Thermal power plant flue gas heat exchanger control method based on dynamic matrix control

Abstract

The invention discloses a thermal power plant flue gas heat exchanger control method based on dynamic matrix control, and relates to the technical field of heat exchanger control. LSTM is adopted to construct a flue gas temperature prediction model, the influence of boiler load fluctuation and fire coal characteristic change on the flue gas temperature is captured, and the future temperature change trend is predicted in advance; a temperature change adjusting instruction is generated, so that the heat exchanger can actively adapt to flue gas temperature fluctuation, and adjusting lag is reduced; a heat exchanger state space model is established based on an LSTM prediction result, historical data and future prediction input are combined, an optimal heat exchange medium flow adjusting instruction is calculated, overshoot or lag of PID control under the condition of sudden load change is avoided, and the temperature adjusting precision is improved; and a rolling optimization strategy is adopted, so that the heat exchanger control is integrated with the current state, the heat exchange medium flow is optimized in a plurality of time steps, and the perspectiveness and stability of control are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of heat exchanger control, and particularly to a control method for a flue gas heat exchanger in a thermal power plant based on dynamic matrix control. Background Art

[0002] In a thermal power plant, the high-temperature flue gas waste heat recovery system undertakes the important tasks of improving energy utilization efficiency and reducing fuel consumption; the function of the heat exchanger is to utilize the high-temperature flue gas discharged from the boiler to efficiently transfer heat to the working medium, thereby reducing energy loss.

[0003] However, this process is far affected by multiple variables such as flue gas temperature, flow rate, coal combustion characteristics, and boiler operating load. In the actual operation of a thermal power plant, the rapid fluctuation of the boiler load will cause an instantaneous change in the flue gas temperature, and the fluctuation is more obvious during the start-stop process or when the coal quality changes; when the boiler load increases, the combustion intensity increases, and the discharged flue gas temperature rises rapidly. The heat exchanger needs to adapt to the new heat load within a short time, otherwise the outlet temperature may rise significantly, affecting the stability of the subsequent system; on the contrary, when the load decreases, the flue gas temperature drops. If the heat exchanger cannot quickly adjust its heat transfer capacity, the outlet temperature may decrease briefly, affecting the waste heat utilization efficiency and even the normal operation of downstream equipment such as the denitration system.

[0004] In the existing control methods, PID control is the most common adjustment means. However, the flue gas temperature does not change smoothly, but is affected by boiler load adjustment, coal type change, and even combustion organization mode, showing a large amplitude of dynamic fluctuation. Traditional PID control relies on fixed parameters, and when facing rapidly changing working conditions, the adjustment response is often lagged. In addition to PID control, bypass flue gas regulation is also a common supplementary means, but this method relies on the adjustment of mechanical baffles and has a certain inertia, and the response is slow when dealing with sudden changes in boiler working conditions; there is also a relatively common adjustment method, which is the variable frequency control of the induced draft fan. By adjusting the flue gas flow rate, the heat transfer intensity of the heat exchanger is indirectly affected, and the residence time of the flue gas in the heat exchanger is reduced. However, the response inertia of the induced draft fan is large, and it is difficult to meet the requirements of rapidly changing boiler working conditions. When the coal quality changes or the boiler load is adjusted rapidly, the adjustment of the induced draft fan mostly lags behind the change of the flue gas temperature;

[0005] In response to the above problems, some traditional solutions introduce adaptive PID to automatically adjust the control parameters according to the change of working conditions, improving the adaptability of the adjustment. However, adaptive PID still has certain limitations when dealing with non-linear working conditions and is difficult to completely eliminate overshoot and lag problems. Intelligent bypass regulation is limited by the mechanical adjustment speed and is difficult to adapt to violent fluctuations within a short time.

[0006] In summary, in the operating environment of a thermal power plant, the rapid fluctuations of boiler load, the changes in coal combustion characteristics, and the non-linear dynamic changes in flue gas temperature lead to a lag in the response of traditional control strategies when dealing with rapidly changing operating conditions, insufficient adjustment accuracy, large fluctuations in the outlet temperature of the heat exchanger, affecting the waste heat recovery efficiency, and causing a chain reaction to subsequent denitrification, emission control and other systems, affecting the overall stability and economy of the thermal power plant. Therefore, there is an urgent need for a more intelligent and faster-responsive control strategy for the flue gas heat exchanger in thermal power plants based on dynamic matrix control to improve the adaptive adjustment ability of the heat exchanger. Summary of the Invention

[0007] In view of the above existing problems, the present invention is proposed.

