A nonlinear model predictive control method for low-temperature multi-effect distillation desalination system

Abstract

The invention discloses a nonlinear model predictive control method for a low-temperature multi-effect distillation seawater desalination system, which relates to the technical field of seawater desalination. The specific steps are: defining system parameters, deriving and establishing a dynamic model of the system, designing a nonlinear model predictive control controller, and designing a low-temperature multi-effect distillation seawater desalination system experiment. Considering the thermodynamic characteristics of the low-temperature multi-effect distillation seawater desalination system, a nonlinear MPC algorithm equation is established, and a NMPC module is used to calculate the power steam flow rate W. ms With the feed seawater flow rate W sw Two major types of control inputs are used to achieve automatic control of the maximum water production ratio. The method of the present invention can effectively overcome the uncertainty, nonlinearity and parallelism of the process, and conveniently handle various constraints in the controlled variables and manipulated variables in the process to achieve a control effect with good stability and strong robustness.

Description

Technical Field

[0001] The invention relates to the technical field of seawater desalination, and in particular to a nonlinear model predictive control method for a low-temperature multi-effect distillation seawater desalination system. Background Art

[0002] Low-temperature multi-effect distillation technology is a widely used thermal desalination technology in practical engineering. Its main principle is: after preheating the raw seawater through the condenser, the seawater is evaporated step by step through multiple evaporators in series, and the final fresh water is collected and discharged from the last evaporator. If a steam ejector is added, part of the steam generated by one of the effect chambers can be extracted as heating steam for the first effect evaporator, further reducing heat energy consumption. This technology is characterized by high distilled water quality, high thermal efficiency, and high water production ratio; but the system mechanism is complex, the parameters are highly coupled, and it is difficult to control.

[0003] At present, the control research of low-temperature multi-effect distillation seawater desalination system is still focused on improving the traditional PID. However, in actual applications, since the controlled process is often complex in mechanism, with characteristics such as high nonlinearity, time-varying uncertainty and pure lag, especially under the influence of factors such as noise and load disturbance, process parameters and even model structure will change with time and working environment. Traditional control methods often cannot meet the control requirements. Summary of the invention

[0004] The present invention provides a nonlinear model predictive control method for a low-temperature multi-effect distillation seawater desalination system, which solves the problem that the traditional PID control method proposed in the above background technology has the disadvantages of high nonlinearity, time-varying uncertainty and pure hysteresis, and under the influence of factors such as noise and load disturbance, process parameters and even model structure will change with time and working environment, and cannot meet control requirements.

[0005] The present invention provides the following technical solution: a nonlinear model predictive control method for a low-temperature multi-effect distillation seawater desalination system, wherein the low-temperature multi-effect distillation seawater desalination system comprises a condenser, an evaporator and an ejector, and is characterized in that the nonlinear model predictive control method for the low-temperature multi-effect distillation seawater desalination system comprises the following steps:

[0006] Step 1: Define system parameters;

[0007] The condenser parameters include the condenser volume V con , seawater density in the condenser tube ρ con , condenser heat transfer rate Q con , feed seawater flow rate W sw , specific heat of seawater at constant pressure C psw , raw water temperature T feed , sea water temperature T f, the heat energy of seawater in the condenser tube h con , the seawater temperature in the condenser tube is T con , condenser area A c , condenser product water level L d , product water leaving the condenser flow rate W d , product water density ρ d , the total output of product water D; the evaporator has a total of six effects, and the parameters of each effect evaporator include the brine water level L i , steam temperature of each effect T i , each effective brine salinity X i , each effect evaporator area A s,i , effective brine density ρ f,i , steam density of each effect ρ v,i , each effective brine temperature T b,i , the calorific value of brine in each effect and the calorific value of brine leaving each effect h b,i , the calorific value of steam in each effect and the calorific value of steam leaving each effect h v,i , each effect boiling point rise BPE i , steam height of each effect L v,i , feed water flow rate W for each effect feed,i , feed water heat energy h feed , the brine flow rate W leaving each effect b,i , power steam flow W ms , the steam flow rate W leaving each effect v,i , heat transfer rate of each effect Q E,i ; The ejector parameters include the power steam pressure P m , compression steam pressure P s , injection steam pressure P ev , the ejection steam temperature, that is, the steam temperature of the last effect evaporator T ev ;

[0008] Step 2: Design a dynamic model based on the nonlinear model predictive control method. The dynamic model includes the condenser tube bundle and inner dynamic model, the evaporator tube bundle dynamic model, the brine and steam dynamic model, and the ejector dynamic model. In each iteration of the predictive control algorithm, the dynamic model is established; the material balance, energy balance, and salinity balance equations of each part are established; after simplification, the dynamic model of the system regarding water level, temperature, and salinity is obtained;

[0009] Step 3: Design a nonlinear model predictive controller based on step 2; design a corresponding nonlinear model predictive controller;

[0010] Step 4. Design a low-temperature multi-effect distillation seawater desalination system experiment; set the system experimental site, seawater temperature, MPC solver sampling time, predictive control algorithm step range, and solver convergence time; if the solver does not converge, the solver will be reinitialized according to the current system parameters.

