measurement of boiler combustion state. This method can be used in the stability control system of a thermal power plant. In one embodiment, Figure 1 As shown, the method includes the following steps:

[0082] Step 1: Obtain the structural data of the furnace and establish a grid model.

[0083] This step is mainly achieved using geometric modeling and meshing software.

[0084] Specifically, the furnace structural data can be obtained from the boiler manufacturer or on-site, and may include information such as structural dimensions and material properties. After obtaining the furnace structural data, a three-dimensional geometric model of the furnace can be created based on the acquired structural data using geometric modeling software (such as AutoCAD, SolidWorks, Pro / E, etc.). Next, the geometric three-dimensional model is imported into meshing software. Based on the complexity of the furnace and the computational requirements, the appropriate mesh type and mesh size are selected to create a meshed model. Commonly used meshing software includes ANSYS ICEM CFD, HyperMesh, Gridgen, etc.

[0085] Step 2: Based on the grid model, a reaction model inside the furnace is established, and the three-dimensional distribution data of the furnace temperature field, O2 concentration field and CO concentration field under various operating conditions are obtained through numerical simulation.

[0086] The reaction models within the furnace include: a gas-phase turbulence model, a gas-solid two-phase flow model, a particle pyrolysis model, a combustion model, and a radiation heat transfer model. These models can be established using CFD simulation software based on a gridded model. Specific model formulas are known in the art and can be found in relevant literature in this field, so they will not be detailed here.

[0087] After establishing the above model, numerical simulation can be used to obtain three-dimensional distribution data for the furnace temperature field, O2 concentration field, and CO concentration field under various operating conditions. This numerical simulation can be performed using, for example, CFD numerical simulation methods. The numerical simulation uses coal quality parameters, boiler coal feed rate, boiler primary and secondary air flow rates, and boiler burnout air flow rates under various operating conditions. By using the established reaction model within the furnace, three-dimensional distribution data for the furnace temperature field, O2 concentration field, and CO concentration field under the corresponding operating conditions are obtained.

[0088] Step 3: Using the three-dimensional distribution data of the temperature field, O2 concentration field, and CO concentration field under the various operating conditions as output and the unit operating parameters corresponding to the various operating conditions as input, the BP neural network is trained and verified to obtain a temperature component field prediction model, which is used to predict the real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field, and CO concentration field based on the real-time operating conditions of the monitored thermal power unit.

[0089] Specifically, the specific temperature values, O2 concentration values, and CO concentration values at the custom coordinate positions are extracted from the three-dimensional distribution data of the temperature field, O2 concentration field, and CO concentration field under various operating conditions obtained by numerical simulation, and the position coordinates, temperature data, O2 concentration data, and CO concentration data are arranged in a fixed order to obtain an arithmetic matrix of the position coordinates and temperature, O2 concentration, and CO concentration. The unit operating parameters are then added to the arithmetic matrix to obtain a data set corresponding to the temperature field, O2 concentration field, and CO concentration field, and the BP neural network is trained using this data set.

[0090] Furthermore, the unit operating parameters corresponding to various operating conditions may include boiler load, coal quality parameters, boiler coal feed rate, boiler secondary air damper opening, boiler primary and secondary air flow, etc. These unit operating parameters can be used as inputs of the BP neural network.

[0091] By training the BP neural network with the above input and output data, a combustion state prediction model based on the BP neural network can be obtained. This combustion state prediction model can be used to calculate the three-dimensional distribution data of temperature field, O2 concentration field, and CO concentration under all operating conditions.

[0092] In the BP neural network, Logsig is selected as the neuron activation function of the input parameter, Tansig is selected as the neuron activation function of the hidden layer, Purelin is selected as the neuron activation function of the output layer, the back propagation algorithm Trainlm function is selected as the training function of the neural network, the deviation learning function selects the gradient descent weight learning algorithm Learndm function with an additional momentum factor, and the performance evaluation function selects the mean square error performance algorithm Mse function.

[0093] Preferably, multiple temperature, O2 and CO measuring points need to be installed at the furnace fire viewing port to verify the accuracy of the predicted three-dimensional distribution data of temperature field, O2 concentration field and CO concentration field through field tests.

[0094] Step 4: Use the two-level fuzzy comprehensive evaluation method to calculate the real-time furnace combustion stability coefficient based on the predicted real-time three-dimensional distribution data of the temperature field, O2 concentration field and CO concentration field.

