Circulating fluidized bed boiler main steam outlet temperature control distribution method, equipment, medium and product

Abstract

The present application discloses a circulating fluidized bed boiler main steam outlet temperature control distribution method, equipment, medium and product, and relates to the field of circulating fluidized bed boilers. The method determines the modeling features that affect the main steam outlet temperature and obtains the modeling feature observation sequence; sets the target value of the main steam outlet temperature at a preset time in the future; obtains all the first and second level flow combination values ​​in the combination space formed by the first and second level cooling water flow search space, and constitutes the values ​​of multiple modeling features at the next moment; thereby using the variable TFT timing prediction model to output the predicted value of the main steam outlet temperature at a preset time in the future; according to the predicted value and target value of the main steam outlet temperature at a preset time in the future, selects the first and second level flow combination value that minimizes the loss function, and adjusts the first level cooling water flow and / or the switch state of the second level cooling water valve at the next moment. The present application can realize that both the first level cooling water flow and the second level cooling water flow are controlled for the main steam outlet temperature.

Description

Technical Field

[0001] The present application relates to the field of circulating fluidized bed boilers, and in particular to a method, equipment, medium and product for controlling and distributing the main steam outlet temperature of a circulating fluidized bed boiler. Background Art

[0002] Platen heating surfaces are widely used in the furnace of circulating fluidized bed boilers to maintain reasonable bed temperature and furnace outlet temperature. Platen superheaters are also commonly called semi-radiant superheaters, which are arranged at the upper part of the furnace or at the smoke window at the furnace outlet, and are heating surfaces that receive both direct radiation heat from the furnace and convection heat from the flue gas. Keeping a large distance between two adjacent platens of the platen superheater can reduce the flue gas temperature at the furnace outlet and slag condensation, and prevent slagging on the heating surface behind. At the same time, it is also the main component of the heating surface of the circulating fluidized bed boiler superheater.

[0003] If the outlet temperature of the first and second stage platen superheaters of circulating fluidized bed boilers is too high or too low, it will affect the equipment life, safety, economic benefits and other aspects. Therefore, it is very important to keep the outlet temperature of the platen superheater stable. Circulating fluidized bed boilers all use the method of adjusting the cooling water flow rate to control the inlet and outlet temperatures of the platen superheater within the target range. According to the current development status of the platen superheater outlet temperature control technology, in the main steam outlet temperature control project of circulating fluidized bed boilers, the PID controller is integrated into the DCS (Distributed Control System), and the PID is used in the DCS to adjust the cooling water flow rate of each stage to control the outlet temperature of each stage of the platen superheater.

[0004] like Figure 1 As shown in the figure, the points involved in the first and second stage platen superheaters of the circulating fluidized bed boiler are: first stage desuperheating water flow, first stage platen superheater inlet temperature, first stage platen superheater outlet temperature, second stage desuperheating water flow, second stage platen superheater inlet temperature, main steam outlet temperature (second stage platen superheater outlet temperature). According to the design safety and production requirements, there is a safety range for the inlet and outlet temperatures of the first and second stage platen superheaters. The traditional desuperheating water control strategy is: the first stage desuperheating water flow is used to ensure the safety and stability of the inlet and outlet temperatures of the first stage platen superheater; the second stage desuperheating water flow is used to ensure the safety and stability of the inlet and outlet temperatures of the second stage platen superheater. The above control requires setting the target value of the first stage platen superheater outlet temperature and the target value of the main steam outlet temperature at the same time. When only the first stage desuperheating water is involved in the control, only the outlet temperature of the first stage platen superheater can be controlled, and the main steam outlet temperature cannot be feedback-regulated; when only the second stage desuperheating water is involved in the control, the inlet and outlet temperatures of the first stage platen superheater cannot be intervened, and a safety alarm may occur. Summary of the invention

[0005] The purpose of this application is to provide a circulating fluidized bed boiler main steam outlet temperature control distribution method, equipment, medium and product, which can realize the control of both the primary cooling water flow rate and the secondary cooling water flow rate according to the main steam outlet temperature.

