A reboiler energy consumption analysis and control method and system

Abstract

The present invention provides a reboiler energy consumption analysis and control method and system, which relates to the technical field of intelligent control. The method includes: constructing a reboiler energy consumption model; analyzing the reboiler energy consumption model to determine the total energy consumption of the reboiler during operation; constructing a reboiler control module based on a PID neural network; aiming at reducing the total energy consumption of the reboiler during operation, increasing the heat flux in the evaporation zone and the heat transfer coefficient in the evaporation zone, optimizing the reboiler control module based on the PID neural network; setting the target steam mass flow output by the reboiler; obtaining the actual steam mass flow output by the reboiler; and controlling the reboiler through the optimized reboiler control module based on the PID neural network. The present invention overcomes the limitation of setting control parameters depending on experience, has better effects when facing complex dynamic characteristics and nonlinear problems, and can fully cope with the changes of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control, and particularly to a reboiler energy consumption analysis and control method and system. Background Art

[0002] With the continuous advancement of the industrialization process, the demand for efficient energy consumption management and optimization in industries such as chemical engineering, petroleum, and energy is increasing day by day. As a key heat exchange device in these industries, reboilers are widely used in heating, evaporation, and separation processes. Since it consumes a large amount of energy during the production process, the energy consumption management and optimization of reboilers directly affect production efficiency and operating costs. Therefore, how to reduce the total energy consumption of reboilers and improve their energy efficiency has become an important research direction in the industrial field.

[0003] In the existing technical solutions, traditional reboiler energy consumption analysis and control methods usually rely on empirical setting of control parameters, and have limited effects when facing complex dynamic characteristics and nonlinear problems, and cannot fully cope with system changes.

[0004] In addition, existing energy consumption management and optimization technologies often lack the ability of dynamic adjustment and are difficult to achieve comprehensive optimization of reboiler energy efficiency. Summary of the Invention

[0005] In order to solve the technical problems that traditional reboiler energy consumption analysis and control methods usually rely on empirical setting of control parameters, have limited effects when facing complex dynamic characteristics and nonlinear problems, cannot fully cope with system changes, and existing energy consumption management and optimization technologies often lack the ability of dynamic adjustment and are difficult to achieve comprehensive optimization of reboiler energy efficiency, the present invention provides a reboiler energy consumption analysis and control method and system.

[0006] The technical solutions provided by the embodiments of the present invention are as follows:

[0007] First aspect:

[0008] A reboiler energy consumption analysis and control method provided by an embodiment of the present invention includes:

[0009] S1: Construct a reboiler energy consumption model;

[0010] S2: Analyze the reboiler energy consumption model to determine the total energy consumption of the reboiler during operation;

[0011] S3: Construct a reboiler control module based on a PID neural network;

[0012] S4: Optimize the reboiler control module based on the PID neural network with the goal of reducing the total energy consumption of the reboiler during operation, increasing the heat flux in the evaporation zone, and the heat transfer coefficient in the evaporation zone;

[0013] S5: Set the target steam mass flow rate output by the reboiler;

[0014] S6: Obtain the actual steam mass flow rate output by the reboiler;

[0015] S7: Control the reboiler through the optimized reboiler control module based on the PID neural network.

[0016] Second aspect:

[0017] A reboiler energy consumption analysis and control system provided by an embodiment of the present invention includes: a memory and one or more processors;

[0018] One or more application programs are stored in the memory, and the one or more application programs are adapted to be executed by the one or more processors to implement the above-mentioned reboiler energy consumption analysis and control method.

[0019] The beneficial effects brought by the technical solution provided by the embodiment of the present invention at least include:

[0020] In the present invention, by constructing a reboiler control module based on the PID neural network and using the PID neural network control module to replace the traditional control method, the limitation of relying on empirical setting of control parameters is overcome, and it has better effects when facing complex dynamic characteristics and nonlinear problems, and can fully cope with the changes of the system. Further, with the goal of reducing the total energy consumption of the reboiler during operation, increasing the heat flux in the evaporation zone, and the heat transfer coefficient in the evaporation zone, the reboiler control module based on the PID neural network is optimized, and the reboiler is controlled through the optimized reboiler control module based on the PID neural network, so as to provide dynamic adjustment ability for energy consumption management and optimization technology, and it is easier to achieve the overall optimization of the reboiler energy efficiency. Description of the Drawings

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0022] Figure 1 It is a schematic flowchart of a reboiler energy consumption analysis and control method provided by an embodiment of the present invention;

[0023] Figure 2 It is a schematic structural diagram of a reboiler energy consumption analysis and control system provided by an embodiment of the present invention. Detailed Embodiments

[0024] The technical solutions in the present invention will be described below with reference to the accompanying drawings.