[0011] Preferably, the dynamic model in step 2 includes a condenser tube bundle and inner lining dynamic model, an evaporator tube bundle dynamic model, a brine and steam dynamic model, and an ejector dynamic model.

[0012] Preferably, the step 2 specifically includes the following steps:

[0013] Step 21. During each iteration of the predictive control algorithm, a dynamic model of each effect evaporator and condenser is established;

[0014] Step 22: Establish a dynamic model of the ejector and calculate the ejector injection ratio;

[0015] Step 2: Simplify the ejector equations and use linear equations to reduce the complexity of the nonlinear model to reduce the convergence time of the MPC solver.

[0016] Preferably, the nonlinear equation of each effective evaporator established in step 2 is:

[0017]

[0018]

[0019] The nonlinear equation of the condenser established is:

[0020]

[0021]

[0022] in:

[0023] The calculation formula of the ejector injection ratio established is:

[0024]

[0025] Preferably, PCF and TCF are both based on P m and P ev is the empirical formula of the parameters, where:

[0026] PCF = 3 × 10 -7 (P m ) 2 -0.0009(P m )+1.6101 (26);

[0027] TCF = 2 × 10 -8 (T ev ) 2 -0.0006(T ev )+1.0047 (27).

[0028] Preferably, the step three specifically includes the following steps:

[0029] Step 31: Establish and optimize the system cost function and determine the cost function formula as shown in formula (28):

[0030]

[0031] Step 32: Determine the control variables, sampling intervals and initial conditions of the nonlinear model predictive controller;

[0032] Step 3: Establish the constraints of the nonlinear model predictive controller.

[0033] Preferably, the power steam flow rate W is determined ms With seawater flow W sw is the control variable of the nonlinear model predictive controller; determine the sampling T of the nonlinear model predictive controller s The interval is 0.5s.

[0034] Preferably, the initial condition of the nonlinear model predictive controller is determined as:

[0035]

[0036]

[0037]

[0038] Preferably, the constraints of the nonlinear model predictive control are:

[0039] L i ∈N(29);

[0040] T i ∈[0,70](30);

[0041] ω∈[0,1]

[0042] (31).

[0043] Preferably, the system experiment site in step 4 is set to be outdoor, the season is summer, the seawater temperature range is 25°C-35°C, and the MPC solver sampling time is T s=0.5s, the predictive control algorithm step is N=30, and the control level is 2; the average convergence time of the solver is 400s, the frequency is 20Hz, and the sampling time provides a buffer time for the calculation error value.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] 1. The nonlinear model predictive control method of the low-temperature multi-effect distillation desalination system considers the thermodynamic characteristics of the low-temperature multi-effect distillation desalination system, establishes the nonlinear MPC algorithm equation, and the NMPC module is based on the power steam flow W ms With the feed seawater flow rate W sw Two major types of control inputs are used to achieve automatic control of the maximum water production ratio.

[0046] 2. The nonlinear model predictive control method of the low-temperature multi-effect distillation seawater desalination system can effectively overcome the uncertainty, nonlinearity and parallelism of the process, and conveniently handle various constraints in the controlled variables and manipulated variables in the process to achieve a control effect with good stability and strong robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 The effect of feed seawater temperature fluctuation on water production ratio;

[0048] Figure 2 The effect of feed seawater flow fluctuation on water production ratio;

[0049] Figure 3 The influence of power steam mass flow rate fluctuation on water production ratio;

[0050] Figure 4 The influence of the brine flow fluctuation of the last-effect evaporator on the water production ratio;

[0051] Figure 5 This is a flowchart of the nonlinear model predictive control method described in the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] The present invention provides a nonlinear model predictive control method for a low-temperature multi-effect distillation seawater desalination system. The low-temperature multi-effect distillation seawater desalination system comprises a condenser, an evaporator and an ejector. The nonlinear model predictive control method for the low-temperature multi-effect distillation seawater desalination system comprises the following steps:

[0054] Step 1: Define system parameters;

[0055] The condenser parameters include the condenser volume V con , seawater density in the condenser tube ρ con , condenser heat transfer rate Q con , feed seawater flow rate W sw , specific heat of seawater at constant pressure C psw , raw water temperature T feed , sea water temperature T f , the heat energy of seawater in the condenser tube h con , the seawater temperature in the condenser tube is T con , condenser area A c , condenser product water level L d , product water leaving the condenser flow rate W d , product water density ρ d , the total output of product water D; the evaporator has a total of six effects, and the parameters of each effect evaporator include the brine water level L i , steam temperature of each effect T i , each effective brine salinity X i , each effect evaporator area A s,i , effective brine density ρ f,i , steam density of each effect ρ v,i , each effective brine temperature T b,i , the calorific value of brine in each effect and the calorific value of brine leaving each effect h b,i , the calorific value of steam in each effect and the calorific value of steam leaving each effect h v,i , each effect boiling point rise BPE i , steam height of each effect L v,i , feed water flow rate W for each effect feed,i , feed water heat energy h feed , the brine flow rate W leaving each effect b,i , power steam flow W ms , the steam flow rate W leaving each effect v,i , heat transfer rate of each effect Q E,i ; The ejector parameters include the power steam pressure P m , compression steam pressure P s , injection steam pressure P ev , the ejection steam temperature, that is, the steam temperature of the last effect evaporator T ev ;

[0056] Step 2: Design a dynamic model based on the nonlinear model predictive control method. The dynamic model includes the condenser tube bundle and its inner dynamic model, the evaporator tube bundle dynamic model, the brine and steam dynamic model, and the ejector dynamic model. In each iteration of the predictive control algorithm, a dynamic model is established; the material balance, energy balance, and salinity balance equations of each part are established; after simplification, the system's dynamic model of water level, temperature, and salinity is obtained.

[0057] Step 2 specifically includes the following steps:

[0058] Step 21. Establish dynamic models of each effect evaporator and condenser; in each iteration of the predictive control algorithm, establish dynamic models as shown in equations (1) to (5), wherein equations (1) to (3) are nonlinear equations of each effect evaporator, and equations (4) to (5) are nonlinear equations of the condenser;

[0059]

[0060]

[0061]

[0062]

[0063]

[0064] In the above formula, D is the total product water output, which is the condensed water produced after the steam entering each effect evaporator and condenser is completely condensed;

[0065] The above formulas (1) to (3) are derived from formulas (7) to (12):

[0066] Z 1,i =a 1,i c 3,i -c 1,i a 3,i (7); Z 2,i =b 1,i c 3,i -c 1,i b 3,i (8); Z 3,i =d 1,i c 3,i -c 1,i d 3,i (9);

[0067] Z 4,i =a 2,i c 3,i -c 2,i a 3,i(10); Z 5,i =b 2,i c 3,i -c 2,i b 3,i (11); Z 6,i =d 2,i c 3,i -c 2,i d 3,i (12).

[0068] The above: a 1,i =A s,i ×ρ f,i -A s,i ×ρ v,i (13);

[0069]

[0070]

[0071]

[0072] a 2,i =A s,i ×h b,i ×ρ f,i -A s,i ×h v,i ×ρ v,i (17);

[0073]

[0074]

[0075]

[0076] a 3,i =A s,i ×X i ×ρ f,i (twenty one);

[0078]

[0079]

[0080]

[0081] Step 2: Establish the dynamic model of the ejector and calculate the ejection ratio; simplify the ejector equation and use linear equations to reduce the complexity of the nonlinear model to reduce the convergence time of the MPC solver; the simplified ejector ejection ratio linear equation is shown in formula (25):

[0082]

[0083] Where: ω is the ejector injection ratio, PCF and TCF are both based on P m and P ev is the empirical formula of the parameters.

[0084] PCF = 3 × 10 -7 (P m ) 2 -0.0009(P m )+1.6101 (26);

[0085] TCF = 2 × 10 -8 (T ev ) 2 -0.0006(T ev )+1.0047

[0086] (27).

[0087] Step three: Based on step two, design a nonlinear dynamic model and a corresponding nonlinear model predictive controller; the controller includes establishing and optimizing a system cost function; determining nonlinear system control variables, sampling intervals and initial conditions; and establishing constraints for nonlinear model predictive control.