[0095] According to the two-level fuzzy comprehensive evaluation method, its basic idea is:

[0096] (1) The boiler furnace combustion stability evaluation index is divided into three factors according to the attributes, P = (p1, p2, p3), each factor p i It also contains three secondary evaluation factors p i =(p i1 ,p i2 ,p i3 ).

[0097] (2) For each factor p of the evaluation index system i , according to the single-level evaluation model, its three evaluation sub-factors are evaluated, and the sub-factor weight coefficient c i =(c i1 ,c i2 ,c i3 ), satisfying ∑c ij =1, by s i =c i ·p i T The evaluation vector of each factor is obtained as s=(s1,s2,s3). ij It can be set by experts in the industry or determined using the analytic hierarchy process.

[0098] (3) For the three evaluation factors p i , i=(1,2,3) for comprehensive evaluation, its factor weight coefficient c=(c1,c2,c3), satisfying ∑c i =1, from B = c·s T Get the judgment value B. Among them, c i It can be set by experts in the industry or determined using the analytic hierarchy process.

[0099] Specifically, this step includes:

[0100] Step 41. Based on the predicted real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field and CO concentration field, calculate and obtain the following indicators at the current moment: the temperature average value, temperature variance, optimal temperature area ratio, O2 concentration average value, O2 concentration economic operation range area ratio, O2 high concentration range area ratio, CO concentration average value, CO concentration economic operation range area ratio and CO high concentration range area ratio.

[0101] In this step, the calculation method of each indicator value is as follows:

[0102] (1) The average temperature of the two-dimensional furnace temperature in the plane where the flame deflection angle is located x 11 Expressed as:

[0103]

[0104] Where: t is the temperature value at the center of the grid, and n is the number of horizontal cross-section grids.

[0105] (2) Furnace temperature variance x 12 Expressed as:

[0106]

[0107] (3) Optimal temperature area ratio:

[0108] The temperature value in the furnace is in the optimal combustion temperature range of pulverized coal under the operating conditions [t a ,t b The percentage of the number of grids in the plane to the total number of grids is the optimal temperature area ratio. The optimal combustion temperature range of pulverized coal under operating conditions is usually a dynamic range and needs to be adjusted according to various factors such as coal type, coal fineness, concentration, and combustion conditions. In actual applications, it is usually necessary to determine the optimal combustion temperature range through experiments and practice. The optimal temperature area ratio x 13 Expressed as:

[0109]

[0110] Among them, t a and t b are the lower and upper limits of the optimal combustion temperature range of pulverized coal, respectively. ε(x) is a step function. The specific formula is as follows:

[0111]

[0112] (4) Average O2 concentration in the furnace x 21 Expressed as:

[0113]

[0114] Where: is the O2 concentration value at the center of the grid, and n is the number of horizontal cross-section grids.

[0115] (5) O2 concentration economic operation area ratio:

[0116] During the operation of coal-fired units, the O2 concentration needs to be maintained within a certain range. Too low a concentration will lead to incomplete combustion of pulverized coal, and too high a concentration will cause increased heat loss of the unit. The reasonable range of O2 concentration is the economic operating area (or interval) under the operating conditions of O2 concentration.

[0117] O2 concentration in the furnace Located in the economic operation range under operating conditions [C a ,C b The percentage of the number of grids in the plane to the total number of grids is the area rate of the economic operation area of O2 concentration x 22 , expressed as:

[0118]

[0119] Among them, C a and C b They are respectively the lower limit and upper limit of the economic operation range under operating conditions.

[0120] (6) Area ratio of high O2 concentration range:

[0121] O2 concentration in the furnace Greater than the maximum threshold of CO concentration in the furnace The percentage of the number of grids in the plane to the total number of grids is the area ratio of high CO concentration range x 23 , expressed as:

[0122]

[0123] (7) Average CO concentration in the furnace x 31 Expressed as:

[0124]

[0125] Where: C CO,i is the temperature value at the center of the grid, and n is the number of horizontal cross-section grids.

[0126] (8) CO concentration economic operation area ratio:

[0127] Similarly, during the operation of coal-fired units, the CO concentration must also be maintained within a certain range. The reasonable range of CO concentration is the economic operating area (or interval) under the operating conditions of CO concentration.