[0008] The present invention provides a control method for a flue gas heat exchanger in a thermal power plant based on dynamic matrix control to solve the problems of lag in traditional PID control, insufficient adjustment accuracy of bypass flue gas, slow response of induced draft fans, and difficulty in coping with rapid fluctuations in flue gas temperature, which affect the stability of the outlet temperature of the heat exchanger.

[0009] To solve the above technical problems, the present invention provides the following technical solutions:

[0010] An embodiment of the present invention provides a control method for a flue gas heat exchanger in a thermal power plant based on dynamic matrix control, which includes:

[0011] Step S1, temperature sensors, flow meters and pressure sensors are respectively arranged at the inlet and outlet of the flue gas heat exchanger to collect flue gas sensing data and boiler operation data in real time, and generate a current operating condition data set;

[0012] Step S2, based on the current operating condition data set in Step S1, use the long short-term memory network LSTM to construct a flue gas temperature prediction model to predict the change trend of flue gas temperature in the future time period, and generate a temperature change adjustment instruction;

[0013] Step S3, receive the temperature change adjustment instruction generated in Step S2, and use dynamic matrix control DMC to perform predictive control on the outlet temperature of the heat exchanger to generate a heat transfer medium flow rate adjustment instruction;

[0014] Step S4, adopt the heat transfer medium flow rate adjustment instruction in Step S3 to control the multi-channel heat exchange tube bundle structure inside the heat exchanger, and dynamically distribute the flue gas flow rate entering different heat exchange areas;

[0015] Step S5, based on the temperature adjustment feedback data at the outlet of the heat exchanger, adaptively optimize the DMC control parameters in Step S3 based on the reinforcement learning algorithm.

[0016] As a preferred embodiment of the flue gas heat exchanger control method for thermal power plants based on dynamic matrix control according to the present invention, wherein: the flue gas sensing data includes flue gas temperature, flue gas flow rate, flue gas pressure, and heat exchange medium temperature;

[0017] The boiler operation data includes boiler load changes, coal combustion characteristics, and air flow rate.

[0018] As a preferred embodiment of the flue gas heat exchanger control method for thermal power plants based on dynamic matrix control according to the present invention, wherein: the steps of constructing a flue gas temperature prediction model using a long short-term memory network (LSTM) are as follows:

[0019] Construct a flue gas temperature prediction model and collect the current working condition data set D t , for LSTM network training, and represent the data set as D t :

[0020] D t ={T g , Q g , P g , T m , L, C, Q a},

[0021] wherein, T g represents the flue gas temperature, Q g represents the flue gas flow rate, P g represents the flue gas pressure, T m represents the heat exchange medium temperature, L represents the boiler load, C represents the coal combustion characteristics, and Q a represents the air flow rate;

[0022] The state update process of the LSTM cell includes:

[0023] i t =σ(W i [h t-1 , x t +b i ),

[0024] f t =σ(W f [h t-1 , x t +b f ),

[0025] o t =σ(W o [h t-1 , x t +b o ),

[0026]

[0027] h t = o t ⊙tanh(C t )

[0028] wherein, i t represents the input gate activation value, f t represents the forget gate activation value, o t represents the output gate activation value, represents the candidate state value, C t represents the cell state value, C t1 represents the cell state value at the previous time step, h t represents the hidden state value, h t1 represents the hidden state value at the previous time step, W i , W f , W o , W c are weight matrices, b i , b f , b o , b c are bias terms, x t represents the input data, σ represents the Sigmoid activation function, tanh represents the hyperbolic tangent activation function, and ⊙ represents the element-wise multiplication operation.

[0029] As a preferred solution of the flue gas heat exchanger control method based on dynamic matrix control described in the present invention, wherein: the step of predicting the change trend of the flue gas temperature in the future time period and generating a temperature change adjustment instruction is as follows,

[0030] Based on the hidden state h t calculated by the LSTM, predict the flue gas temperature at the future time step, and the prediction formula is:

[0031]

[0032] wherein, represents the predicted flue gas temperature at time t+τ, f(·) is the LSTM prediction model, and θ is the set of parameters obtained by LSTM training,

[0033] Generate a temperature change adjustment instruction, and the generation process includes:

[0034] The temperature change trend, and the calculation formula is:

[0035]

[0036] wherein, ΔT g is the flue gas temperature change amount, is the predicted flue gas temperature, T g,tis the flue gas temperature at the current time step t,

[0037] If |ΔT g | > δ T , then generate a temperature change adjustment instruction:

[0038] U T = g(ΔT g ),

[0039] where δ T is the temperature change threshold, U T represents the temperature change adjustment instruction, and g(·) is the adjustment function.