[0006] To achieve the above objectives, this application provides the following solutions:

[0007] In a first aspect, the present application provides a method for controlling and allocating the main steam outlet temperature of a circulating fluidized bed boiler, comprising: determining a modeling feature that affects the main steam outlet temperature; the modeling feature at least includes: a primary cooling water flow rate and a secondary cooling water flow rate; setting a target value for the main steam outlet temperature at a preset time in the future; obtaining the value of the modeling feature at a current moment and a preset number of moments before the current moment to form a modeling feature observation sequence; generating a search space for the primary cooling water flow rate at the next moment according to the value of the primary cooling water flow rate at the current moment, and discretizing it; generating a search space for the secondary cooling water flow rate at the next moment according to the value of the secondary cooling water flow rate at the current moment The first-stage cooling water flow search space is discretized; all the first-stage and second-stage flow combination values ​​are obtained in the combination space formed by the discretized first-stage cooling water flow search space and the discretized second-stage cooling water flow search space, and each first-stage and second-stage flow combination value is used to form the value of the modeling feature at the next moment; the value of each modeling feature at the next moment is input into the variable TFT time series prediction model together with the modeling feature observation sequence, and the predicted values ​​of the main steam outlet temperature at multiple future preset times are output; according to the predicted value of the main steam outlet temperature at each future preset time and the target value of the main steam outlet temperature, the loss function is selected. The first and second level flow combination value with the smallest number is selected; if the value of the first level cooling water flow at the next moment in the selected first and second level flow combination value is greater than the upper limit of the flow threshold, and the second level cooling water valve is in a closed state, the second level cooling water valve is normally opened, and the flow after the second level cooling water valve is normally opened is used as the value of the second level cooling water flow at the current moment, and the step of "generating a search space for the second level cooling water flow at the next moment according to the value of the second level cooling water flow at the current moment, and discretizing it" is returned to obtain the optimal value of the first level cooling water flow at the next moment, and then adjusting the first level cooling water flow at the next moment according to the optimal value of the first level cooling water flow at the next moment. level cooling water flow; if the value of the level one cooling water flow at the next moment in the selected level one and level two flow combination values ​​is less than the lower limit of the flow threshold, and the level two cooling water valve is in the normally open state, the level two cooling water valve is normally closed, and the flow after the level two cooling water valve is normally closed is used as the value of the level two cooling water flow at the current moment, and return to the step "according to the value of the level two cooling water flow at the current moment, generate the level two cooling water flow search space at the next moment, and discretize it" to obtain the optimal level one cooling water flow value at the next moment, and then adjust the level one cooling water flow at the next moment according to the optimal level one cooling water flow value at the next moment.

[0008] In a second aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the circulating fluidized bed boiler main steam outlet temperature control distribution method described above.

[0009] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned circulating fluidized bed boiler main steam outlet temperature control distribution method.

[0010] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above-mentioned circulating fluidized bed boiler main steam outlet temperature control distribution method.

[0011] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0012] The present application provides a method, device, medium and product for controlling and distributing the main steam outlet temperature of a circulating fluidized bed boiler. The modeling characteristics of at least the primary cooling water flow rate and the secondary cooling water flow rate are used as the input of a variable TFT timing prediction model, and the main steam outlet temperature is used as the output of the variable TFT timing prediction model, so that both the primary cooling water and the secondary cooling water are strongly associated with the main steam outlet temperature. This breaks the traditional practice that the primary cooling water only controls the primary screen superheater outlet temperature and the secondary cooling water only controls the main steam outlet temperature. By setting the target value of the main steam outlet temperature at a preset future time, the setting of multiple target values ​​is changed to setting only the main steam outlet temperature target value. According to the predicted value of the main steam outlet temperature at the preset future time and the target value of the main steam outlet temperature, the values ​​of the primary and secondary cooling water flows at the next moment that minimize the loss function are selected. The primary cooling water is used as the main control device and the secondary cooling water is used as the auxiliary control device, and the primary and secondary cooling water flows are further adjusted, thereby increasing the feedback adjustment of the main steam outlet temperature when only the primary cooling water is controlled. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0014] Figure 1 This is a point diagram involved in the first and second stage screen superheater of a circulating fluidized bed boiler;

[0015] Figure 2This is an application environment diagram of a circulating fluidized bed boiler main steam outlet temperature control distribution method in one embodiment of the present application;

[0016] Figure 3 A schematic flow chart of a circulating fluidized bed boiler main steam outlet temperature control distribution method provided in one embodiment of the present application;

[0017] Figure 4 A schematic diagram of the principle of a circulating fluidized bed boiler main steam outlet temperature control distribution method provided in one embodiment of the present application;

[0018] Figure 5 A structural diagram of a variable TFT timing prediction model provided by another embodiment of the present application;

[0019] Figure 6 A diagram of a gated residual network structure provided in another embodiment of the present application;

[0020] Figure 7 A feature selection network structure diagram provided for another embodiment of the present application;

[0021] Figure 8 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0023] The present application aims to optimize the cooling water control and the first and second cooling water distribution strategies, so as to solve the problem that the first cooling water only controls the outlet temperature of the first screen superheater, which will affect the main steam temperature but cannot be feedback-regulated, as well as the problem of optimal distribution of the first and second cooling water involved.