[0025] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as an "example" in the present invention should not be construed as being more preferred or more advantageous than other embodiments or design solutions. Rather, the use of the word "example" is intended to present concepts in a specific manner. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two.

[0026] To make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0027] Refer to the attached Figure 1 illustrates a schematic flow chart of a reboiler energy consumption analysis and control method provided by an embodiment of the present invention.

[0028] The embodiments of the present invention provide a reboiler energy consumption analysis and control method, which can be implemented by a reboiler energy consumption analysis and control device. The reboiler energy consumption analysis and control device can be a terminal or a server. The processing flow of the reboiler energy consumption analysis and control method can include the following steps:

[0029] S1: Build a reboiler energy consumption model.

[0030] It should be noted that a reboiler is an important heat exchange device, usually installed at the bottom of a distillation column or a separation column. Its main function is to generate steam by heating a liquid mixture, so as to maintain the distillation or separation process in the column. The reboiler heats the bottom liquid, causing the volatile components therein to evaporate, forming rising steam, and these steam return to the column for further separation or purification.

[0031] Furthermore, by building a reboiler energy consumption model, various energy consumptions during the operation of the reboiler can be comprehensively understood, such as liquid heat energy, steam energy consumption, heat loss, and auxiliary equipment energy consumption.

[0032] Among them, the liquid heat energy refers to the heat required for the liquid entering the reboiler, used to heat the liquid to a certain temperature or cause partial evaporation of the liquid.

[0033] Among them, the steam energy consumption refers to the amount of steam consumed by the reboiler during operation, used to provide sufficient heat for the liquid to generate steam.

[0034] Among them, the heat loss refers to the part of the heat that cannot be effectively utilized during the operation of the reboiler.

[0035] Among them, the energy consumption of auxiliary equipment includes the energy consumed by auxiliary equipment such as circulation pumps and control systems during the operation of the reboiler. Although this part of the energy consumption is relatively small, it is still part of the overall energy consumption, especially in the case of long-term operation, it will accumulate into significant energy consumption.

[0036] S2: Analyze the energy consumption model of the reboiler to determine the total energy consumption during the operation of the reboiler.

[0037] Optionally, the total energy consumption includes: liquid thermal energy, steam energy consumption, heat loss, and auxiliary equipment energy consumption.

[0038] Among them, the calculation method of the total energy consumption is specifically:

[0039] Q total = Q liquid + Q steam + Q loss + Q aux

[0040] Among them, Q total represents the total energy consumption, Q liquid represents the liquid thermal energy, Q steam represents the steam energy consumption, Q loss represents the heat loss, Q aux represents the auxiliary equipment energy consumption.

[0041] Among them, the calculation method of the liquid thermal energy is specifically:

[0042] Q liquid = m C · c p · (T pinch - T in )

[0043] Among them, m C represents the mass of the liquid entering the reboiler, c p represents the specific heat capacity of the liquid, T pinch represents the pinch temperature, T in represents the temperature of the liquid when it enters the reboiler.

[0044] Among them, the pinch temperature is the intersection temperature of the two linear temperature trends in the heating zone and the evaporation zone during the heat exchange process.

[0045] Among them, the calculation method of the steam energy consumption is specifically:

[0046]

[0047] Among them, represents the mass flow rate of the steam generated by the reboiler, H steam represents the enthalpy value of the steam.

[0048] Among them, the calculation method of heat loss is specifically as follows:

[0049] Q loss = U surface ·A surface ·ΔT surface

[0050] Among them, U surface represents the overall heat transfer coefficient on the surface of the reboiler, A surface represents the outer surface area of the reboiler, and ΔT surface represents the temperature difference between the surface of the reboiler and the environment.

[0051] Among them, the calculation method of the energy consumption of auxiliary equipment is specifically as follows:

[0052]

[0053] Among them, and respectively represent the powers of the circulation pump and the control system, and t pump and t control respectively represent the operating times of the circulation pump and the control system.

[0054] In the present invention, by decomposing the total energy consumption into liquid thermal energy, steam energy consumption, heat loss, and auxiliary equipment energy consumption, the energy consumption situation of the reboiler in each link can be understood in detail. This helps to identify which parts are the main energy consumption sources and which parts may have room for optimization, thus laying a foundation for improving the overall system efficiency. After clarifying the energy consumption of each part, the operating parameters of the reboiler can be optimized more targeted. For example, by reducing heat loss or optimizing steam energy consumption, the total energy consumption can be significantly reduced. Precise energy consumption analysis provides a scientific basis for the subsequent optimization of control strategies and avoids waste of resources caused by blind adjustment.

[0055] S3: Construct a reboiler control module based on a PID neural network.