[0040] As a preferred solution of the flue gas heat exchanger control method based on dynamic matrix control according to the present invention, wherein: in step S3, establish a state space model of the heat exchanger, comprehensively calculate the heat transfer medium flow adjustment scheme based on historical temperature change data, current measurement data and future predicted temperature trends, and generate a heat transfer medium flow adjustment instruction.

[0041] As a preferred solution of the flue gas heat exchanger control method based on dynamic matrix control according to the present invention, wherein: the step of using dynamic matrix control DMC to perform predictive control on the outlet temperature of the heat exchanger and generate a heat transfer medium flow adjustment instruction is as follows:

[0042] Establish a state space model of the heat exchanger, and the heat exchanger state space equation is defined as:

[0043] X t+1 = AX t + BU t ,

[0044] Y t = CX t + DU t ,

[0045] where X t+1 represents the state variable vector of the next time step, X t represents the state variable vector of the current time step, U t represents the input variable vector of the current time step, Y t represents the output variable vector of the current time step, A represents the heat exchanger state transition matrix, describing the evolution relationship of the heat exchanger state over time, B represents the input influence matrix, describing the influence of the heat transfer medium flow on the heat exchanger state, C represents the measurement matrix, mapping the state variable to the output variable, D represents the input direct influence matrix, describing the direct influence of the input on the output,

[0046] Predict the temperature change based on historical data and future input, and the prediction formula is:

[0047]

[0048] Among them, is the temperature of the heat exchange medium predicted for the future time step t+τ, X t+τ is the state variable at time step t+τ, U t+τ is the input variable at time step t+τ,

[0049] DMC adopts a rolling optimization strategy to calculate the optimal heat exchange medium flow rate adjustment command. The calculation formula is:

[0050]

[0051] Among them, is the optimal heat exchange medium flow rate adjustment command, is the desired heat exchange medium temperature, N is the prediction step length, w τ is the weight coefficient at time step τ, λ is the regularization coefficient, ΔU m is the change in the heat exchange medium flow rate control input, U m represents the control input of the heat exchange medium flow rate.

[0052] As a preferred scheme of the method for controlling a flue gas heat exchanger in a thermal power plant based on dynamic matrix control according to the present invention, wherein:

[0053] In step S4, the heat exchange area is divided into a high-temperature area, a medium-temperature area, and a low-temperature area;

[0054] In step S4, according to the heat exchange medium flow rate adjustment command, an adjustable baffle or a rotary valve is used to adjust the flue gas flow rate ratio entering each zone, and heat exchanger outlet temperature adjustment feedback data is generated.

[0055] As a preferred scheme of the method for controlling a flue gas heat exchanger in a thermal power plant based on dynamic matrix control according to the present invention, wherein: the step of using the heat exchange medium flow rate adjustment command in step S3 to control the multi-channel heat exchange tube bundle structure inside the heat exchanger and dynamically distribute the flue gas flow rate entering different heat exchange areas is,

[0056] The internal area of the heat exchanger is divided into:

[0057] A high-temperature area for initially reducing the flue gas temperature,

[0058] A medium-temperature area for further heat exchange,

[0059] And a low-temperature area for final flue gas exhaust;

[0060] The flue gas flow rate ratio is adjusted by using an adjustable baffle or a rotary valve. The adjustment process is expressed as:

[0061] Q g,i= α i Q g ,

[0062] wherein, Q g,i represents the flue gas flow rate entering the i-th heat exchange area, and Q g represents the total flue gas flow rate, and α i is the distribution coefficient, satisfying ∑ i α i = 1.

[0063] As a preferred solution of the method for controlling a flue gas heat exchanger in a thermal power plant based on dynamic matrix control according to the present invention, wherein:

[0064] In step S5, compare the deviation between the predicted temperature and the actual outlet temperature to determine the error range of the DMC control strategy;

[0065] In step S5, update the DMC control weight based on the error range and adjust the partition heat exchange regulation parameter.