[0024] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0025] The circulating fluidized bed boiler main steam outlet temperature control distribution method provided in the embodiment of the present application can be applied to Figure 2In the application environment shown, the terminal 102 communicates with the server 104 through a network. The data storage system can store data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the modeling feature observation sequence to the server 104. After the server 104 receives the modeling feature observation sequence, the server 104 determines the modeling features that affect the main steam outlet temperature; the modeling features include at least: a primary cooling water flow rate and a secondary cooling water flow rate; set a target value of the main steam outlet temperature at a preset time in the future; generate a primary cooling water flow search space at the next moment according to the value of the primary cooling water flow at the current moment, and discretize it; generate a secondary cooling water flow search space at the next moment according to the value of the secondary cooling water flow at the current moment, and discretize it; obtain all primary and secondary flow combination values ​​in the combination space formed by the discretized primary cooling water flow search space and the discretized secondary cooling water flow search space, and each primary and secondary flow combination value constitutes the value of the modeling feature at the next moment; input each modeling feature value at the next moment together with the modeling feature observation sequence into the variable TFT time series prediction model, and output multiple predicted values ​​of the main steam outlet temperature at the preset time in the future; select the primary and secondary flow combination value that minimizes the loss function according to the predicted value of the main steam outlet temperature at each preset time in the future and the target value of the main steam outlet temperature ; If the value of the first-level cooling water flow at the next moment in the selected first-level and second-level flow combination values ​​is greater than the upper limit of the flow threshold, and the second-level cooling water valve is in a closed state, the second-level cooling water valve is normally opened, and the flow after the second-level cooling water valve is normally opened is used as the value of the second-level cooling water flow at the current moment, and return to the step "according to the value of the second-level cooling water flow at the current moment, generate the search space for the second-level cooling water flow at the next moment, and discretize it" to obtain the optimal value of the first-level cooling water flow at the next moment, and then adjust the first-level cooling water flow at the next moment according to the optimal value of the first-level cooling water flow at the next moment. If the value of the first-level cooling water flow rate at the next moment in the selected first-level and second-level flow combination values ​​is less than the lower limit of the flow threshold, and the second-level cooling water valve is in the normally open state, the second-level cooling water valve is normally closed, and the flow rate after the second-level cooling water valve is normally closed is used as the value of the second-level cooling water flow rate at the current moment, and the step of "generating the search space of the second-level cooling water flow rate at the next moment according to the value of the second-level cooling water flow rate at the current moment, and discretizing it" is returned to obtain the optimal value of the first-level cooling water flow rate at the next moment, and then adjusting the first-level cooling water flow rate at the next moment according to the optimal value of the first-level cooling water flow rate at the next moment. The server 104 can feed back the first-level cooling water flow rate and / or the switch state control data of the second-level cooling water valve at the next moment to the terminal 102.In addition, in some embodiments, the circulating fluidized bed boiler main steam outlet temperature control distribution method can also be implemented solely by the server 104, and the server 104 obtains the modeling feature observation sequence from the data storage system, and performs the first-level cooling water flow and / or the second-level cooling water valve switch state processing on the modeling feature observation sequence.

[0026] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart vehicle-mounted devices, etc. The portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.

[0027] In an exemplary embodiment, Figure 3 As shown, a circulating fluidized bed boiler main steam outlet temperature control distribution method is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 2 The server 104 in the example is used as an example to illustrate, including the following steps 1 to 10. Among them:

[0028] Step 1: Determine the modeling features that affect the main steam outlet temperature; the modeling features include: primary desuperheating water flow rate and secondary desuperheating water flow rate.

[0029] Step 2: Set the target value of the main steam outlet temperature at a preset time in the future;

[0030] Step 3: Obtain the values ​​of the modeling features at the current moment and a preset number of moments before the current moment to form a modeling feature observation sequence.

[0031] Step 4: According to the value of the primary and secondary cooling water flow at the current moment, generate the search space of the primary and secondary cooling water flow at the next moment and discretize it.

