Indirect energy recycling system and method for deamination tower

By introducing indirect energy reuse system and BO-LSTM intelligent control into the ammonia deamination tower, the problem of large steam consumption is solved, the significant reduction in energy consumption and the improvement of system stability is achieved, and the energy utilization of the ammonia deamination tower is improved Efficiency and economic benefits.

CN118987904AInactive Publication Date: 2024-11-22JIANGSU KECHENG ENERGY SAVING TECH CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411136952.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Excessive deamination towers use too much steam when increasing the temperature and pressure of absorbents lead to increased energy consumption and operational complexity, affecting the stable operation of the equipment.

Method used

The indirect energy reuse system of deamination tower is adopted, and a closed cycle is formed through the tower top heat collector and the tower bottom reboiler, and the energy reuse is used by the steam compressor and material separator, combining Bayesian optimization and long-term short-term memory network (BO-LSTM) intelligent control unit adjusts the temperature of the overhead heat collector and the bottom reboiler and the compression ratio of the vapor compressor in real time to optimize energy utilization efficiency.

Benefits of technology

The system steam consumption is significantly reduced, and it is expected to reduce the energy consumption of deamination towers by about 70-80%, improve the stability and response speed of system operation, improve energy utilization efficiency, and reduce energy waste and operating costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118987904A_ABST
    Figure CN118987904A_ABST
Patent Text Reader

Abstract

The invention discloses an indirect energy recycling system and method for a deamination tower. The indirect energy recycling system comprises the deamination tower, a tower top heat collector, a material separator I, a material separator II, a steam compressor, a tower bottom reboiler and a condenser, the tower top heat collector and the tower bottom reboiler are respectively arranged at the upstream and the downstream of the deamination tower and are both communicated with the deamination tower, the tower top heat collector is respectively connected with the condenser and the material separator I, and the material separator I is communicated with the steam compressor so as to feed working substances into the steam compressor; the steam compressor is communicated with the tower bottom reboiler, the tower bottom reboiler is connected with the pressure reducer, and the pressure reducer is communicated with the deamination tower. According to the indirect energy recycling system of the deamination tower, most energy of the tower bottom reboiler is provided by heat exchange of a compressed working medium, and a small amount of energy is provided by steam. The steam consumption of the system is greatly reduced, and about 70-80% of energy consumption of the deamination tower can be expected to be reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of deamination tower energy recycling, and in particular relates to a deamination tower indirect energy recycling system and a method thereof. Background Art

[0002] A deamination tower is a commonly used equipment in the industrial field. It can help us remove ammonia from some harmful gases, thereby achieving the purpose of purifying the air. A deamination tower is usually used to remove ammonia in industrial production. Its operating principle is to use chemical absorbents to absorb ammonia and convert it into other substances. In some cases, in order to increase the efficiency of deamination, it may be necessary to use steam to increase the temperature and pressure of the absorbent to promote the reaction.

[0003] However, there are also some disadvantages in using too much steam. First, too much steam will increase energy consumption and operating costs, which is not conducive to energy conservation and environmental protection and reducing production costs. Second, too much steam may make the operation of the deamination tower too complicated, prone to operating errors or failures, and affect the stable operation of the equipment.

[0004] Therefore, the present invention aims to provide a deamination tower indirect energy recovery system and method thereof to solve the problem of large steam loss in the prior art. Summary of the invention

[0005] The present invention aims to provide a deamination tower indirect energy recycling system and method thereof to solve the technical problems raised in the background technology.

[0006] To achieve the above object, the specific technical scheme of the present invention is as follows: comprising a deamination tower, a tower top heat collector, a material separator 1, a material separator 2, a steam compressor, a heat exchanger and a condenser;

[0007] The tower top heat collector and tower bottom reboiler are respectively arranged upstream and downstream of the deamination tower and are both connected to the deamination tower. The tower top heat collector is respectively connected to the condenser and the material separator one. The material separator one is connected to the steam compressor to feed the working material to the steam compressor. The steam compressor is connected to the tower bottom reboiler.