[0088] Step three specifically includes the following steps:

[0089] Step 3: 1. Establish system cost function. The seawater desalination system generally takes maximizing the water production ratio as the control target, and the cost function is determined as shown in formula (28):

[0090] Step 3.2: Determine the power steam flow rate W ms With the feed seawater flow rate W sw is the control variable of the nonlinear model predictive controller, and the sampling T of the nonlinear model predictive controller s The interval is 0.5s.

[0091] The above-mentioned evaporator dynamic model, condenser dynamic model and steam ejector dynamic model together constitute the mathematical model of the entire system, and the control variables are mainly determined by selecting multiple input variables after the model is established and testing the influence of the variables on the system performance. Because the seawater desalination system needs to operate all year round and adjust production capacity according to water supply demand, the system's feed seawater temperature and feed seawater flow rate can be changed with external conditions, and the system's power steam is an external driving steam that can be artificially controlled. At the same time, the seawater desalination system discharges the concentrated brine out of the system in the last effect, and the flow rate can also be artificially controlled. Therefore, the feed seawater temperature T f , feed seawater flow rate W sw, power steam flow W ms And the brine flow rate W of the final evaporator b,n as a control variable.

[0092] Figures 1 to 4 The effect of the above four preparatory control variables on the system performance indicator water production ratio D when the increase of 5% at 7000s is shown. As can be seen from the figure, after the feed seawater temperature fluctuated at 7000s, the water production ratio decreased by 0.3% compared with the original; after the feed seawater flow rate fluctuated at 7000s, the water production ratio decreased by 1% compared with the original; after the power steam mass flow rate fluctuated at 7000s, the water production ratio increased by 0.4% compared with the original; after the final brine flow rate fluctuated at 7000s, the water production ratio increased by 0.2% compared with the original. It can be seen that the power steam flow rate W ms With the feed seawater flow rate W sw It is a key factor affecting the overall performance of the system. Therefore, the power steam flow rate W is determined ms With the feed seawater flow rate W sw Control variables for nonlinear model predictive controllers.

[0093] Step 3: Determine the initial conditions of the nonlinear model predictive controller. The initial conditions are shown in equations (29) to (31):

[0094]

[0095]

[0096]

[0097] Step 3: Determine the constraints of the nonlinear model predictive controller. Equations (32) to (34) are the constraints:

[0098] L i ∈N(32);

[0099] T i ∈[0,70](33);

[0100] ω∈[0,1]

[0101] (34).

[0102] Step 4. Design a low-temperature multi-effect distillation seawater desalination system experiment; set the system experimental site, seawater temperature, MPC solver sampling time, predictive control algorithm step range, and solver convergence time; if the solver does not converge, the solver will be reinitialized according to the current system parameters.

[0103] The system experiment site in step 4 is set to be outdoor, the season is summer, the seawater temperature range is 25℃-35℃, and the MPC solver sampling time is T s=0.5s, the predictive control algorithm step is N=30, and the control level is 2; the average convergence time of the solver is 400s, the frequency is 20Hz, and the sampling time provides a buffer time for the calculation error value.

[0104] It can be known from the above description that a nonlinear model predictive control (MPC) method designed by the present invention is applied to a low-temperature multi-effect distillation desalination system, a dynamic model of the system is derived and established, and a nonlinear model predictive control controller is designed. In practical applications, predictive control is implemented by an MPC algorithm. The specific implementation of MPC can adopt the MPC module in simulink, and only the relevant parameters in step three need to be input to achieve control. The control mechanism is simple and can meet actual needs. The method of the present invention can effectively overcome the uncertainty, nonlinearity and parallelism of the process, and conveniently handle various constraints in the controlled variables and manipulated variables in the process to achieve a control effect with good stability and strong robustness.

[0105] The contents not described in detail in this specification belong to the prior art known to professional and technical personnel in the field. Although the embodiments of the present invention have been shown and described, it is understood by ordinary technicians in the field that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the attached claims and their equivalents.