[0066] As a preferred solution of the method for controlling a flue gas heat exchanger in a thermal power plant based on dynamic matrix control according to the present invention, wherein: the step of adaptively optimizing the DMC control parameters in step S3 based on the temperature regulation feedback data at the outlet of the heat exchanger and based on the reinforcement learning algorithm is as follows:

[0067] Calculate the prediction error of the heat exchanger outlet temperature at each time step t, and the calculation formula is:

[0068]

[0069] wherein, e t represents the temperature prediction error, represents the temperature of the heat exchange medium predicted at time step t, and T m,t represents the temperature of the heat exchange medium actually measured at time step t,

[0070] Set the reinforcement learning target, construct the objective function J, and measure the optimization degree of the control strategy:

[0071]

[0072] wherein, J represents the control optimization objective function, T is the number of time steps in the optimization period, λ is the regularization coefficient, and Δw τ represents the change amount of the control parameter weight,

[0073] Update the DMC control weight using the gradient descent method, and the update formula is:

[0074]

[0075] wherein, Denotes the weights of the DMC control parameters after the update at time step t+1, Denotes the weights of the current DMC control parameters at time step t, and η is the learning rate, Denotes the gradient of the loss function J with respect to w τ of,

[0076] Based on the optimized DMC control weights, adjust the flue gas flow distribution in the heat exchange area. The adjustment formula is:

[0077]

[0078] where, Denotes the flue gas flow distribution coefficient of the i-th heat exchange area after the update at time step t+1, Denotes the flue gas flow distribution coefficient of the i-th heat exchange area at time step t, and γ is the adjustment step size.

[0079] The beneficial effects of the present invention are as follows: In the present invention, an LSTM is used to construct a flue gas temperature prediction model to capture the influence of boiler load fluctuations and changes in coal combustion characteristics on the flue gas temperature, predict the future temperature change trend in advance, and generate temperature change adjustment instructions, enabling the heat exchanger to actively adapt to flue gas temperature fluctuations and reducing adjustment lag; Based on the LSTM prediction results, a state space model of the heat exchanger is established, and combined with historical data and future prediction inputs, the optimal heat transfer medium flow adjustment instructions are calculated to avoid overshoot or lag in the case of load mutations in PID control and improve the temperature adjustment accuracy.

[0080] In the present invention, a rolling optimization strategy is adopted to enable the heat exchanger control to comprehensively consider the current state and optimize the heat transfer medium flow within multiple time steps, improving the forward-looking and stability of the control; At the heat transfer execution level, combined with the partition dynamic heat transfer strategy, the inside of the heat exchanger is divided into a high-temperature zone, a medium-temperature zone, and a low-temperature zone, and an adjustable baffle or a rotary valve is used to dynamically adjust the flue gas flow distribution in each zone according to the heat transfer medium flow adjustment instructions calculated by DMC, optimize the heat load distribution, improve the heat transfer efficiency, and reduce the situation of local overheating or uneven temperature distribution.

[0081] In the present invention, a reinforcement learning algorithm is used to optimize the DMC control weights. Through a feedback adjustment strategy, the control parameters are dynamically adjusted according to historical errors, improving the adaptive ability of the heat exchanger under different boiler load conditions; Specifically, the reinforcement learning objective function calculates the temperature prediction error and optimizes the DMC weights in combination with the gradient descent method. During long-term operation, the adjustment strategy is continuously optimized to enhance the stability of temperature control. Compared with traditional PID and bypass flue gas adjustment schemes, it can respond faster, reduce large fluctuations in the outlet temperature, and improve the waste heat utilization efficiency. Description of the Drawings

[0082] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0083] Figure 1 It is a schematic flow chart of the control method for the flue gas heat exchanger in a thermal power plant based on dynamic matrix control of the present invention. Specific embodiments

[0084] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will make a detailed description of the specific embodiments of the present invention in conjunction with the drawings of the specification.

[0085] Many specific details are set forth in the following description to facilitate a thorough understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0086] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude other embodiments.

[0087] Embodiment 1, referring to Figure 1 , this embodiment provides a control method for the flue gas heat exchanger in a thermal power plant based on dynamic matrix control, including the following steps:

[0088] Step S1, respectively set temperature sensors, flow meters, and pressure sensors at the inlet and outlet of the flue gas heat exchanger, collect flue gas sensing data and boiler operation data in real time, and generate a current working condition data set;

[0089] The flue gas sensing data includes flue gas temperature, flue gas flow rate, flue gas pressure, and heat transfer medium temperature;

[0090] The boiler operation data includes boiler load changes, coal combustion characteristics, and air flow rate;

[0091] Step S2, based on the current working condition data set in step S1, use the long short-term memory network LSTM to construct a flue gas temperature prediction model, predict the change trend of the flue gas temperature in the future time period, and generate a temperature change adjustment instruction;

[0092] Construct a flue gas temperature prediction model using the long short-term memory network (LSTM) to predict the change trend of flue gas temperature in the future time period. The steps to generate a temperature change adjustment instruction are as follows:

[0093] Construct a flue gas temperature prediction model and collect the current working condition dataset D t , which is used for LSTM network training. Denote the dataset as D t :