[0056] Among them, the PID neural network (PID Neural Network, PID-NN) is an intelligent control system that combines traditional PID controllers and neural network technologies. It uses the self-learning ability of neural networks to optimize the parameters of PID controllers, thereby improving the control performance of the system. The PID neural network can better adapt to the dynamic characteristics of the system and external disturbances by adjusting the proportional, integral, and derivative parameters of the PID controller in real time, and achieve more precise control effects.

[0057] Introduce the working principle of the reboiler control module based on the PID neural network. The PID controller is a classic feedback controller that controls the output of the system by adjusting three parameters: Proportional, Integral, and Derivative. The goal of the PID controller is to make the system output (such as steam mass flow) close to the set value (target value). Traditional PID controllers may not perform ideally in complex, non-linear systems because their parameters are usually fixed, while the reboiler system has dynamic and non-linear characteristics. To enhance the adaptability of the system, the PID neural network combines the self-learning function of the neural network, enabling the PID parameters to be dynamically adjusted according to the operating state of the system. The PID neural network usually consists of an input layer, a hidden layer, and an output layer. The input layer receives the error signal of the system (such as the difference between the target steam mass flow and the actual steam mass flow), the hidden layer learns the relationship between the error and the control signal, and the output layer generates a new control signal.

[0058] In a possible implementation, the reboiler control module based on the PID neural network is specifically used for:

[0059] Obtain the input of each neuron in the input layer:

[0060] x1(k) = r(k)

[0061] x2(k) = y(k)

[0062] Where x1(k) represents the input of the first neuron in the input layer at the k-th sampling moment, r(k) represents the target steam mass flow set for the reboiler at the k-th sampling moment, x2(k) represents the input of the second neuron in the input layer at the k-th sampling moment, and y(k) represents the actual steam mass flow output by the reboiler at the k-th sampling moment.

[0063] Calculate the input of each neuron in the hidden layer:

[0064]

[0065] Where n i (k) represents the input of the i-th neuron in the hidden layer at the k-th sampling moment, w ij represents the connection weight between the j-th neuron in the input layer and the i-th neuron in the hidden layer, and x j (k) represents the input of the j-th neuron in the input layer at the k-th sampling moment.

[0066] According to the input of each neuron in the hidden layer, calculate the output of each neuron in the hidden layer:

[0067] u1(k) = n1(k)

[0068] u2(k) = n2(k) + u2(k - 1)

[0069] u3(k) = n3(k) - n3(k - 1)

[0070] Among them, u1(k) represents the output of the hidden layer proportional neuron at the k - th sampling moment, n1(k) represents the input of the hidden layer proportional neuron at the k - th sampling moment, u2(k) represents the output of the hidden layer integral neuron at the k - th sampling moment, n2(k) represents the input of the hidden layer integral neuron at the k - th sampling moment, u2(k - 1) represents the output of the hidden layer integral neuron at the (k - 1)-th sampling moment, u3(k) represents the output of the hidden layer differential neuron at the k - th sampling moment, n3(k) represents the input of the hidden layer differential neuron at the k - th sampling moment, and n3(k - 1) represents the input of the hidden layer differential neuron at the (k - 1)-th sampling moment.

[0071] Calculate the output of the output layer according to the output of each neuron in the hidden layer:

[0072]

[0073] Among them, u(k) represents the control signal output by the output layer at the k - th sampling moment, w i represents the connection weight between the i - th neuron in the hidden layer and the output layer, and u i (k) represents the output of the i - th neuron in the hidden layer at the k - th sampling moment.

[0074] In the present invention, the control module based on the PID neural network combines the advantages of the fast response of the PID controller and the learning and memory characteristics of the neural network. This combination can enhance the adaptability of the control system to nonlinear and complex dynamic characteristics while maintaining the fast response of the control system, ensuring the stability of the system. Through more precise control, the PID neural network can optimize the operation efficiency of the reboiler, reduce unnecessary energy consumption, and thus achieve more efficient energy consumption management. This is of great significance for saving energy and reducing operating costs.

[0075] S4: Optimize the reboiler control module based on the PID neural network with the goal of reducing the total energy consumption during the operation of the reboiler, increasing the heat flux in the evaporation zone, and the heat transfer coefficient in the evaporation zone.

[0076] It should be noted that the heat flux and heat transfer coefficient in the evaporation zone have a direct impact on the efficiency and performance of the reboiler.

[0077] Among them, the heat flux in the evaporation zone refers to the heat transferred per unit area in the evaporation zone per unit time. It reflects the efficiency and intensity of the evaporation process and directly affects the evaporation capacity and production efficiency of the reboiler.