[0008] The ammonia-containing water vapor and the working material at the top of the deamination tower are partially heated by the top heat collector, and the working material of the deamination tower after partial heat removal is input into the material separator 1, and the ammonia-containing water vapor after heat removal enters the condenser.

[0009] The material separator 1 separates the working material into gas and liquid, and the separated gas phase enters the steam compressor, and the separated liquid phase is extracted.

[0010] The steam compressor compresses the working substance to increase the pressure and enthalpy value, and performs heat exchange in the tower bottom reboiler.

[0011] The reboiler at the bottom of the tower exchanges heat with part of the compressed working material and then takes heat again in the heat collector at the top of the tower to form a closed cycle. At the same time, part of the working material after heat exchange is input into the material separator 2.

[0012] The working material after heat exchange in the reboiler at the bottom of the material separator tower 2 undergoes gas-liquid separation, the separated liquid phase enters the deamination tower, and the separated gas phase is produced.

[0013] An indirect energy recycling system for a deamination tower, characterized in that:

[0014] Step 1: Data collection and preprocessing

[0015] In the deamination tower indirect energy recycling system, data collection and preprocessing are important links in optimizing system operation efficiency:

[0016] Data collection

[0017] The operating data of each part is collected in real time through the temperature sensors, pressure sensors and flow sensors in the system. These sensors are installed in key positions, such as the top and bottom of the deamination tower, the top heat collector, the bottom reboiler and the steam compressor. The collected raw data is expressed as:

[0018] D={(T1,P1,F1),(T2,P2,F2),…,(T n ,P n ,F n )}

[0019] D is the entire data set, T is temperature, P is pressure, F is flow, and n is the number of the data in the data set;

[0020] Data preprocessing

[0021] 1. Normalization

[0022] Normalization is to eliminate the influence of different dimensions and make the data comparable. The formula is as follows:

[0023]

[0024] x′ is the normalized data, x is the original data, and x min is the minimum value in the data set, x max is the maximum value in the data set;

[0025] After normalizing each collected data, we can get:

[0026]

[0027] T i ′ is the normalized temperature value, Ti is the original collected temperature value, T min is the minimum value of all temperature values ​​in the temperature data set, T max is the maximum value of all temperature values ​​in the temperature data set;

[0028]

[0029] P i ′ is the normalized pressure value, P i is the original collected pressure value, P min is the minimum value of all pressure values ​​in the pressure data set, P max is the maximum value of all pressure values ​​in the pressure data set;

[0030]

[0031] F i ′ is the normalized flow value, F i is the original collected flow value, F min is the minimum value of all flow values ​​in the flow data set, F max is the maximum value of all flow values ​​in the flow data set;

[0032] 2. Denoising

[0033] De-noising is used to eliminate noise in data and improve data accuracy;

[0034] Moving average method:

[0035] The sliding average method smoothes the data by calculating the average value of the data in the window. The formula is:

[0036]

[0037] is the i-th data point after smoothing, k is half of the window size, is the normalization coefficient,

[0038] For all x in the range from -k to k for j i+j The values ​​are summed, x i+j is the i+jth data point in the original data set;

[0039] Kalman filter:

[0040] Kalman filtering is a recursive filtering method suitable for dynamic systems;

[0041] Status prediction:

[0042]

[0043] is the predicted state vector, A is the state transfer matrix, is the updated state estimate at time k-1, B is the control input matrix, u k-1 is the control input vector at time k-1;

[0044]

[0045] is the prediction error covariance matrix, A is the state transfer matrix, P k-1 is the updated error covariance matrix at time k-1, A T is the transpose of the state transfer matrix, Q is the process noise covariance matrix;

[0046] Status Update:

[0047]

[0048] K k is the Kalman gain at time k, is the prediction error covariance matrix, H T is the transpose of the measurement matrix, H is the measurement matrix, is the prediction error covariance matrix, H T is the transpose of the observation matrix, R is the observation noise covariance matrix;

[0049]

[0050] is the updated state estimate at time k, is the predicted state vector, K k is the Kalman gain, z k is the observation vector at time k, H is the observation matrix, is the predicted state vector;