[0032] Step 5: According to the value of the secondary cooling water flow at the current moment, generate the search space of the secondary cooling water flow at the next moment and discretize it.

[0033] Step 6: Obtain all the first- and second-level flow combination values ​​in the combination space formed by the discretized first-level cooling water flow search space and the discretized second-level cooling water flow search space, and use each first- and second-level flow combination value to form the value of the modeling feature at the next moment.

[0034] Step 7: Input the value of each modeling feature at the next moment together with the modeling feature observation sequence into the variable TFT time series prediction model, and output the predicted values ​​of the main steam outlet temperature at multiple future preset times.

[0035] Step 8: According to the predicted value of the main steam outlet temperature at each future preset time and the target value of the main steam outlet temperature, select the first and second level flow combination values ​​that minimize the loss function.

[0036] Step 9: If the value of the first-level cooling water flow rate in the selected first-level and second-level flow combination values ​​at the next moment is greater than the upper limit of the flow threshold, and the second-level cooling water valve is in a closed state, the second-level cooling water valve is normally opened, and the flow rate after the second-level cooling water valve is normally opened is used as the value of the second-level cooling water flow rate at the current moment, and return to the step "Based on the value of the second-level cooling water flow rate at the current moment, generate the search space for the second-level cooling water flow rate at the next moment, and discretize it" to obtain the optimal value of the first-level cooling water flow rate at the next moment, and then adjust the first-level cooling water flow rate at the next moment according to the optimal value of the first-level cooling water flow rate at the next moment.

[0037] Step 10: If the value of the first-level cooling water flow rate in the selected first-level and second-level flow combination values ​​at the next moment is less than the lower limit of the flow threshold, and the second-level cooling water valve is in the normally open state, the second-level cooling water valve is normally closed, and the flow rate after the second-level cooling water valve is normally closed is used as the value of the second-level cooling water flow rate at the current moment, and return to the step "according to the value of the second-level cooling water flow rate at the current moment, generate the search space for the second-level cooling water flow rate at the next moment, and discretize it" to obtain the optimal value of the first-level cooling water flow rate at the next moment, and then adjust the first-level cooling water flow rate at the next moment according to the optimal value of the first-level cooling water flow rate at the next moment.

[0038] Implement the above-mentioned steps 1 to 10. The control strategy of cooling water in this application adopts primary cooling water and secondary cooling water to control the main steam outlet temperature, which effectively avoids the situation where the main steam outlet temperature cannot participate in feedback regulation when only the first level is controlled. At the same time, try to avoid the situation where only the second level is controlled. Even if controlled in this way, the flow rate of the primary cooling water must be adjusted according to the actual operating conditions.

[0039] In another exemplary embodiment of the present application, in order to specifically determine the modeling features that affect the main steam outlet temperature, the above step 1 is replaced by the following steps 1.1 to 1.3:

[0040] Step 1.1: Preliminary determination of modeling features that affect the main steam outlet temperature.

[0041] Step 1.2: Perform a single-dimensional analysis on the historical data of the preliminarily determined modeling features, eliminate the modeling features that are the same in 90% of the historical data, and obtain the modeling features for one screening.

[0042] Step 1.3: Perform a multi-dimensional correlation analysis on the modeling features screened once, retain the modeling features whose correlation is greater than the correlation threshold and have advance properties, and finally obtain the modeling features that affect the main steam outlet temperature; the modeling features finally obtained include primary air volume, secondary air volume, primary cooling water flow rate, secondary cooling water flow rate, load, cooling water temperature and cooling water pressure; the advance property refers to the property of the main steam outlet temperature changing again after the modeling features change for a period of time.

[0043] Furthermore, in order to accurately use the model to predict the main steam outlet temperature, it is necessary to clarify the changing trend of the influence of the modeling features on the main steam outlet temperature, analyze the time lag and reaction time of the modeling features relative to the main steam outlet temperature, that is, when one or several features change and other features remain unchanged, when does the influence on the main steam outlet temperature begin and end, and then determine the number of historical sampling moments (preset number of moments) and future prediction moments (number of moments in the future preset time) of the past observed features when using the model for prediction. Therefore, it is also necessary to perform a time lag analysis on the modeling features finally obtained to determine the changing trend of the influence of each modeling feature on the main steam outlet temperature; the influence change trend includes time lag and reaction time; based on the influence change trend, determine the number of moments in the preset number of moments and the number of moments in the future preset time.