[0078] Among them, the heat transfer coefficient in the evaporation zone is an important parameter to measure the heat transfer capacity of the reboiler evaporation zone, indicating the efficiency of heat transfer from the heating medium to the liquid through the reboiler wall.

[0079] Optionally, the calculation method of the heat flux in the evaporation zone is specifically as follows:

[0080] Calculate the heat transfer area of the evaporation zone of the reboiler:

[0081] A EZ = W·h EZ ·(2n s - 2)

[0082] Among them, A EZ represents the heat transfer area of the evaporation zone, W represents the width of the reboiler, h EZ represents the height of the evaporation zone, and n s represents the number of pillow plates in the reboiler.

[0083] Calculate the heat flow rate of the heating zone of the reboiler:

[0084]

[0085] Among them, represents the heat flow rate of the heating zone, represents the mass flow rate of the liquid entering the reboiler, c p represents the specific heat capacity of the liquid, T pinch represents the pinch temperature, and T in represents the temperature of the liquid when it enters the reboiler.

[0086] Based on the steam mass flow rate generated by the reboiler, calculate the total heat flow rate of the reboiler:

[0087]

[0088] Among them, represents the total heat flow rate of the reboiler, represents the mass flow rate of the steam generated by the reboiler, and ΔH steam represents the liquid vaporization potential.

[0089] According to the heat flow rate of the heating zone and the total heat flow rate of the reboiler, calculate the heat flow rate of the evaporation zone:

[0090]

[0091] Among them, represents the heat flow rate of the evaporation zone, represents the total heat flow rate of the reboiler.

[0092] According to the heat flow rate of the evaporation zone and the heat transfer area of the evaporation zone, calculate the heat flux in the evaporation zone:

[0093]

[0094] Among them, represents the heat flux in the evaporation zone.

[0095] Optionally, the calculation method of the heat transfer coefficient in the evaporation zone is specifically as follows:

[0096] Calculate the average temperature difference in the evaporation zone according to the pinch temperature:

[0097] ΔT EZ = T HS -(T pinch + T out ) / 2

[0098] Among them, ΔT EZ represents the average temperature difference in the evaporation zone, T HS represents the temperature of the heating surface, and T out represents the temperature when the steam leaves the reboiler.

[0099] Calculate the heat transfer coefficient in the evaporation zone according to the heat flow rate in the evaporation zone, the heat transfer area in the evaporation zone, and the average temperature difference in the evaporation zone:

[0100]

[0101] Among them, U EZ represents the heat transfer coefficient in the evaporation zone.

[0102] In the present invention, by optimizing the control strategy to reduce the total energy consumption of the reboiler, the energy efficiency of the equipment can be significantly improved. This not only helps to reduce energy consumption and operating costs, but also improves the overall economic efficiency of the reboiler and realizes a more green and sustainable production. Increasing the heat flux in the evaporation zone means that heat energy can be more effectively converted into steam, improving the evaporation efficiency. Through the optimized control module, the reboiler can utilize the input heat energy more efficiently, produce more steam, and meet the production requirements. By increasing the heat transfer coefficient in the evaporation zone, the heat transfer effect from the heating surface to the liquid can be enhanced. This will help to reduce the heat loss during the evaporation process, ensure that more heat is effectively utilized, and thus improve the overall performance of the reboiler.

[0103] In a possible implementation manner, S4 specifically includes sub-steps S401 to S403:

[0104] S401: Construct an objective function with the goal of reducing the total energy consumption of the reboiler during operation, increasing the heat flux in the evaporation zone and the heat transfer coefficient in the evaporation zone during the operation of the reboiler.

[0105] Among them, the objective function is specifically:

[0106]

[0107] Among them, f represents the objective function, represents the heat flux in the evaporation zone, and U EZ represents the heat transfer coefficient in the evaporation zone, and Q total represents the total energy consumption, α1 represents the weight parameter of the heat flux in the evaporation zone, α2 represents the weight parameter of the heat transfer coefficient in the evaporation zone, and α3 represents the weight parameter of the total energy consumption.

[0108] Among them, those skilled in the art can set the magnitudes of the weight parameter α1 of the heat flux in the evaporation zone, the weight parameter α2 of the heat transfer coefficient in the evaporation zone, and the weight parameter α3 of the total energy consumption according to the actual situation, and the present invention does not make any limitations.

[0109] In the present invention, the constructed objective function combines the heat flux in the evaporation zone, the heat transfer coefficient in the evaporation zone, and the total energy consumption to achieve multi-objective optimization. This method can minimize energy consumption while maintaining high efficiency, ensuring that the system reaches the optimal state in multiple key performance indicators.