[0051]

[0052] P k is the updated error covariance matrix at time k, I is the identity matrix, K k is the Kalman gain, H is the observation matrix, is the prediction error covariance matrix;

[0053] Step 2: Model training

[0054] Model training uses the collected data to train an LSTM model, adjusts the hyperparameters of the LSTM model through Bayesian optimization, and further optimizes the Bayesian optimization process using the Parrot optimization algorithm. The following are the detailed steps:

[0055] LSTM model training

[0056] 1. Input gate:

[0057] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0058] i t is the activation value of the input gate at time t, σ is the Sigmoid activation function, W i is the weight matrix of the input gate, h t-1 is the hidden state of the previous moment, x t is the input at the current moment, b i is the bias term of the input gate;

[0059] 2. Forget Gate:

[0060] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0061] f t is the activation value of the forget gate at time t, σ is the Sigmoid activation function, W f is the weight matrix of the forget gate, h t-1 is the hidden state of the previous moment, x t is the input at the current moment, b f is the bias term of the forget gate;

[0062] 3. Candidate memory cell status:

[0063]

[0064] is the value of the candidate memory cell state at time t, tanh is the Tanh activation function, W C is the weight matrix of the candidate memory cell state, h t-1 is the hidden state of the previous moment, x t is the input at the current moment, b C is the bias term of the candidate memory cell state;

[0065] 4. Memory cell status update:

[0066]

[0067] C t is the value of the memory cell state at time t, ft is the activation value of the forget gate, C t-1 is the memory cell state at the previous moment, i t is the activation value of the input gate, is the value of the candidate memory cell state;

[0068] 5. Output Gate:

[0069] o t =σ(W o ·[h t-1 ,x t ]+b o )

[0070] o t is the activation value of the output gate at time t, σ is the Sigmoid activation function, W o is the weight matrix of the output gate, h t-1 is the hidden state of the previous moment, x t is the input at the current moment, b o is the bias term of the output gate;

[0071] 6.Hide status update:

[0072] h t =o t *tanh(C t )

[0073] h t is the value of the hidden state at time t, o t is the activation value of the output gate, tanh is the Tanh activation function, C t is the value of the memory cell state;

[0074] Bayesian optimization to tune hyperparameters

[0075] 1. Objective function definition:

[0076]

[0077] Loss(θ) is the objective function, θ is the hyperparameter set of the model, N is the number of samples, is the summation formula, y i is the true value, is the predicted value;

[0078] 2. Gaussian Process Regression Model:

[0079]

[0080] f(θ) is the predicted distribution of the objective function, is a Gaussian process, μ(θ) is the mean function, and k(θ,θ′) is the similarity between hyperparameters θ and θ′;

[0081] 3. Expectation Enhancement Criteria:

[0082]

[0083] EI(θ) is the expected improvement criterion, is the expectation operator, max(0,f(θ)-f(θ + )) is the maximum value function, f(θ) is the objective function, f(θ + ) is the objective function at the current optimal hyperparameter θ + The predicted value under ;

[0084] Parrot Optimization Algorithm Optimizes Bayesian Optimization Process

[0085] 1. Candidate solution update:

[0086] P new =P old +α(G best -P old )+β(P mean -P old )

[0087] P new is a new candidate solution, P old is the current solution, α is the control parameter used to adjust the weight of global exploration, G best is the global optimal solution, β is the control parameter used to adjust the weight of local development, P mean is the population average solution;

[0088] 2. Fitness function:

[0089] f(P)=Performance Metric(P)

[0090] f(P) is the fitness function, Performance Metric(P) is the performance metric;

[0091] Step 3: Real-time optimization control

[0092] In the deamination tower indirect energy recycling system, the intelligent control unit adjusts the system operating parameters in real time according to the trained BO-LSTM model to maximize energy utilization and minimize energy consumption. The following are the detailed steps:

[0093] Real-time monitoring

[0094] The intelligent control unit monitors the key parameters of the system in real time through sensors:

[0095] D(t)={T(t),P(t),F(t)}

[0096] D(t) is the system operation data set collected at time t, T(t) is the temperature data collected at time t, P(t) is the pressure data collected at time t, and F(t) is the flow data collected at time t;

[0097] Model predictions

[0098] Using the trained BO-LSTM model, we can make predictions based on the real-time collected data to get an estimate of the future state of the system:

[0099]

[0100] is the predicted value of the system state at time t, BO-LSTM is a prediction model based on Bayesian optimization and long short-term memory network, and D(t) is the system operation data collected at time t;

[0101] Optimization goals

[0102] The optimization objective function is defined to maximize energy utilization efficiency and minimize energy consumption. The objective function is expressed as:

[0103]

[0104] J is the optimization objective function value, max is the maximization operator, is the energy efficiency, min is the minimization operator, E loss For energy loss;

[0105] Control parameter adjustment

[0106] Based on the optimization objective function J, the system state predicted by the BO-LSTM model is used to adjust the system operating parameters θ to achieve the optimization goal:

[0107]

[0108] θ(t) is the set of system operating parameters adjusted at time t, argmax θ To express finding the parameter θ that maximizes the objective function J, is the system state based on the prediction at time t and the optimization objective function value calculated by the control parameter θ;

[0109] Execution Control

[0110] According to the optimized control parameter θ(t), the intelligent control unit issues control instructions to adjust the system operating parameters in real time:

[0111]

[0112] T top (t) is the adjusted temperature of the top heat exchanger at time t, is the optimized temperature parameter of the top heat exchanger at time t, T bottom (t) is the adjusted bottom reboiler temperature at time t, is the optimized bottom reboiler temperature parameter at time t, r comp (t) is the compression ratio of the steam compressor after adjustment at time t, is the compression ratio parameter of the steam compressor after optimization at time t

[0113] The deamination tower indirect energy recycling system of the present invention has the following advantages:

[0114] In this deamination tower indirect energy recycling system, most of the energy of the tower bottom reboiler is provided by heat exchange of compressed working fluid, and a small amount of energy is provided by steam. This greatly reduces the amount of steam used in the system and is expected to reduce the energy consumption of the deamination tower by about 70-80%.

[0115] The system introduces an intelligent control unit based on Bayesian optimization combined with long short-term memory network (BO-LSTM) and parrot optimization algorithm (POA). The intelligent control unit maximizes energy utilization efficiency and minimizes energy consumption loss by real-time monitoring and adjusting key parameters such as the operating temperature of the top heat collector and the bottom reboiler, the compression ratio of the steam compressor, etc. Specifically, the intelligent control unit uses the real-time data collected by the sensor, inputs the trained BO-LSTM model for prediction after data preprocessing, and adjusts the control parameters in real time by optimizing the objective function, realizing the adaptive optimization control of the system. This intelligent control method not only improves the stability and response speed of the system operation, but also significantly improves the energy utilization efficiency, further reducing energy waste and operating costs. In addition, the intelligent algorithm used in the system can continuously learn and optimize, adapt to the operating requirements under different working conditions, and ensure that the system is always in the best operating state. Combining the above advantages, the deamination tower indirect energy recycling system of the present invention has significant advantages in energy saving and consumption reduction and intelligent control, which greatly improves the economic and environmental benefits of the system.

[0116] The deamination tower indirect energy recycling system of the present invention realizes the maximization of energy utilization efficiency and the minimization of energy consumption by combining the advanced BO-LSTM (Bayesian optimization-long short-term memory network) algorithm. The system includes components such as a deamination tower, a tower top heat collector, a material separator 1, a material separator 2, a steam compressor, a tower bottom reboiler, a pressure reducer and a condenser. The system operating parameters, such as the operating temperature of the tower top heat collector and the tower bottom reboiler and the compression ratio of the steam compressor, are monitored and adjusted in real time through an intelligent control unit to ensure that the system can maintain optimal performance under various operating conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0117] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0118] Figure 1 It is the overall system principle block diagram of the present invention;

[0119] Figure 2 This is a flow chart for optimizing the energy recycling of the deamination tower of the present invention. DETAILED DESCRIPTION

[0120] In the following, only some exemplary embodiments are briefly described. As those skilled in the art will appreciate, the described embodiments may be modified in various ways without departing from the spirit or scope of the embodiments of the present invention. Therefore, the drawings and descriptions are considered to be exemplary and non-restrictive in nature.