[0128] CO concentration in the furnace C CO,i Located in the economic operation range under operating conditions [C c ,C d The percentage of the number of grids in ] to the total number of grids in the plane is the area rate of the economic operation area of O2 concentration.

[0129]

[0130] Among them, C c and C d They are respectively the lower limit and upper limit of the economic operation range under operating conditions.

[0131] (9) Area ratio of high CO concentration range:

[0132] CO concentration in the furnace C CO,i Greater than the maximum threshold value of CO concentration in the furnace C CO,max The percentage of the number of grids in the total number of grids in the plane is the area ratio of high CO concentration range.

[0133]

[0134] Step 42: According to the correlation between the various indicators and the combustion stability, different membership functions are selected to normalize the various indicators at the current moment.

[0135] Since the dimensions of the sub-factor indicators are different, it is necessary to select the membership function according to their correlation with combustion stability, normalize them separately, and then perform the secondary fuzzy comprehensive evaluation calculation.

[0136] Among them, the average temperature, average O2 concentration, and average CO concentration index values achieve the best results within a specific range, so these indicators are normalized using a trapezoidal membership function. The trapezoidal membership function (denoted as μ1(x)) is expressed as:

[0137]

[0138] The larger the area ratio of the optimal temperature region, the area ratio of the economic operating range of O2 concentration, and the area ratio of the economic operating range of CO concentration, the more conducive it is to improving system performance. Therefore, a large-scale membership function is used to normalize these indicators. Among them, the large-scale membership function (denoted as μ2(x)) is expressed as:

[0139]

[0140] The smaller the values of temperature variance, O2 high concentration area ratio, and CO high concentration area ratio, the more conducive to improving system performance. Therefore, a small-scale membership function is selected to normalize these indicators. Among them, the small-scale membership function (denoted as μ3(x)) is expressed as:

[0141]

[0142] In the above trapezoidal membership function μ1(x), the relatively large membership function μ2(x), and the relatively small membership function μ3(x), x is the index to be normalized; x min 、x max are the minimum and maximum values of the corresponding indicators to be normalized; x a 、x b are the lower and upper limits of the normalized index when the unit is in the reasonable operating range; x min 、x max 、x a and x b The reasonable operating range is determined according to the unit design parameters.

[0143] By selecting appropriate membership functions to normalize each indicator separately, we can accurately reflect the characteristics of the indicators, enhance the flexibility of the evaluation, and improve the robustness of the evaluation.

[0144] Step 43: Perform weighted summation on the normalized temperature average value, temperature variance, and optimal temperature region area ratio at the current moment to obtain the temperature evaluation factor at the current moment.

[0145] Among them, when performing weighted summation, the sum of the weight coefficients corresponding to the normalized temperature average, temperature variance, and optimal temperature area ratio is 1.

[0146] Step 44: Perform weighted summation on the normalized average O2 concentration, the area ratio of the economic operation range of O2 concentration, and the area ratio of the high O2 concentration range at the current moment to obtain the O2 concentration evaluation factor at the current moment.

[0147] Among them, when performing weighted summation, the sum of the weight coefficients corresponding to the normalized average O2 concentration, the area ratio of the economic operation range of O2 concentration, and the area ratio of the high O2 concentration range is 1.

[0148] Step 45 : Perform a weighted summation of the normalized average CO concentration, the CO concentration economic operation range area ratio, and the CO high concentration range area ratio at the current moment to obtain a CO concentration evaluation factor at the current moment.

[0149] Among them, when performing weighted summation, the sum of the weight coefficients corresponding to the normalized average CO concentration, the area ratio of the CO concentration economic operation range, and the area ratio of the CO high concentration range is 1.

[0150] Step 46: Perform weighted summation on the temperature evaluation factor, O2 concentration evaluation factor, and CO concentration evaluation factor at the current moment to obtain the furnace combustion stability coefficient at the current moment.

[0151] Among them, when performing weighted summation, the sum of the weight coefficients corresponding to the temperature evaluation factor, the O2 concentration evaluation factor and the CO concentration evaluation factor is 1.

[0152] Step 5: After receiving the thermal power unit cut-off instruction, sort the real-time furnace combustion stability coefficients of the units in the power station from small to large, and give priority to cutting off the unit with the smallest combustion stability coefficient until the load cutting demand is met.