[0094] D t ={T g ,Q g ,P g ,T m ,L,C,Q a},

[0095] where T g represents the flue gas temperature, Q g represents the flue gas flow rate, P g represents the flue gas pressure, T m represents the heat transfer medium temperature, L represents the boiler load, C represents the coal combustion characteristics, and Q a represents the air flow rate;

[0096] The state update process of the LSTM cell includes:

[0097] i t =σ(W i [h t-1 ,x t +b i ),

[0098] f t =σ(W f [h t-1 ,x t +b f ),

[0099] o t =σ(W o [h t-1 ,x t +b o ),

[0100]

[0101] h t =o t ⊙tanh(C t ),

[0102] where i t represents the input gate activation value, f t represents the forget gate activation value, o t represents the output gate activation value, Represents the candidate status value, C t Represents the unit status value, C t1 Represents the unit status value at the previous time step, h t Represents the hidden status value, h t1 Represents the hidden status value at the previous time step, W i ,W f ,W o ,W c Is the weight matrix, b i ,b f ,b o ,b c Is the bias term, x t Represents the input data, σ represents the Sigmoid activation function, tanh represents the hyperbolic tangent activation function, and ⊙ represents the element-wise multiplication operation;

[0103] The hidden status h calculated based on the LSTM t , predicts the flue gas temperature at future time steps, and the prediction formula is:

[0104]

[0105] Where, Represents the predicted flue gas temperature at time t+τ, f(·) is the LSTM prediction model, and θ is the set of parameters obtained by LSTM training,

[0106] Generates a temperature change adjustment instruction, and the generation process includes:

[0107] The temperature change trend, and the calculation formula is:

[0108]

[0109] Where, ΔT g Is the flue gas temperature change amount, Is the predicted flue gas temperature, T g,t Is the flue gas temperature at the current time step t,

[0110] If |ΔT g |>δ T , then generate a temperature change adjustment instruction:

[0111] U T =g(ΔT g ),

[0112] Where, δ T Is the temperature change threshold, and U T Represents the temperature change adjustment instruction, and g(·) is the adjustment function;

[0113] Specifically, in this step, a long short-term memory network is used to model the trend of flue gas temperature changes. A time series dataset is constructed based on boiler operation data and flue gas sensing data. The LSTM captures the long-term dependence relationship of temperature changes through a gating mechanism and predicts the flue gas temperature at future time steps based on the hidden state. When the temperature prediction error is large, the system generates a temperature change adjustment instruction to dynamically adjust the flue gas temperature;

[0114] Step S3: Receive the temperature change adjustment instruction generated in step S2, and use dynamic matrix control (DMC) to perform predictive control on the outlet temperature of the heat exchanger to generate a heat transfer medium flow rate adjustment instruction;

[0115] In step S3, a state space model of the heat exchanger is established. By integrating historical temperature change data, current measurement data, and future predicted temperature trends, a heat transfer medium flow rate adjustment plan is calculated and a heat transfer medium flow rate adjustment instruction is generated;

[0116] The steps of using dynamic matrix control (DMC) to perform predictive control on the outlet temperature of the heat exchanger and generating a heat transfer medium flow rate adjustment instruction are as follows:

[0117] Establish a state space model of the heat exchanger. The state space equation of the heat exchanger is defined as:

[0118] X t+1 =AX t +BU t ,

[0119] Y t =CX t +DU t ,

[0120] where X t+1 represents the state variable vector at the next time step, X t represents the state variable vector at the current time step, U t represents the input variable vector at the current time step, Y t represents the output variable vector at the current time step, A represents the heat exchanger state transition matrix, describing the evolution relationship of the heat exchanger state over time, B represents the input influence matrix, describing the influence of the heat transfer medium flow rate on the heat exchanger state, C represents the measurement matrix, mapping the state variable to the output variable, and D represents the input direct influence matrix, describing the direct influence of the input on the output.