[0121] In the description of the embodiments of the present invention, it should be understood that the terms "length", "vertical", "horizontal", "top", "bottom", etc. indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the embodiments of the present invention.

[0122] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0123] In the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, an electrical connection, or a communication; it can be a direct connection, or an indirect connection through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the embodiments of the present invention can be understood according to specific circumstances.

[0124] The disclosure below provides many different embodiments or examples to implement different structures of the embodiments of the present invention. In order to simplify the disclosure of the embodiments of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the embodiments of the present invention. In addition, the embodiments of the present invention can repeat reference numbers and / or reference letters in different examples, and this repetition is for the purpose of simplification and clarity, which itself does not indicate the relationship between the various embodiments and / or settings discussed.

[0125] In order to better understand the purpose, structure and function of the present invention, the deamination tower indirect energy recycling system of the present invention is further described in detail below in conjunction with the accompanying drawings.

[0126] like Figure 1 As shown, the deamination tower indirect energy recycling system of the present invention comprises a deamination tower, a tower top heat collector, a material separator 1, a material separator 2, a steam compressor, a tower bottom reboiler and a condenser; the tower top heat collector and the tower bottom reboiler are respectively arranged at the upstream and downstream of the deamination tower and are both connected to the deamination tower, the tower top heat collector is respectively connected to the condenser and the material separator 1, the material separator 1 is connected to the steam compressor to feed the working substance to the steam compressor, and the steam compressor is connected to the tower bottom reboiler. The ammonia-containing water vapor and the working substance at the top of the deamination tower are partially heated by the tower top heat collector, and the working substance of the deamination tower after partial heat extraction is input into the material separator 1, and the ammonia-containing water vapor after heat extraction enters the condenser. The material separator 1 performs gas-liquid separation on the working substance, and the separated gas phase enters the steam compressor, and the separated liquid phase is extracted. The steam compressor compresses the working substance to increase the pressure and enthalpy value, and performs heat exchange with the tower bottom reboiler. The reboiler at the bottom of the tower exchanges heat for part of the compressed working material and then takes heat again in the heat collector at the top of the tower, forming a closed cycle. At the same time, part of the working material after heat exchange is input into the material separator 2. The material separator 2 separates the working material after heat exchange in the reboiler at the bottom of the tower into gas and liquid. The separated liquid phase enters the deamination tower, and the separated gas phase is extracted.

[0127] The specific working principle of the present invention is as follows:

[0128] The ammonia-containing water vapor and the working substance at the top of the deamination tower are partially heated by the heat collector at the top of the tower. The working substance after heat extraction is input into the material separator 1. The material separator 1 performs gas-liquid separation on the working substance. The separated gas phase enters the steam compressor. The separated liquid phase is extracted. The steam compressor compresses the working substance to increase the pressure and enthalpy value, and the bottom reboiler is used for heat exchange. The bottom reboiler exchanges heat with part of the compressed working substance and then takes heat again in the heat collector at the top of the tower to form a closed cycle. At the same time, the material separator 2 performs gas-liquid separation on the working substance after heat exchange in the bottom reboiler. The separated liquid phase enters the deamination tower, and the separated gas phase is extracted. A small part of the energy of the bottom reboiler is provided by the steam in the deamination tower, and most of the energy is provided by heat exchange of the compressed working substance.

[0129] After the top of the tower is heated by the tower top heat collector, the ammonia-containing vapor enters the condenser for condensation, and part of it is refluxed and part of it is extracted.

[0130] It is to be understood that the present invention is described by some embodiments, and it is known to those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the present invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the scope of protection of the present invention.