[0044] The more detailed implementation process of step 1 is as follows: Based on expert experience, preliminarily determine the modeling features, such as: primary air volume, secondary air volume, load, cooling water flow, etc., which may affect the main steam outlet temperature; analyze historical data: 1) Single-dimensional analysis, among the features that can be collected, check the distribution of features, and eliminate the features whose values ​​remain unchanged for 90% of the time; 2) Multi-dimensional correlation analysis, check the correlation between features and the main steam outlet temperature, and retain features with high correlation and advance (features that change after a period of time before the main steam outlet temperature changes. For some features that change after a period of time before the main steam outlet temperature changes, these features may not be suitable although they have high correlation); 3) Time lag analysis, through the trend chart combined with the optimization algorithm to check the influence of each feature on the main steam outlet temperature change trend (time lag and reaction time), that is, when one or several features change and other features remain unchanged, when the influence on the main steam outlet temperature begins and ends. Based on the above expert experience and data analysis operations, determine the features that affect the main steam outlet temperature, as well as the time lag and reaction time of each feature on the main steam outlet temperature.

[0045] In another exemplary embodiment of the present application, in order to clarify in more detail the process of forming the value of the modeling feature at the next moment, steps 4 to 6 can be understood as follows:

[0046] According to the actual control requirements of the first and second level cooling water flow, when both the first and second level valves are automatically controlled, it is necessary to find the appropriate first and second level cooling water flow at the next moment. Assuming that the features 1, 2, etc. at the next moment remain unchanged, the first and second level cooling water flow are traversed within a certain range (such as the current value of the first level cooling water flow is f1, and the current value of the second level cooling water is f2. The search space for the first and second level cooling water flow at the next moment is generated [f1-m, f1+m], and discretized with 0.1 or 0.3 as the dividing point and [f2-q, f2+q] with 0.1 or 0.3 as the dividing point. The lengths are 2m and 2q respectively, and a two-dimensional matrix of 2m×2q is generated. In this two-dimensional space, an attempt is made to find the combination of the first and second level cooling water that minimizes the loss function, which is used as the basis for adjusting the first and second level cooling water flow at the next moment).

[0047] Obtain all first- and second-level flow combination values ​​in the combination space formed by the discretized first-level cooling water flow search space and the discretized second-level cooling water flow search space; set the values ​​of modeling features other than the first-level cooling water flow and the second-level cooling water flow at the next moment to be equal to their respective values ​​at the current moment, and form the values ​​of the modeling features at the next moment with each first- and second-level flow combination value.

[0048] In another exemplary embodiment of the present application, the specific implementation process of step 7 is as follows: Based on the above analysis results, a model for predicting the main steam outlet temperature in the future is constructed for the primary and secondary cooling water flow rates respectively. The model adopts a variable TFT timing prediction model, and the analysis modeling input is the primary air volume, secondary air volume, primary cooling water flow rate, secondary cooling water flow rate, load, cooling water temperature, cooling water pressure and other related characteristics, and the output is the predicted value of the main steam outlet temperature in the future ([VT+1, VT+2, ... VT+k], where T is the current moment, and the main steam outlet temperature value is predicted for the next k moments, that is, the latter has a predicted value for each moment in the future as stated in the notes). The structure of the variable TFT timing prediction model is as follows Figure 5 shown.

[0049] Figure 5 The GRN in is a gated residual network (such as Figure 6 As shown in the figure), VSN is a feature selection network (as shown in the figure), Figure 7As shown in the figure), LstmEncoder is the Lstm encoder, Lstm Decoder is the Lstm decoder, add&Norm is multi-value superposition and summation followed by normalization, Multi-headAttention is the multi-head attention mechanism, and Dense is the fully connected layer. The above model description: To predict the main steam outlet temperature value at k moments in the future, the sequence values ​​of each observation feature at t moments before the current moment (such as the t-time value of the primary air volume, etc.) are known as past observation features, and the feature values ​​known at k moments in the future (such as the time value at k moments, etc.) are called future known features. 1) Feature selection (VSN) is performed on the observation features of each moment in the past, and then they are encoded using LSTM, and then they are superimposed and normalized with the features before encoding, and then enter the gated residual network. 2) Feature selection (VSN) is performed on the future known features, and then they are decoded using LSTM, and then they are superimposed and normalized with the features before encoding, and then enter the gated residual network. 3) Input the outputs of 1) and 2) into the multi-head attention mechanism, superimpose and normalize the multiple arrays generated at each future moment, then connect them to a gated residual network, connect the outputs of these k moments to a fully connected network, and finally predict the main steam outlet temperature values ​​at the next k moments.