[0153] In this step, the shedding principle is to shear off the running units from small to large according to the combustion stability coefficient calculated in combination with the soft measurement parameters of the units until the load shedding requirement is met, where the combustion stability coefficient is a real number in the interval (0, 1).

[0154] By accurately judging the combustion stability of the units through the combustion stability coefficient and giving priority to cutting off the units with poor combustion conditions, the losses caused by cutting off the units in coal-fired power plants can be reduced.

[0155] Example 2

[0156] This embodiment provides a device for removing a coal-fired unit based on the soft measurement of boiler combustion state using the method described in Example 1. Figure 2 , the device comprises:

[0157] The gridding module is used to obtain the structural data of the furnace and establish a gridding model;

[0158] The sample acquisition module is used to establish a reaction model inside the furnace based on the grid model, and obtain the three-dimensional distribution data of the furnace temperature field, O2 concentration field and CO concentration field under various operating conditions through numerical simulation;

[0159] a training module for training and validating a BP neural network using the three-dimensional distribution data of the temperature field, O2 concentration field, and CO concentration field under the multiple operating conditions as output and the unit operating parameters corresponding to the multiple operating conditions as input, thereby obtaining a temperature component field prediction model for predicting the real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field, and CO concentration field based on the monitored real-time operating conditions of the thermal power unit;

[0160] The evaluation module is used to calculate the real-time furnace combustion stability coefficient based on the predicted real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field and CO concentration field using a two-level fuzzy comprehensive evaluation method;

[0161] The shedding module is used to sort the real-time furnace combustion stability coefficients of each unit after receiving the shedding instruction of the thermal power unit, and give priority to shedding one or more units with the smallest combustion stability coefficients until the load shedding demand is met.

[0162] Example 3

[0163] Embodiment 3 of the present invention provides a computer-readable storage medium.

[0164] A computer-readable storage medium stores a program thereon, which, when executed by a processor, implements the steps of the method for removing a coal-fired unit based on soft measurement of boiler combustion state as described in embodiment 1 of the present invention.

[0165] The detailed steps are the same as those of the method for removing a coal-fired unit based on soft measurement of boiler combustion state provided in Example 1, and are not described again here.

[0166] The beneficial effects of the present invention are as follows:

[0167] 1. The distribution of internal furnace temperature, O2 concentration and CO concentration under all working conditions can be obtained through numerical simulation, which can reduce the number of temperature, O2 concentration and CO concentration detection equipment installed.

[0168] 2. Through the BP neural network prediction method, the amount of numerical simulation calculations can be reduced and the calculation costs can be saved.

[0169] 3. Through the two-level fuzzy comprehensive evaluation method, a variety of coal-fired unit operating parameter indicators are integrated to make the calculation of furnace combustion stability coefficient more accurate.

[0170] 4. The furnace combustion stability coefficient is used to determine the units that need to be cut off for stabilization and control, and the units with poor combustion conditions are cut off first, which can reduce the loss of coal-fired power plants.

[0171] 5. By selecting appropriate membership functions to normalize each indicator separately, it can accurately reflect the characteristics of the indicator, enhance the flexibility of the evaluation, and improve the robustness of the evaluation.

[0172] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0173] Computer-readable storage medium can be a tangible device that can keep and store the instruction used by the instruction execution device.Computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device or any suitable combination thereof.The more specific example (non-exhaustive list) of computer-readable storage medium includes: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, for example, a punch card or a convex structure in a groove having instructions stored thereon and any suitable combination thereof.Computer-readable storage medium used herein is not interpreted as a transient signal itself, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagated by waveguides or other transmission media (for example, light pulses by fiber optic cables) or electrical signals transmitted by wires.

[0174] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0175] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages such as Smalltalk, C++, and conventional procedural programming languages such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., utilizing an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions. The electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0176] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for removing coal-fired units based on soft measurement of boiler combustion state, characterized in that: The following steps are involved: Obtain the structural data of the furnace and establish a grid model; Based on the grid model, a reaction model inside the furnace was established, and the three-dimensional distribution data of the furnace temperature field, O2 concentration field, and CO concentration field under various operating conditions were obtained through numerical simulation. The BP neural network is trained and verified using the three-dimensional distribution data of the temperature field, O2 concentration field, and CO concentration field under the multiple operating conditions as output and the unit operating parameters corresponding to the multiple operating conditions as input to obtain a temperature component field prediction model for predicting the real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field, and CO concentration field based on the monitored real-time operating conditions of the thermal power unit; A two-level fuzzy comprehensive evaluation method is used to calculate the real-time furnace combustion stability coefficient based on the predicted real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field, and CO concentration field. After receiving the thermal power unit cut-off instruction, the real-time furnace combustion stability coefficients of each unit are sorted, and one or more units with the smallest combustion stability coefficients are preferentially cut off until the load cut-off demand is met.