[0121] Predict the temperature change based on historical data and future inputs. The prediction formula is:

[0122]

[0123] where, is the predicted heat transfer medium temperature at the future time step t + τ, X t+τis the state variable at time step t+τ, U t+τ is the input variable at time step t+τ,

[0124] DMC adopts a rolling optimization strategy to calculate the optimal heat transfer medium flow rate adjustment command. The calculation formula is:

[0125]

[0126] where, is the optimal heat transfer medium flow rate adjustment command, is the desired heat transfer medium temperature, N is the prediction step length, w τ is the weight coefficient at time step τ, λ is the regularization coefficient, ΔU m is the change in the heat transfer medium flow rate control input, U m represents the control input of the heat transfer medium flow rate;

[0127] Specifically, based on the state space model of the heat exchanger, dynamic matrix control is used for predictive control of the outlet temperature. DMC predicts the future outlet temperature of the heat exchanger through rolling optimization and calculates the optimal heat transfer medium flow rate adjustment command according to the prediction results. Compared with the traditional PID control, this method can effectively consider the future temperature change trend, reduce temperature fluctuations, and improve the stability of the heat exchanger outlet temperature. The establishment of the state space model provides dynamic input for DMC, enabling the control system to adapt to complex working conditions changes, thereby improving the energy efficiency and response ability of the heat exchanger;

[0128] Step S4, adopt the heat transfer medium flow rate adjustment command in step S3 to control the multi-channel heat transfer tube bundle structure inside the heat exchanger and dynamically distribute the flue gas flow rate entering different heat transfer areas;

[0129] In step S4, the heat transfer area is divided into a high-temperature area, a medium-temperature area, and a low-temperature area;

[0130] In step S4, according to the heat transfer medium flow rate adjustment command, use adjustable baffles or rotary valves to adjust the flue gas flow rate ratio entering each partition and generate heat exchanger outlet temperature adjustment feedback data;

[0131] The steps of adopting the heat transfer medium flow rate adjustment command in step S3 to control the multi-channel heat transfer tube bundle structure inside the heat exchanger and dynamically distribute the flue gas flow rate entering different heat transfer areas are,

[0132] Divide the internal area of the heat exchanger into:

[0133] A high-temperature area for initially reducing the flue gas temperature,

[0134] A medium-temperature area for further heat transfer,

[0135] And a low-temperature area for final flue gas exhaust;

[0136] Adjust the flue gas flow rate ratio by using a regulating baffle or a rotary valve. The adjustment process is expressed as:

[0137] Q g,i = α i Q g ,

[0138] where Q g,i represents the flue gas flow rate entering the i-th heat exchange area, Q g represents the total flue gas flow rate, and α i is the distribution coefficient, satisfying ∑ i α i = 1;

[0139] Specifically, based on the heat transfer medium flow rate adjustment instruction calculated in step S3, dynamically control the distribution of the flue gas flow rate inside the heat exchange tube bundle structure, divide it into a high-temperature zone, a medium-temperature zone, and a low-temperature zone, and use an adjustable baffle or a rotary valve to dynamically adjust the flue gas flow rate entering different zones to optimize the heat transfer efficiency;

[0140] Step S5: Based on the temperature adjustment feedback data at the outlet of the heat exchanger, adaptively optimize the DMC control parameters in step S3 based on the reinforcement learning algorithm;

[0141] In step S5, compare the deviation between the predicted temperature and the actual outlet temperature to determine the error range of the DMC control strategy;

[0142] In step S5, update the DMC control weights based on the error range and adjust the partition heat exchange adjustment parameters;

[0143] The steps for adaptively optimizing the DMC control parameters in step S3 based on the temperature adjustment feedback data at the outlet of the heat exchanger and based on the reinforcement learning algorithm are as follows:

[0144] Calculate the prediction error of the heat exchanger outlet temperature at each time step t. The calculation formula is:

[0145]

[0146] where e t represents the temperature prediction error, represents the temperature of the heat transfer medium predicted at time step t, T m,t represents the temperature of the heat transfer medium actually measured at time step t,

[0147] Set the reinforcement learning goal and construct the objective function J to measure the optimization degree of the control strategy:

[0148]

[0149] Among them, J represents the control optimization objective function, T is the number of time steps in the optimization period, λ is the regularization coefficient, and Δw τ represents the change in the control parameter weight,

[0150] The DMC control weight is updated using the gradient descent method, and the update formula is:

[0151]

[0152] Wherein, represents the DMC control parameter weight updated at time step t + 1, represents the current DMC control parameter weight at time step t, η is the learning rate, represents the gradient of the loss function J with respect to w τ of,

[0153] Based on the optimized DMC control weight, the flue gas flow distribution in the heat exchange area is adjusted, and the adjustment formula is:

[0154]

[0155] Wherein, represents the flue gas flow distribution coefficient of the i-th heat exchange area updated at time step t + 1, represents the flue gas flow distribution coefficient of the i-th heat exchange area at time step t, and γ is the adjustment step size;

[0156] Specifically, the control parameters of the dynamic matrix control are optimized using the reinforcement learning algorithm. Based on the feedback data of the outlet temperature of the heat exchanger, the DMC control strategy is adaptively adjusted. By calculating the temperature prediction error and using the gradient descent method to update the DMC control weight, the control strategy can be optimized and adjusted according to the historical error, and the control parameters can be adaptively adjusted to improve the self-adaptability of the system.