Claims

1. A nonlinear model predictive control method for a low-temperature multi-effect distillation seawater desalination system, wherein the low-temperature multi-effect distillation seawater desalination system comprises a condenser, an evaporator and an ejector, characterized in that: The nonlinear model predictive control method for the low-temperature multi-effect distillation seawater desalination system comprises the following steps: Step 1: Define system parameters; The condenser parameters include the condenser volume V con , seawater density in the condenser tube ρ con , condenser heat transfer rate Q con , feed seawater flow rate W sw , specific heat of seawater at constant pressure C psw , raw water temperature T feed , sea water temperature T f , the heat energy of seawater in the condenser tube h con , the seawater temperature in the condenser tube is T con , condenser area A c , condenser product water level L d , product water leaving the condenser flow rate W d , product water density ρ d , the total output of product water D; the evaporator has a total of six effects, and the parameters of each effect evaporator include the brine water level L i , steam temperature of each effect T i , each effective brine salinity X i , each effect evaporator area A s,i , effective brine density ρ f,i , each effective steam density ρ v,i , each effective brine temperature T b,i , the calorific value of brine in each effect and the calorific value of brine leaving each effect h b,i , the calorific value of steam in each effect and the calorific value of steam leaving each effect h v,i , each effect boiling point rise BPE i , steam height of each effect L v,i , feed water flow rate W for each effect feed,i , feed water heat energy h feed , the brine flow rate W leaving each effect b,i , power steam flow W ms , the steam flow rate W leaving each effect v,i , heat transfer rate of each effect Q E,i The ejector parameters include the power steam pressure P m , compression steam pressure P s , injection steam pressure P ev , the ejection steam temperature, that is, the steam temperature of the last effect evaporator T ev ; Step 2: Design a dynamic model based on the nonlinear model predictive control method. The dynamic model includes the condenser tube bundle and the inner dynamic model, the evaporator tube bundle, the brine and steam dynamic model and the ejector dynamic model. In each iteration of the predictive control algorithm, the dynamic model is established; the material balance, energy balance and salinity balance equations of each part are established; after simplification, the dynamic model of the system about water level, temperature and salinity is obtained; The nonlinear equations of each effect evaporator are established as follows: The nonlinear equation of the condenser established is: The calculation formula of the ejector injection ratio established is: Step 3: Design a nonlinear model predictive controller based on step 2; design a corresponding nonlinear model predictive controller; determine the power steam flow rate W ms With seawater flow W sw is the control variable of the nonlinear model predictive controller; determine the sampling T of the nonlinear model predictive controller s The interval is 0.5s; the initial conditions of the nonlinear model predictive controller are determined as: The constraints for establishing a nonlinear model predictive controller are: L i ∈N(29); T i ∈[0,70](30)； ω∈[0,1](31); Step 4. Design a low-temperature multi-effect distillation seawater desalination system experiment; set the system experimental site, seawater temperature, MPC solver sampling time, predictive control algorithm step range, and solver convergence time; if the solver does not converge, the solver will be reinitialized according to the current system parameters.

2. The nonlinear model predictive control method for a low-temperature multi-effect distillation seawater desalination system according to claim 1, characterized in that: The dynamic model in step 2 includes a condenser tube bundle and inner lining dynamic model, an evaporator tube bundle dynamic model, a brine and steam dynamic model, and an ejector dynamic model.

3. The nonlinear model predictive control method for a low-temperature multi-effect distillation seawater desalination system according to claim 2, characterized in that: The step 2 specifically includes the following steps: Step 21. During each iteration of the predictive control algorithm, a dynamic model of each effect evaporator and condenser is established; Step 22: Establish a dynamic model of the ejector and calculate the ejector injection ratio; Step 2: Simplify the ejector equations and use linear equations to reduce the complexity of the nonlinear model to reduce the convergence time of the MPC solver.

4. The nonlinear model predictive control method for a low-temperature multi-effect distillation seawater desalination system according to claim 1, characterized in that: PCF and TCF are both based on P m and P ev is the empirical formula of the parameters, where: PCF=3×10 -7 (P m ) 2 -0.0009(P m )+1.6101 (26); TCF=2×10 -8 (T ev ) 2 -0.0006(T ev )+1.0047 (27)。 5. The nonlinear model predictive control method for a low-temperature multi-effect distillation seawater desalination system according to claim 1, characterized in that:

6. The nonlinear model predictive control method for a low-temperature multi-effect distillation seawater desalination system according to claim 1, characterized in that: The step three specifically includes the following steps: Step 31: Establish and optimize the system cost function and determine the cost function formula as shown in formula (28): Step 32: Determine the control variables, sampling intervals and initial conditions of the nonlinear model predictive controller; Step 3: Establish the constraints of the nonlinear model predictive controller.

7. The nonlinear model predictive control method for a low-temperature multi-effect distillation seawater desalination system according to claim 1, characterized in that: The system experiment site in step 4 is set to be outdoor, the season is summer, the seawater temperature range is 25°C-35°C, and the MPC solver sampling time is T s =0.5s, the predictive control algorithm step is N=30, and the control level is 2; the average convergence time of the solver is 400s, the frequency is 20Hz, and the sampling time provides a buffer time for the calculation error value.

Images