[0131] Numerical case analysis:

[0132] Experimental setup

[0133] 1. Equipment parameters:

[0134] Deamination tower: processing capacity is 1000kg / h

[0135] Tower top heat collector: initial temperature is set to 150℃

[0136] Bottom reboiler: initial temperature is set to 180℃

[0137] Steam compressor: Initial compression ratio is set to 3:1

[0138] 2. Sensor data collection:

[0139] Temperature sensor: installed at the top heat collector and the bottom reboiler

[0140] Pressure sensor: installed at key locations of the deamination tower

[0141] Flow sensors: monitor the flow of materials and working fluids

[0142] 3. BO-LSTM model training:

[0143] Dataset: collected temperature, pressure and flow data within one week, totaling 10,000 data points

[0144] Hyperparameter optimization: Use Bayesian optimization and Parrot optimization algorithms to adjust the learning rate, number of hidden layer units and other hyperparameters of the LSTM model

[0145] Experimental Results

[0146] 1. Energy efficiency:

[0147] Before optimization: Energy efficiency is 65%

[0148] After optimization: Energy utilization efficiency increased to 85%

[0149] 2. Reduced energy consumption:

[0150] Before optimization: total system energy consumption is 200kWh

[0151] After optimization: total system energy consumption was reduced to 150kWh, a 25% reduction

[0152] 3. System stability:

[0153] Before optimization: The temperature and pressure fluctuated greatly during system operation, with fluctuations ranging between ±10℃ and ±0.5MPa.

[0154] After optimization: The system runs more stably, and the temperature and pressure fluctuation range is reduced to ±2℃ and ±0.1MPa.

[0155] Data analysis

[0156] Through real-time monitoring and optimization, the intelligent control unit dynamically adjusts the temperature of the top heat exchanger and the bottom reboiler and the compression ratio of the steam compressor according to the prediction results of the BO-LSTM model, so that the system energy utilization efficiency is increased from 65% to 85%, significantly improving the operating efficiency of the system. At the same time, the total energy consumption of the system is reduced from 200kWh to 150kWh, saving 25% of energy consumption. In addition, the optimized system operates more stably, the temperature and pressure fluctuation range is greatly reduced, and the reliability and safety of the system are improved.

Claims

1. The deamination tower indirect energy recycling system is characterized by: It includes a deamination tower, a tower top heat collector, a material separator, a steam compressor, a tower bottom reboiler, a pressure reducer and a condenser; The tower top heat collector and tower bottom reboiler are respectively arranged upstream and downstream of the deamination tower and are both connected to the deamination tower. The tower top heat collector is respectively connected to the condenser and the material separator one. The material separator one is connected to the steam compressor to feed the working material to the steam compressor. The steam compressor is connected to the tower bottom reboiler. The pressure reducer is arranged between the deamination tower and the tower bottom reboiler.

2. The deamination tower indirect energy recycling system according to claim 1, characterized in that: The ammonia-containing water vapor and the working material at the top of the deamination tower are partially heated by the top heat collector, and the working material of the deamination tower after partial heat removal is input into the material separator 1, and the ammonia-containing water vapor after heat removal enters the condenser.

3. The deamination tower indirect energy recycling system according to claim 1, characterized in that: The material separator 1 separates the working material into gas and liquid, and the separated gas phase enters the steam compressor, and the separated liquid phase is extracted.

4. The deamination tower indirect energy recycling system according to claim 1, characterized in that: The steam compressor compresses the working substance to increase the pressure and enthalpy value, and performs heat exchange in the tower bottom reboiler.

5. The deamination tower indirect energy recycling system according to claim 1, characterized in that: The reboiler at the bottom of the tower exchanges heat with part of the compressed working material and then takes heat again in the heat collector at the top of the tower to form a closed cycle. At the same time, part of the working material after heat exchange is input into the material separator 2.

6. The deamination tower indirect energy recycling system according to claim 1, characterized in that: Material separator 2. The working material after heat exchange in the bottom reboiler is separated into gas and liquid. The separated liquid phase enters the deamination tower, and the separated gas phase is produced.