[0050] Figure 6 This is the structure diagram of the GRN gated residual network. The input vector a passes through a fully connected network and uses ELU to make nonlinear changes. Then it is connected to a layer of fully connected network and passes through a layer of Gate structure (i.e. value*sigmoid(value)) and then is superimposed and normalized with the original input vector a.

[0051] Figure 7 This is the VSN feature selection network structure diagram. The purpose is to screen out the more important features at each moment, pass them through the gated residual network, and then multiply them with the values ​​of the softmax mapping after the full amount of features pass through the gated residual network. After superposition, they are used as the output of the feature selection structure VSN.

[0052] The goal of the model is to minimize the loss function, which is as follows:

[0053]

[0054] Among them, L is the loss value, y i is the target value of the main steam outlet temperature at the i-th moment in the future, is the predicted value of the main steam outlet temperature at the i-th moment in the future, i is the i-th moment in the future, and k is the total number of k moments that need to be predicted in the future (the number of moments within the preset time in the future), that is, the square error between the predicted value and the actual value at the first moment in the future plus the square error between the predicted value and the actual value at the second moment in the future until the square error between the predicted value and the actual value at the k-th moment in the future. Model training attempts to find the set of parameters that minimize this loss function (i.e., model training).

[0055] The input of the variable TFT timing prediction model during application: feature 1, feature 2, ...., primary cooling water flow, secondary cooling water flow, and the output: main steam outlet flow at the next k moments.

[0056] In another exemplary embodiment of the present application, Figure 4 As shown, the above steps 8 to 10 can be specifically implemented by the following process: Based on the above main steam temperature prediction model, the first-level cooling water change (searched within the interval range) can be set when other features remain unchanged at the next moment, and the first-level cooling water change corresponding to the predicted main steam temperature value closest to the target value is found, that is, the cooling water flow change required to adjust only the first-level cooling water at the next moment. Similarly, the cooling water flow change required to adjust only the second-level cooling water at the next moment can be found. Based on the statistical analysis of historical data combined with expert experience, it can be known that the time to adjust the first-level cooling water is longer than the time to adjust the second-level cooling water, that is, most of the time, the first-level cooling water can be adjusted. Therefore, it is determined that the first-level cooling water is the main control device and the second-level cooling water is the auxiliary control device. The flow of the first-level cooling water is often greater than the flow of the second-level cooling water. When the flow of the first-level cooling water exceeds the set threshold, the second-level cooling water needs to change the control state, and then combined with historical information to determine whether the second-level cooling water valve is normally open or normally closed. That is, when the primary cooling water flow is greater than the set threshold value, the secondary cooling water valve is in a closed state, and the secondary valve is normally open (known part: directly open the valve to the minimum water unit, such as: when the valve is closed, the cooling water flow is 0t; the valve opening is opened to 1%, the cooling water flow is still 0t; until the valve opening is opened to 2% (or 8% according to the actual situation), the cooling water flow immediately changes from 0->4t; the valve opening is opened to 3%, and the cooling water flow is 5t. According to the known, the secondary cooling water flow is locked at 4t (the secondary valve opening is locked at 2%) to search for the primary cooling water flow that minimizes the above loss function); when the primary cooling water flow is less than the set threshold value, the secondary cooling water valve is in a normally open state, and the secondary valve is normally closed (the explanation is the same as above); in other cases, the valve switch state remains unchanged.

[0057] The following is a specific example to illustrate the implementation process of the method of the present application.

[0058] Assume that the upper limit of the flow threshold of the first-stage cooling water flow is 20 (if it is greater than 20, the second stage is normally open), and the lower limit of the flow threshold is 15 (if it is less than 15, the second stage is normally closed).

[0059] If the primary cooling water flow rate is 18 t / h, the secondary cooling water flow rate is 0 t / h, the search interval of the primary cooling water is [15,21], the search interval of the secondary cooling water is [0,0], and the optimal value of the primary cooling water flow rate is 21, then the secondary cooling water flow rate needs to be set to 4, and then the primary cooling water is searched in the search interval [15,21] and the secondary cooling water search interval is [4,4]. The retrieved combination may be [15,4].