[0110] It should be noted that the training of the PID neural network usually adopts the gradient descent method. Although this method can adaptively update the weights and biases of the neural network, there is no fixed algorithm for the selection of the initial controller parameters, and they are usually randomly selected. Since the initial controller parameters determine the convergence direction and the starting point of learning of the neural network, the randomly selected initial values may cause the network to start training from an uncertain direction. This will increase the training time because more iterations and learning are required to find the correct convergence direction. In addition, the randomly selected initial parameters may also lead to the risk of the weights falling into local optimal solutions, thereby reducing the overall convergence speed of the network.

[0111] In order to improve the control effect of the PID neural network, appropriate initial controller parameters must be selected. The present invention innovatively proposes a method for optimizing the initial controller parameters through the cuckoo search algorithm.

[0112] S402: According to the objective function, determine the initial controller parameters of the reboiler control module based on the PID neural network through the cuckoo search algorithm.

[0113] It should be noted that the Cuckoo Search Algorithm is a heuristic optimization algorithm that mimics the parasitic breeding behavior of cuckoos to solve optimization problems. Cuckoos lay their eggs in the nests of other birds and achieve reproduction by randomly placing the eggs and taking advantage of the brood-rearing behavior of the host birds. In the algorithm, the solution space is regarded as the bird's nest, and the solution to the optimization problem is regarded as the cuckoo's egg. The core steps of the algorithm include randomly generating an initial solution (bird's nest), generating a new solution (egg) through the cuckoo's strategy, and evaluating and updating the position of the bird's nest using the fitness function. The Cuckoo Search Algorithm is widely used in function optimization and solving complex problems due to its global search ability and simple structure.

[0114] In a possible implementation, S402 specifically includes S4021 to S4025:

[0115] S4021: Initialize the iteration number t, the discovery probability Pa, the nest size N, and use the objective function as the fitness function of the Cuckoo Search Algorithm.

[0116] Optionally, the discovery probability is set to 0.25.

[0117] S4022: Calculate the fitness value of each initialized bird's nest according to the fitness function.

[0118] S4023: Select the position of the bird's nest with the highest fitness value as the initial bird's nest position.

[0119] S4024: Update the position of the bird's nest:

[0120]

[0121] Levy: μ = t (-λ) , 1 ≤ λ ≤ 3

[0122] Where represents the position of the i-th bird's nest at the (t + 1)-th iteration, represents the i-th bird's nest

[0123] at the t-th iteration, α represents the step size control quantity, represents element-wise multiplication, Levy(λ) represents the Levy random search path, μ represents the random perturbation value of the Levy flight, and λ represents the parameter controlling the step size distribution.

[0124] It should be noted that Levy Flight is a stochastic process used to describe the non-uniform random movement of an object in space. Its characteristic is that the displacement step size follows the Levy distribution, which has a long-tailed property, meaning that most step sizes are short, but occasionally there are longer step sizes. The Levy flight model is usually used to simulate the foraging behavior of organisms or other stochastic processes with long-range dependence characteristics. Compared with the standard random walk, Levy flight can more effectively explore a large range of solution spaces, thus improving the global search ability of the optimization algorithm.

[0125] S4025: Calculate the fitness value of the updated bird nest, and compare the fitness value of the updated bird nest with the fitness value of the initial bird nest. Select the bird nest with the largest fitness value as the current bird nest position.

[0126] S4026: Randomly generate a random number R between 0 and 1 at the current bird nest position, and compare the random number with the discovery probability Pa. In the case where the random number is greater than the discovery probability, update the bird nest position and calculate the fitness value of the updated bird nest. If the fitness value of the updated bird nest is greater than the fitness value of the current bird nest position, update the fitness value. In the case where the random number is less than or equal to the discovery probability, keep the current bird nest position unchanged.

[0127] S4027: Determine whether the maximum number of iterations has been reached. If so, output the optimal fitness value as the initial controller parameter. Otherwise, return to S4024 until the maximum number of iterations is reached.

[0128] It should be noted that by using the objective function as the fitness function of the cuckoo search algorithm, the selection of the controller parameters can be directly associated with the performance optimization objectives of the reboiler (such as reducing the total energy consumption, improving the heat transfer efficiency, etc.). This enables the optimization process to more precisely meet specific process requirements. The Levy flight mechanism in the algorithm allows the use of random search paths, which explore and develop the solution space through a combination of long and short step sizes. Through this adaptive step size adjustment mechanism, it is possible to converge to the optimal solution faster and improve the search efficiency.

[0129] In the present invention, the initial controller parameters of the reboiler control module based on the PID neural network are determined through the cuckoo search algorithm, determining the convergence direction and the starting point of learning of the neural network, facilitating quickly finding the correct convergence direction, and significantly improving the training efficiency, convergence speed, and final control performance of the network.