[0157] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A control method for flue gas heat exchanger in a thermal power plant based on dynamic matrix control, characterized in that: The method steps include: Step S1, respectively setting a temperature sensor, a flow meter and a pressure sensor at the inlet and outlet of the flue gas heat exchanger to collect flue gas sensing data and boiler operation data in real time to generate a current operating condition data set; Step S2, based on the current working condition data set of step S1, a flue gas temperature prediction model is constructed using a long short-term memory network LSTM to predict the flue gas temperature change trend in the future time period and generate a temperature change adjustment instruction; Step S3, receiving the temperature change adjustment instruction generated in step S2, using dynamic matrix control DMC to perform predictive control on the outlet temperature of the heat exchanger, and generating a heat exchange medium flow adjustment instruction; Step S4, using the heat exchange medium flow rate adjustment instruction of step S3 to control the multi-channel heat exchange tube bundle structure inside the heat exchanger, and dynamically distribute the flue gas flow entering different heat exchange areas; Step S5, based on the temperature adjustment feedback data at the heat exchanger outlet, the DMC control parameters of step S3 are adaptively optimized based on the reinforcement learning algorithm.

2. A control method for a flue gas heat exchanger in a thermal power plant based on dynamic matrix control as claimed in claim 1, characterized in that: The flue gas sensing data includes flue gas temperature, flue gas flow, flue gas pressure and heat exchange medium temperature; The boiler operation data includes boiler load changes, coal combustion characteristics, and air flow.

3. A control method for a flue gas heat exchanger in a thermal power plant based on dynamic matrix control as claimed in claim 2, characterized in that: The steps of constructing a flue gas temperature prediction model using a long short-term memory network LSTM are as follows: Construct a flue gas temperature prediction model and collect the current operating condition data set D t , used for LSTM network training, the data set is represented as D t : D t ={T g ,Q g ,P g ,T m ,L,C,Q a }, Among them, T g represents the flue gas temperature, Q g Indicates the flue gas flow rate, P g Indicates the flue gas pressure, T m represents the temperature of the heat exchange medium, L represents the boiler load, C represents the coal characteristics, Q a Indicates air flow rate; The state update process of LSTM cells includes: i t =σ(W i [h t-1 ,x t ]+b i ), f t =σ(W f [h t-1 ,x t ]+b f ), the t =σ(W o [h t-1 ,x t ]+b o ), h t =o t ⊙tanh(C t ), Among them, i t represents the input gate activation value, f t represents the activation value of the forget gate, o t represents the output gate activation value, represents the candidate state value, C t Indicates the unit status value, C t1 represents the cell state value at the previous time step, h t represents the hidden state value, h t1 represents the hidden state value of the previous time step, W i , W f , W o , W c is the weight matrix, b i , b f , b o , b c is the bias term, x t Represents input data, σ represents the Sigmoid activation function, tanh represents the hyperbolic tangent activation function, and ⊙ represents element-by-element multiplication operation.

4. A control method for a flue gas heat exchanger in a thermal power plant based on dynamic matrix control as claimed in claim 3, characterized in that: The step of predicting the flue gas temperature change trend in the future time period and generating the temperature change adjustment instruction is as follows: The hidden state h calculated based on LSTM t , predict the smoke temperature in the future time step, the prediction formula is: in, represents the predicted smoke temperature at time t+τ, f(·) is the LSTM prediction model, θ is the parameter set obtained by LSTM training, Generate temperature change adjustment instructions, the generation process includes: Temperature change trend, calculated as: Where, ΔT g is the flue gas temperature change, To predict the flue gas temperature, T g，t is the flue gas temperature at the current time step t, If |ΔT g |>δ T , then generate temperature change adjustment instructions: U T =g(ΔT g ), Among them, δ T is the temperature change threshold, U T represents the temperature change regulation instruction, and g(·) is the regulation function.

5. A control method for a flue gas heat exchanger in a thermal power plant based on dynamic matrix control as claimed in claim 4, characterized in that: In step S3, a state space model of the heat exchanger is established, historical temperature change data, current measurement data and future predicted temperature trends are integrated to calculate a heat exchange medium flow adjustment plan and generate a heat exchange medium flow adjustment instruction.