[0060] If the primary cooling water flow is 15t / h, the secondary cooling water flow is 4t / h, the primary cooling water search interval is [12,18], the secondary cooling water search interval is [4,6], and the search result is [13,5], then the secondary cooling water flow is set to 0, and then the primary cooling water search interval is [12,18], and the secondary cooling water search interval is [0,0]. The subsequent secondary valve is always open, which means that the secondary cooling water only needs to have a flow (the default minimum is 4t / h). If there is a flow, it can be adjusted and searched, as long as the primary cooling water flow is greater than the threshold, the secondary cooling water has a flow (greater than 4t / h).

[0061] When controlling the main steam outlet temperature of the circulating fluidized bed boiler, the traditional practice of using the first-stage cooling water to only control the first-stage screen superheater outlet temperature and the second-stage cooling water to only control the main steam outlet temperature is broken. Both the first-stage cooling water and the second-stage cooling water are used to build a strong correlation model with the main steam outlet temperature, and the feedback adjustment of the main steam outlet temperature is increased when only the first-stage cooling water is controlled (the core of the main steam outlet temperature project is to control the main steam outlet temperature. The previous practice was that the second-stage cooling water was used to control the main steam outlet temperature, and the first-stage cooling water was used to control the first-stage screen superheater outlet temperature. The main steam outlet temperature was not directly controlled. The corresponding relationship between the first-stage screen superheater outlet temperature and the main steam outlet temperature under different working conditions was inconsistent. Therefore, the previous solution required manual real-time adjustment of the target value of the first-stage screen superheater outlet temperature. Now the first-stage cooling water directly controls the main steam outlet temperature, that is, the feedback of the main steam outlet temperature is increased when only the first-stage cooling water is controlled).

[0062] According to this new control allocation strategy for primary and secondary cooling water, human operations are minimized, such as: changing the setting of multiple target values ​​to only setting the target value of the main steam outlet temperature, changing the manual setting of whether to control the primary cooling water or the secondary cooling water to controlling both the primary and secondary cooling water, and whether to control depends on the allocation strategy. In other words, the existing technical solution is to manually set which one to control, while the technical solution of this application is to control both cooling waters, and decide whether to control and how to control through the allocation strategy. It can also avoid the problem that the main steam outlet temperature cannot participate in feedback control and the inlet and outlet temperatures of the primary screen superheater exceed the limit.

[0063] Based on the same inventive concept, the embodiment of the present application also provides a distribution device for implementing the above-mentioned circulating fluidized bed boiler main steam outlet temperature control distribution method. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above-mentioned method, so the specific limitations in the distribution device embodiment provided below can refer to the limitations of the circulating fluidized bed boiler main steam outlet temperature control distribution method above, and will not be repeated here.

[0064] In an exemplary embodiment, a distribution device is provided, including: a server, a database, and a DCS. The server executes the above-mentioned circulating fluidized bed boiler main steam outlet temperature control distribution method, obtains the primary cooling water flow rate and / or the secondary cooling water valve switch state control data at the next moment, and stores them in the database. After the DCS obtains the primary cooling water flow rate and / or the secondary cooling water valve switch state control data at the next moment from the database, it controls the primary cooling water valve and the secondary cooling water valve according to the obtained data.

[0065] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 8As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store a circulating fluidized bed boiler main steam outlet temperature control distribution method. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a circulating fluidized bed boiler main steam outlet temperature control distribution method is implemented.