[0130] S403: Use the initial controller parameters as the starting parameters of the reboiler control module based on the PID neural network. Through the gradient descent algorithm, iteratively update the controller parameters of the reboiler control module based on the PID neural network to optimize the reboiler control module based on the PID neural network.

[0131] It should be noted that the Gradient Descent Algorithm is an iterative algorithm for optimization problems, aiming to minimize the value of the objective function by gradually adjusting the parameters. The algorithm calculates the gradient of the objective function at the current point, indicating the direction in which the function rises fastest, and then adjusts the parameters in the opposite direction of the gradient, thereby gradually approaching the minimum value of the objective function. The adjustment amplitude is determined by the learning rate. The gradient descent algorithm is widely used in the training of machine learning and deep learning models, and can effectively find the optimal parameters of the model, minimize the loss function, and thus improve the prediction performance of the model.

[0132] In a possible implementation, S403 specifically includes sub-steps S4031 and S4032:

[0133] S4031: Construct the loss function of the gradient descent algorithm:

[0134]

[0135] Among them, L represents the loss function, θ represents the controller parameters, r(k) represents the target steam mass flow output by the reboiler set at the k sampling moment, y(k) represents the actual steam mass flow output by the reboiler at the k sampling moment, and K represents the total observation duration.

[0136] The specific way to update the controller parameters in S4032 is:

[0137]

[0138] Among them, θ t+1 represents the controller parameters at the (t + 1)-th iteration, θ t represents the controller parameters at the t-th iteration, θ t-1 represents the controller parameters at the (t - 1)-th iteration, η t represents the adaptive learning rate at the t-th iteration, represents the gradient of the loss function L with respect to the controller parameters θ t and β represents the momentum coefficient.

[0139] In the present invention, momentum optimization can utilize the gradient accumulation effect of previous iterations, enabling the acceleration of parameter updates along long gradient paths. Especially in the case of approaching the convergence point, momentum can help the model jump out of local optima more quickly and move towards the global optimum direction. Introducing the momentum term can help the model reduce oscillation phenomena during the parameter update process when the gradient direction is unstable or there is significant noise, thereby making the training process smoother and more stable.

[0140] Optionally, the calculation method of the adaptive learning rate is as follows:

[0141]

[0142] where η min represents the minimum value of the learning rate, η max represents the maximum value of the learning rate, T max represents the maximum number of iterations, and cos represents the cosine function.

[0143] In the present invention, as the training process progresses, the learning rate smoothly decreases from an initial large value to a small value. This smooth decay method enables the model to quickly learn global features in the early stage of training and refine the learning process in the later stage of training, avoiding missing tiny optimal solutions. At the same time, due to the properties of the cosine function, the adjustment of the learning rate is very smooth, avoiding the training instability that may be caused by sudden changes in the learning rate. Compared with piecewise linear decay or exponential decay, cosine annealing can provide a more stable training process.

[0144] S5: Set the target steam mass flow rate output by the reboiler.

[0145] It should be noted that the target steam mass flow rate refers to the steam mass flow rate that the reboiler aims to achieve.

[0146] S6: Obtain the actual steam mass flow rate output by the reboiler.

[0147] It should be noted that the actual steam mass flow rate output by the reboiler can be monitored through a sensor.

[0148] S7: Control the reboiler through the optimized reboiler control module based on the PID neural network.

[0149] The working principle of how the reboiler control module based on the PID neural network controls the reboiler according to the target steam mass flow rate and the actual steam mass flow rate has been described above. To avoid repetition, the present invention will not elaborate further.

[0150] In the present invention, by setting the target steam mass flow rate and comparing it with the actual steam mass flow rate, the optimized PID neural network control module can adjust the control parameters in a timely manner according to the error, achieving precise control of the reboiler. This closed-loop control method ensures that the amount of steam output by the reboiler strictly conforms to the set target value, avoiding problems of overage or shortage. Using the optimized PID neural network to control the reboiler, the control module can quickly respond to changes in steam demand by continuously learning and adjusting parameters. This dynamic response ability enables the reboiler to quickly adjust its working state, adapt to load fluctuations during the production process, and maintain the efficient operation of the system.

[0151] The beneficial effects brought by the technical solution provided in the embodiment of the present invention at least include:

[0152] In the present invention, by constructing a reboiler control module based on a PID neural network and using the PID neural network control module to replace the traditional control method, the limitation of relying on empirical setting of control parameters is overcome, and it has better effects in the face of complex dynamic characteristics and nonlinear problems, and can fully cope with system changes. Further, with the goal of reducing the total energy consumption during the operation of the reboiler, increasing the heat flux in the evaporation zone, and the heat transfer coefficient in the evaporation zone, the reboiler control module based on the PID neural network is optimized, and the reboiler is controlled by the optimized reboiler control module based on the PID neural network, thereby providing dynamic adjustment ability for energy consumption management and optimization technology and making it easier to achieve comprehensive optimization of the energy efficiency of the reboiler.