6. A control method for a flue gas heat exchanger in a thermal power plant based on dynamic matrix control as claimed in claim 5, characterized in that: The steps of using dynamic matrix control DMC to predictively control the outlet temperature of the heat exchanger and generate a heat exchange medium flow adjustment instruction are as follows: The state space model of the heat exchanger is established, and the state space equation of the heat exchanger is defined as: X t+1 =AX t +BU t , Y t CX t +YOU t , Among them, X t+1 represents the state variable vector for the next time step, X t represents the state variable vector of the current time step, U t Represents the input variable vector of the current time step, Y t represents the output variable vector of the current time step, A represents the heat exchanger state transfer matrix, which describes the evolution of the heat exchanger state over time, B represents the input influence matrix, which describes the influence of the heat exchange medium flow on the heat exchanger state, C represents the measurement matrix, which maps the state variables to the output variables, and D represents the input direct influence matrix, which describes the direct influence of the input on the output. Based on historical data and future inputs, the temperature change is predicted using the following formula: in, is the predicted temperature of the heat transfer medium at the future time step t+τ, X t+τ is the state variable at time step t+τ, U t+τ is the input variable at time step t+τ, DMC adopts a rolling optimization strategy to calculate the optimal heat exchange medium flow regulation instruction. The calculation formula is: in, is the optimal heat exchange medium flow regulation instruction, is the expected heat transfer medium temperature, N is the prediction step length, w τ is the weight coefficient of the time step τ, λ is the regularization coefficient, ΔU m is the input change of heat exchange medium flow control, U m Represents the control input for the heat transfer medium flow rate.

7. A control method for a flue gas heat exchanger in a thermal power plant based on dynamic matrix control as claimed in claim 6, characterized in that: In step S4, the heat exchange area is divided into a high temperature area, a medium temperature area and a low temperature area; In step S4, according to the heat exchange medium flow adjustment instruction, the adjustable baffle or rotary valve is used to adjust the flue gas flow ratio entering each partition, and the heat exchanger outlet temperature adjustment feedback data is generated.

8. A control method for a flue gas heat exchanger in a thermal power plant based on dynamic matrix control as claimed in claim 7, characterized in that: The step of using the heat exchange medium flow rate adjustment instruction of step S3 to control the multi-channel heat exchange tube bundle structure inside the heat exchanger and dynamically distribute the flue gas flow entering different heat exchange areas is as follows: The internal area of ​​the heat exchanger is divided into: High temperature zone, used to initially reduce the flue gas temperature, Medium temperature zone, used for further heat exchange, and a low temperature zone for final smoke exhaust; The smoke flow ratio is adjusted by adjusting the damper or rotating valve. The adjustment process is expressed as: Q g，i =α i Q g , Among them, Q g，i represents the flue gas flow rate entering the i-th heat exchange area, Q g represents the total flue gas flow rate, α i is the distribution coefficient, satisfying ∑ i α i =1.

9. A control method for a flue gas heat exchanger in a thermal power plant based on dynamic matrix control as claimed in claim 8, characterized in that: In step S5, the deviation between the predicted temperature and the actual outlet temperature is compared to determine the error range of the DMC control strategy; In step S5, the DMC control weight is updated based on the error range, and the partition heat exchange adjustment parameters are adjusted.

10. A control method for a flue gas heat exchanger in a thermal power plant based on dynamic matrix control as claimed in claim 9, characterized in that: The step of adaptively optimizing the DMC control parameters of step S3 based on the temperature adjustment feedback data of the heat exchanger outlet and based on the reinforcement learning algorithm is as follows: The prediction error of the heat exchanger outlet temperature is calculated at each time step t, and the calculation formula is: Among them, e t represents the temperature prediction error, represents the predicted temperature of the heat transfer medium at time step t, T m，t represents the actual measured temperature of the heat transfer medium at time step t, Set the reinforcement learning goal, construct the objective function J, and measure the optimization degree of the control strategy: Where J represents the control optimization objective function, T is the number of time steps in the optimization cycle, λ is the regularization coefficient, and Δw τ represents the change in the weight of the control parameter, The gradient descent method is used to update the DMC control weight, and the update formula is: in, represents the weight of the DMC control parameter after the update at time step t+1, represents the current DMC control parameter weight at time step t, η is the learning rate, Denotes the loss function J with respect to w τ The gradient of Based on the optimized DMC control weight, the flue gas flow distribution in the heat exchange area is adjusted. The adjustment formula is: in, represents the flue gas flow distribution coefficient of the ith heat exchange area after the update at time step t+1, It represents the flue gas flow distribution coefficient of the i-th heat exchange area at time step t, and γ is the adjustment step length.

Images