[0066] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0067] In an exemplary embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0068] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0069] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0070] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0071] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0072] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0073] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0074] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A circulating fluidized bed boiler main steam outlet temperature control distribution method, characterized in that: include: Identify modeling features that affect main steam outlet temperature; The modeling features include at least: primary desuperheating water flow rate and secondary desuperheating water flow rate; Set the target value of the main steam outlet temperature at a preset time in the future; Obtain the values ​​of the modeling feature at the current moment and a preset number of moments before the current moment to form a modeling feature observation sequence; According to the value of the primary cooling water flow at the current moment, the search space of the primary cooling water flow at the next moment is generated and discretized; According to the value of the secondary cooling water flow at the current moment, the search space of the secondary cooling water flow at the next moment is generated and discretized; Obtain all the first- and second-level flow combination values ​​in the combination space formed by the discretized first-level cooling water flow search space and the discretized second-level cooling water flow search space, and use each first- and second-level flow combination value to form the value of the modeling feature at the next moment; The value of each modeling feature at the next moment is input into the variable TFT time series prediction model together with the modeling feature observation sequence, and the predicted values ​​of the main steam outlet temperature at various future preset times are output; According to the predicted value of the main steam outlet temperature at each future preset time and the target value of the main steam outlet temperature, the first and second level flow combination values ​​that minimize the loss function are selected; If the value of the first-level cooling water flow rate at the next moment in the selected first-level and second-level flow combination values ​​is greater than the upper limit of the flow threshold, and the second-level cooling water valve is in a closed state, the second-level cooling water valve is normally opened, and the flow rate after the second-level cooling water valve is normally opened is used as the value of the second-level cooling water flow rate at the current moment, and the step of "generating a search space for the second-level cooling water flow rate at the next moment according to the value of the second-level cooling water flow rate at the current moment, and discretizing it" is returned to obtain the optimal value of the first-level cooling water flow rate at the next moment, and then adjusting the first-level cooling water flow rate at the next moment according to the optimal value of the first-level cooling water flow rate at the next moment; If the value of the first-level cooling water flow rate at the next moment in the selected first-level and second-level flow combination values ​​is less than the lower limit of the flow threshold, and the second-level cooling water valve is in a normally open state, the second-level cooling water valve is normally closed, and the flow rate after the second-level cooling water valve is normally closed is used as the value of the second-level cooling water flow rate at the current moment, and the process returns to step "generating a search space for the second-level cooling water flow rate at the next moment according to the value of the second-level cooling water flow rate at the current moment, and discretizing it", to obtain the optimal value of the first-level cooling water flow rate at the next moment, and then adjusting the first-level cooling water flow rate at the next moment according to the optimal value of the first-level cooling water flow rate at the next moment; Identify modeling features that affect the main steam outlet temperature, including: Preliminary determination of modeling features that affect main steam outlet temperature; Perform a single-dimensional analysis on the historical data of the initially determined modeling features, remove the modeling features that are the same in 90% of the historical data, and obtain the modeling features for one-time screening; A multi-dimensional correlation analysis is performed on the modeling features screened once, and modeling features with correlation greater than the correlation threshold and with advance properties are retained, and finally the modeling features that affect the main steam outlet temperature are obtained; the modeling features finally obtained include the primary cooling water flow rate, the secondary cooling water flow rate, the primary air volume, the secondary air volume, the load, the cooling water temperature and the cooling water pressure; the advance property refers to the property of the main steam outlet temperature changing again after the modeling features change for a period of time; A multi-dimensional correlation analysis is performed on the modeling features screened once, and modeling features with correlation greater than a correlation threshold and with advance properties are retained, and finally modeling features affecting the main steam outlet temperature are obtained, and then the following is further included: Performing a time lag analysis on the modeling features finally obtained to determine the influence change trend of each modeling feature on the main steam outlet temperature; the influence change trend includes time lag and reaction time; According to the impact change trend, a preset number of moments and a number of moments within a preset time in the future are determined.

2. The circulating fluidized bed boiler main steam outlet temperature control distribution method according to claim 1, characterized in that: All the first- and second-level flow combination values ​​are obtained in the combination space formed by the discretized first-level cooling water flow search space and the discretized second-level cooling water flow search space, and each first- and second-level flow combination value is used to form the value of the modeling feature at the next moment, specifically including: Obtain all the first and second level flow combination values ​​in the combination space formed by the discretized first level desuperheating water flow search space and the discretized second level desuperheating water flow search space; The values ​​of the modeling features other than the primary cooling water flow and the secondary cooling water flow at the next moment are set equal to their respective values ​​at the current moment, and together with each primary and secondary flow combination value, constitute the values ​​of the modeling features at the next moment.

3. The circulating fluidized bed boiler main steam outlet temperature control distribution method according to claim 1, characterized in that: The loss function is: Where L is the loss value, y i is the target value of the main steam outlet temperature at the i-th moment in the future, is the predicted value of the main steam outlet temperature at the i-th moment in the future, and k is the number of moments in the future preset time.

4. The circulating fluidized bed boiler main steam outlet temperature control distribution method according to claim 1, characterized in that: When the secondary cooling water valve is normally opened, the secondary valve opening is locked at 2%, and the secondary cooling water flow rate is 4t / h; When the secondary cooling water valve is normally closed, the secondary cooling water flow rate is 0.

5. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the circulating fluidized bed boiler main steam outlet temperature control distribution method according to any one of claims 1 to 4.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the circulating fluidized bed boiler main steam outlet temperature control distribution method according to any one of claims 1 to 4 is implemented.

7. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the circulating fluidized bed boiler main steam outlet temperature control distribution method according to any one of claims 1 to 4 is implemented.

Images