[0153] Refer to the attached Figure 2 illustrates a schematic structural diagram of a reboiler energy consumption analysis and control system provided by the present invention.

[0154] The present invention also provides a reboiler energy consumption analysis and control system 30, including: a memory 303 and one or more processors 301.

[0155] One or more application programs are stored in the memory 303, and the one or more application programs are adapted to be executed by the one or more processors 301 to implement the reboiler energy consumption analysis and control method described in the method embodiment.

[0156] The reboiler energy consumption analysis and control system 30 includes: a processor 301 and a memory 303. Among them, the processor 301 and the memory 303 are connected, such as connected through a bus 302.

[0157] The structure of the reboiler energy consumption analysis and control system 30 does not constitute a limitation to the embodiment of the present invention.

[0158] The processor 301 may be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logic blocks, modules, and circuits described in connection with the disclosure of the present invention. The processor 301 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0159] The bus 302 may include a path for transmitting information between the above components. The bus 302 may be a PCI bus or an EISA bus, etc. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity of representation, only a thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0160] The memory 303 may be a ROM or other types of static storage devices that can store static information and instructions, a RAM, or other types of dynamic storage devices that can store information and instructions. It may also be an EEPROM, a CD-ROM, or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0161] It should be noted that the reboiler energy consumption analysis and control system 30 can implement the above-mentioned reboiler energy consumption analysis and control method and can achieve the same or similar technical effects. To avoid repetition, the present invention will not elaborate further.

[0162] The beneficial effects brought by the technical solution provided by the embodiments of the present invention at least include:

[0163] In the present invention, by constructing a reboiler control module based on a PID neural network and using the PID neural network control module to replace the traditional control method, the limitation of relying on empirical setting of control parameters is overcome, and it has better effects when facing complex dynamic characteristics and nonlinear problems, and can fully cope with the changes of the system. Further, by aiming at reducing the total energy consumption of the reboiler during operation, increasing the heat flux in the evaporation zone, and the heat transfer coefficient in the evaporation zone, the reboiler control module based on the PID neural network is optimized, and the reboiler is controlled through the optimized reboiler control module based on the PID neural network, so as to provide dynamic adjustment ability for energy consumption management and optimization technology, and it is easier to achieve the overall optimization of the reboiler energy efficiency. The present invention also provides a computer-readable storage medium, on which a computer program is stored, and the computer program can be loaded and executed by a processor to implement the reboiler energy consumption analysis and control method described in the first aspect.

[0164] As described above, it is only the specific implementation manner of the present invention. However, the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

[0165] The following points need to be explained:

[0166] (1) The accompanying drawings of the embodiments of the present invention only relate to the structures involved in the embodiments of the present invention, and other structures can refer to the general design.

[0167] (2) For clarity, in the accompanying drawings used to describe the embodiments of the present invention, the thickness of layers or regions is enlarged or reduced, that is, these drawings are not drawn according to the actual scale. It can be understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, the element can be "directly" on or under the other element or there can be intermediate elements.

[0168] (3) Without conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0169] As described above, it is only the specific implementation manner of the present invention. However, the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A reboiler energy consumption analysis and control method, characterized in that Including: S1: Construct a reboiler energy consumption model; S2: Analyze the reboiler energy consumption model to determine the total energy consumption of the reboiler during operation; S3: Construct a reboiler control module based on a PID neural network; S4: Optimize the reboiler control module based on a PID neural network with the goal of reducing the total energy consumption of the reboiler during operation, increasing the heat flux in the evaporation zone, and the heat transfer coefficient in the evaporation zone; S5: Set the target steam mass flow rate output by the reboiler; S6: Obtain the actual steam mass flow rate output by the reboiler; S7: Control the reboiler through the optimized reboiler control module based on a PID neural network; Among them, the specific content of S4 includes: S401: Construct an objective function with the goal of reducing the total energy consumption of the reboiler during operation, increasing the heat flux in the evaporation zone, and the heat transfer coefficient in the evaporation zone of the reboiler during operation; S402: According to the objective function, determine the initial controller parameters of the reboiler control module based on a PID neural network through the cuckoo search algorithm; S403: Use the initial controller parameters as the starting parameters of the reboiler control module based on a PID neural network, and through the gradient descent algorithm, iteratively update the controller parameters of the reboiler control module based on a PID neural network to optimize the reboiler control module based on a PID neural network; Among them, the specific objective function is: where f represents the objective function, represents the heat flux in the evaporation zone, and U EZ represents the heat transfer coefficient in the evaporation zone, and Q total represents the total energy consumption, α1 represents the weight parameter of the heat flux in the evaporation zone, α2 represents the weight parameter of the heat transfer coefficient in the evaporation zone, and α3 represents the weight parameter of the total energy consumption.