7. The method of the deamination tower indirect energy recycling system according to claim 1, characterized in that: The following steps are involved: Step 1: Data collection and preprocessing: In the deamination tower indirect energy recycling system, data collection and preprocessing are important links in optimizing system operation efficiency; Data collection: The temperature sensors, pressure sensors and flow sensors set in the system collect the operating data of each part in real time. These sensors are installed in key positions, such as the top and bottom of the deamination tower, the top heat collector, the bottom reboiler and the steam compressor. The collected raw data is expressed as: D={(T1,P1,F1),(T2,P2,F2),…,(T n ,P n ,F n )} D is the entire data set, T is temperature, P is pressure, F is flow, and n is the number of the data in the data set; Data preprocessing 1. Normalization Normalization is to eliminate the influence of different dimensions and make the data comparable. The formula is as follows: x′ is the normalized data, x is the original data, and x min is the minimum value in the data set, x max is the maximum value in the data set; After normalizing each collected data, we can get: T i ′ is the normalized temperature value, T i is the original collected temperature value, T min is the minimum value of all temperature values ​​in the temperature data set, T max is the maximum value of all temperature values ​​in the temperature data set; P i ′ is the normalized pressure value, P i is the original collected pressure value, P min is the minimum value of all pressure values ​​in the pressure data set, P max is the maximum value of all pressure values ​​in the pressure data set; F i ′ is the normalized flow value, F i is the original collected flow value, F min is the minimum value of all flow values ​​in the flow data set, F max is the maximum value of all flow values ​​in the flow data set; 2. Denoising De-noising is used to eliminate noise in data and improve data accuracy; Moving average method: The sliding average method smoothes the data by calculating the average value of the data in the window. The formula is: is the i-th data point after smoothing, k is half of the window size, is the normalization coefficient, For all x in the range from -k to k for j i+j The values ​​are summed, x i+j is the i+jth data point in the original data set; Kalman filter: Kalman filtering is a recursive filtering method suitable for dynamic systems; Status prediction: is the predicted state vector, A is the state transfer matrix, is the updated state estimate at time k-1, B is the control input matrix, u k-1 is the control input vector at time k-1; is the prediction error covariance matrix, A is the state transfer matrix, P k-1 is the updated error covariance matrix at time k-1, A T is the transpose of the state transfer matrix, Q is the process noise covariance matrix; Status Update: K k is the Kalman gain at time k, is the prediction error covariance matrix, H T is the transpose of the measurement matrix, H is the measurement matrix, is the prediction error covariance matrix, H T is the transpose of the observation matrix, R is the observation noise covariance matrix; is the updated state estimate at time k, is the predicted state vector, K k is the Kalman gain, z k is the observation vector at time k, H is the observation matrix, is the predicted state vector; P k is the updated error covariance matrix at time k, I is the identity matrix, K k is the Kalman gain, H is the observation matrix, is the prediction error covariance matrix; Step 2: Model training uses the collected data to train a long short-term memory network LSTM model, adjusts the hyperparameters of the LSTM model through Bayesian optimization, and uses the Parrot optimization algorithm to further optimize the Bayesian optimization process. The following are the detailed steps: LSTM model training 1. Input gate: i t =σ(W i ·[h t-1 ,x t ]+b i ) i t is the activation value of the input gate at time t, σ is the Sigmoid activation function, W i is the weight matrix of the input gate, h t-1 is the hidden state of the previous moment, x t is the input at the current moment, b i is the bias term of the input gate; 2. Forget Gate: f t =σ(W f ·[h t-1 ,x t ]+b f ) f t is the activation value of the forget gate at time t, σ is the Sigmoid activation function, W f is the weight matrix of the forget gate, h t-1 is the hidden state of the previous moment, x t is the input at the current moment, b f is the bias term of the forget gate; 3. Candidate memory cell status: is the value of the candidate memory cell state at time t, tanh is the Tanh activation function, W C is the weight matrix of the candidate memory cell state, h t-1 is the hidden state at the previous moment, x t is the input at the current moment, b C is the bias term of the candidate memory cell state; 4. Memory