2. The method for removing a coal-fired unit based on soft measurement of boiler combustion state according to claim 1, characterized in that: The reaction model inside the furnace includes: a gas phase turbulence model inside the furnace, a gas-solid two-phase flow model, a particle pyrolysis model, a combustion model and a radiation heat transfer model; The numerical simulation uses coal quality parameters, boiler coal feed rate, boiler primary and secondary air flow rate, and boiler burnout air flow rate under various operating conditions, and obtains three-dimensional distribution data of furnace temperature field, O2 concentration field, and CO concentration field under corresponding operating conditions through the established reaction model inside the furnace; The unit operating parameters include: boiler load, coal quality parameters, boiler coal feed rate, boiler secondary air door opening and boiler primary and secondary air flow rates.

3. The method for removing a coal-fired unit based on soft measurement of boiler combustion state according to claim 1, characterized in that: The BP neural network uses Logsig as the neuron activation function of the input layer, Tansig as the neuron activation function of the hidden layer, Purelin as the neuron activation function of the output layer, Trainlm function as the training function of the BP neural network, Learngdm function as the deviation learning function, and Mse function as the performance evaluation function.

4. The method for removing a coal-fired unit based on soft measurement of boiler combustion state according to claim 1, characterized in that: The BP neural network is trained and verified, and further comprises: Multiple temperature, O2 and CO measuring points are installed at the furnace fire viewing port, and the accuracy of the predicted real-time three-dimensional distribution data of the temperature field, O2 concentration field and CO concentration field is verified through field tests.

5. The method for removing a coal-fired unit based on soft measurement of boiler combustion state according to claim 1, characterized in that: The two-level fuzzy comprehensive evaluation method is used to calculate the real-time furnace combustion stability coefficient based on the predicted real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field and CO concentration field, including: Based on the predicted real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field, and CO concentration field, the following indicators at the current moment are calculated: the average temperature of the two-dimensional furnace temperature at the plane where the flame deflection angle is located, the temperature variance, the area ratio of the optimal temperature area, the average O2 concentration, the area ratio of the economic operating range of O2 concentration, the area ratio of the high O2 concentration range, the average CO concentration, the area ratio of the economic operating range of CO concentration, and the area ratio of the high CO concentration range; According to the correlation between the various indicators and combustion stability, different membership functions are selected to normalize the various indicators at the current moment; The temperature evaluation factor at the current moment is obtained by weighted summing the normalized temperature average, temperature variance, and optimal temperature area ratio at the current moment; The normalized average O2 concentration, the area ratio of the economic operation range of O2 concentration, and the area ratio of the high O2 concentration range are weighted summed to obtain the O2 concentration evaluation factor at the current moment. The CO concentration evaluation factor at the current moment is obtained by weighted summing the normalized average CO concentration, the CO concentration economic operation range area ratio, and the CO high concentration range area ratio. The temperature evaluation factor, O2 concentration evaluation factor and CO concentration evaluation factor at the current moment are weighted and summed to obtain the furnace combustion stability coefficient at the current moment.