2. The reboiler energy consumption analysis and control method according to claim 1, wherein The total energy consumption includes: liquid thermal energy, steam energy consumption, heat loss, and auxiliary equipment energy consumption; The specific calculation method of the total energy consumption is: Q total = Q liquid + Q steam + Q loss + Q aux Among them, Q total represents the total energy consumption, Q liquid represents the liquid heat energy, Q steam represents the steam energy consumption, Q loss represents the heat loss, Q aux represents the energy consumption of auxiliary equipment.

3. According to the reboiler energy consumption analysis and control method described in claim 2, the specific calculation method of the liquid thermal energy is: Q liquid = m C · c p · (T pinch - T in ) Among them, Q liquid represents the liquid heat energy, m C represents the mass of the liquid entering the reboiler, c p represents the specific heat capacity of the liquid, T pinch represents the pinch point temperature, T in represents the temperature of the liquid when it enters the reboiler.

4. According to the reboiler energy consumption analysis and control method described in claim 2, the specific calculation method of the steam energy consumption is: Among them, Q steam represents the steam energy consumption represents the mass flow rate of the steam generated by the reboiler, H steam represents the enthalpy value of the steam 5. According to the reboiler energy consumption analysis and control method described in claim 2, the specific calculation method of the heat loss is: Q loss = U surface · A surface · ΔT surface Among them, Q loss represents the heat loss, U surface represents the overall heat transfer coefficient of the reboiler surface, A surface represents the external surface area of the reboiler, ΔT surface represents the temperature difference between the reboiler surface and the environment.

6. According to the reboiler energy consumption analysis and control method described in claim 2, the specific calculation method of the auxiliary equipment energy consumption is: Among them, Q aux represents the energy consumption of auxiliary equipment, and represent the power of the circulation pump and the control system respectively, t pump and t control represent the running time of the circulation pump and the control system respectively.

7. The reboiler energy consumption analysis and control method according to claim 1, characterized in that The reboiler control module based on a PID neural network is specifically used for: Obtain the input quantities of each neuron in the input layer: x1(k) = r(k) x2(k) = y(k) Among them, x1(k) represents the input quantity of the first neuron in the input layer at the k sampling moment, r(k) represents the target steam mass flow rate output by the reboiler set at the k sampling moment, x2(k) represents the input quantity of the second neuron in the input layer at the k sampling moment, and y(k) represents the actual steam mass flow rate output by the reboiler at the k sampling moment; Calculate the input quantities of each neuron in the hidden layer: where n i (k) represents the input of the i-th neuron in the hidden layer at the k-th sampling moment, and w ij represents the connection weight between the j-th neuron in the input layer and the i-th neuron in the hidden layer, and x j (k) represents the input of the j-th neuron in the input layer at the k-th sampling moment; According to the input quantities of each neuron in the hidden layer, calculate the output quantities of each neuron in the hidden layer: u1(k) = n1(k) u2(k) = n2(k) + u2(k - 1) u3(k) = n3(k) - n3(k - 1) Among them, u1(k) represents the output of the proportional neuron in the hidden layer at the k-th sampling moment, n1(k) represents the input of the proportional neuron in the hidden layer at the k-th sampling moment, u2(k) represents the output of the integral neuron in the hidden layer at the k-th sampling moment, n2(k) represents the input of the integral neuron in the hidden layer at the k-th sampling moment, u2(k - 1) represents the output of the integral neuron in the hidden layer at the (k - 1)-th sampling moment, u3(k) represents the output of the differential neuron in the hidden layer at the k-th sampling moment, n3(k) represents the input of the differential neuron in the hidden layer at the k-th sampling moment, and n3(k - 1) represents the input of the differential neuron in the hidden layer at the (k - 1)-th sampling moment; Calculate the output of the output layer according to the outputs of the neurons in the hidden layer: Among them, u(k) represents the control signal output by the output layer at the k-th sampling moment, and w i represents the connection weight between the i-th neuron in the hidden layer and the output layer, and u i (k) represents the output of the i-th neuron in the hidden layer at the k-th sampling moment.

8. A reboiler energy consumption analysis and control system, characterized in that, Including: A memory and one or more processors; One or more applications are stored in the memory, and the one or more applications are adapted to be executed by the one or more processors to implement the reboiler energy consumption analysis and control method according to any one of claims 1 to 7.

Images