cell status update: C t is the value of the memory cell state at time t, f t is the activation value of the forget gate, C t-1 is the memory cell state at the previous moment, i t is the activation value of the input gate, is the value of the candidate memory cell state; 5. Output Gate: the t =σ(W o ·[h t-1 ,x t ]+b o ) o t is the activation value of the output gate at time t, σ is the Sigmoid activation function, W o is the weight matrix of the output gate, h t-1 is the hidden state at the previous moment, x t is the input at the current moment, b o is the bias term of the output gate; 6.Hide status updates: h t =o t *tanh(C t ) h t is the value of the hidden state at time t, o t is the activation value of the output gate, tanh is the Tanh activation function, C t is the value of the memory cell state; Bayesian optimization to tune hyperparameters 1. Objective function definition: Loss(θ) is the objective function, θ is the hyperparameter set of the model, N is the number of samples, is the summation formula, y i is the true value, is the predicted value; 2. Gaussian Process Regression Model: f(θ) is the predicted distribution of the objective function, is a Gaussian process, μ(θ) is the mean function, and k(θ,θ′) is the similarity between hyperparameters θ and θ′; 3. Expectation Enhancement Criteria: EI(θ) is the expected improvement criterion, is the expectation operator, max(0,f(θ)-f(θ + )) is the maximum value function, f(θ) is the objective function, f(θ + ) is the objective function at the current optimal hyperparameter θ + The predicted value under ; Parrot Optimization Algorithm Optimizes Bayesian Optimization Process 1. Candidate solution update: P.S new JP old +α(G best -P old )+β(P mean -P old ) P new is a new candidate solution, P old is the current solution, α is the control parameter used to adjust the weight of global exploration, G best is the global optimal solution, β is the control parameter used to adjust the weight of local development, P mean is the population average solution; 2. Fitness function: f(P)=Performance Metric(P) f(P) is the fitness function, Performance Metric(P) is the performance indicator; Thus, the BO-LSTM model is built; Step 3 In the deamination tower indirect energy recycling system, the intelligent control unit adjusts the system operating parameters in real time according to the trained BO-LSTM model to maximize energy utilization and minimize energy consumption. The following are the detailed steps: Real-time monitoring The intelligent control unit monitors the key parameters of the system in real time through sensors: D(t)={T(t),P(t),F(t)} D(t) is the system operation data set collected at time t, T(t) is the temperature data collected at time t, P(t) is the pressure data collected at time t, and F(t) is the flow data collected at time t; Model predictions Using the trained BO-LSTM model, we can make predictions based on the real-time collected data to get an estimate of the future state of the system: is the predicted value of the system state at time t, BO-LSTM is a prediction model based on Bayesian optimization and long short-term memory network, and D(t) is the system operation data collected at time t; Optimization goals The optimization objective function is defined to maximize energy utilization efficiency and minimize energy consumption. The objective function is expressed as: J is the optimization objective function value, max is the maximization operator, is the energy efficiency, min is the minimization operator, E loss For energy loss; Control parameter adjustment Based on the optimization objective function J, the system state predicted by the BO-LSTM model is used to adjust the system operating parameters θ to achieve the optimization goal: θ(t) is the set of system operating parameters adjusted at time t, argmax θ To express finding the parameter θ that maximizes the objective function J, is the system state based on the prediction at time t and the optimization objective function value calculated by the control parameter θ; Execution control: Based on the optimized control parameter θ(t), the intelligent control unit issues control instructions to adjust the system operating parameters in real time: T top (t) is the adjusted temperature of the top heat exchanger at time t, is the optimized temperature parameter of the top heat exchanger at time t, T bottom (t) is the adjusted bottom reboiler temperature at time t, is the optimized bottom reboiler temperature parameter at time t, r comp (t) is the compression ratio of the steam compressor after adjustment at time t, is the compression ratio parameter of the steam compressor after optimization at time t.

Citation Information

Patent Citations

  • Deamination treatment system

    CN113880168A

  • Pressure soft sensor of excavator hydraulic system

    CN115563475A

  • System and method for optimisation

    WO2021163769A1