6. The method for removing a coal-fired unit based on soft measurement of boiler combustion state according to claim 5, characterized in that: The average temperature x of the two-dimensional furnace temperature in the plane where the flame deflection angle is located 11 It is obtained by the following formula: Where: t i is the temperature value at the center of the i-th grid, and n is the number of cross-sectional grids in the plane where the flame deflection angle is located; Temperature variance x 12 It is obtained by the following formula: Optimal temperature area ratio x 13 It means that the temperature value in the furnace is in the optimal combustion temperature range of pulverized coal under the operating conditions [t a ,t b ] is the percentage of the number of grids in the plane to the total number of grids, and is obtained by the following formula: Among them, t a and t b are the lower and upper limits of the optimal combustion temperature range of pulverized coal, respectively. ε(·) is a step function. When the independent variable is x: Average O2 concentration x 21 It is obtained by the following formula: Where: is the O2 concentration value at the center of the i-th grid; O2 concentration economic operation area ratio x 22 Refers to the O2 concentration value in the furnace Located in the economic operation range under operating conditions [C a ,C b The percentage of the number of grids in the plane to the total number of grids in the plane is calculated by the following formula: Among them, C a and C b They are the lower and upper limits of the economic operation range under operating conditions; O2 high concentration range area ratio x 23 Refers to the O2 concentration value in the furnace Greater than the maximum threshold of CO concentration in the furnace The percentage of the number of grids in the total number of grids in the plane is calculated by the following formula: Average CO concentration x 31 By the following formula: Where: C CO,i is the CO concentration at the center of the i-th grid, and n is the number of horizontal cross-section grids; CO concentration economic operation area ratio x 32 Refers to the CO concentration value C in the furnace CO,i Located in the economic operation range under operating conditions [C c ,C d The percentage of the number of grids in the plane to the total number of grids in the plane is calculated by the following formula: Among them, C c and C d They are the lower and upper limits of the economic operation range under operating conditions; CO high concentration range area ratio x 33 Refers to the CO concentration value C in the furnace CO,i Greater than the maximum threshold value of CO concentration in the furnace C CO,max The percentage of the number of grids in the total number of grids in the plane is calculated by the following formula:

7. The method for removing a coal-fired unit based on soft measurement of boiler combustion state according to claim 5, characterized in that: According to the correlation between the various indicators and combustion stability, different membership functions are selected to normalize the various indicators at the current moment, including: The trapezoidal membership function is selected to normalize the average temperature, average O2 concentration and average CO concentration of the two-dimensional furnace temperature in the plane where the flame fold angle is located; A relatively large membership function was selected to normalize the area ratio of the optimal temperature region, the area ratio of the economic operation range of O2 concentration, and the area ratio of the economic operation range of CO concentration. A relatively small membership function is selected to normalize the temperature variance, the area ratio of high O2 concentration range, and the area ratio of high CO concentration range.

8. The method for removing a coal-fired unit based on soft measurement of boiler combustion state according to claim 7, characterized in that: The trapezoidal membership function μ1(x) is expressed as: Among them, x is the index to be normalized; x min 、x max are the minimum and maximum values of the corresponding indicators to be normalized; x a 、x b are the lower and upper limits of the normalized index in the reasonable operating range respectively; x min 、x max 、x a and x b All are determined according to the unit design parameters.

9. The method for removing a coal-fired unit based on soft measurement of boiler combustion state according to claim 7, characterized in that: The large-scale membership function μ2(x) is expressed as: Among them, x is the index to be normalized; x min 、x max They are the minimum and maximum values of the corresponding indicators, which are determined according to the unit design parameters.

10. The method for removing a coal-fired unit based on soft measurement of boiler combustion state according to claim 7, characterized in that: The small-scale membership function μ3(x) is expressed as: Among them, x is the index to be normalized; x min 、x max are the minimum and maximum values of the corresponding indicators to be normalized, which are determined according to the unit design parameters.

11. A device for removing a coal-fired unit based on soft measurement of boiler combustion state using the method according to any one of claims 1 to 10, characterized in that: include: The gridding module is used to obtain the structural data of the furnace and establish a gridding model; The sample acquisition module is used to establish a reaction model inside the furnace based on the grid model, and obtain the three-dimensional distribution data of the furnace temperature field, O2 concentration field and CO concentration field under various operating conditions through numerical simulation; a training module for training and validating a BP neural network using the three-dimensional distribution data of the temperature field, O2 concentration field, and CO concentration field under the multiple operating conditions as output and the unit operating parameters corresponding to the multiple operating conditions as input, to obtain a temperature component field prediction model, and for predicting the real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field, and CO concentration field based on the monitored real-time operating conditions of the thermal power unit; The evaluation module is used to calculate the real-time furnace combustion stability coefficient based on the predicted real-time three-dimensional distribution data of the furnace temperature field, O2 concentration field and CO concentration field using a two-level fuzzy comprehensive evaluation method; The shedding module is used to sort the real-time furnace combustion stability coefficients of each unit after receiving the shedding instruction of the thermal power unit, and give priority to shedding one or more units with the smallest combustion stability coefficients until the load shedding demand is met.

12. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1 to 10.

Images