A desuperheating water injection valve control system for a waste heat power generation control system

Abstract

The application discloses a desuperheating water injection valve control system for a waste heat power generation control system, comprising an SMC control module, a saturation limiting module, a rate limiting module and a compensator module. The system introduces a fast adaptive sliding mode variable structure scheme based on differential evolution on the basis of the SMC control module output limited feedforward compensation sliding mode variable structure control, realizes fast optimization by using the population size and the limited iteration number of the evolution module to meet the control requirement through the contraction of the feasible solution space to the optimal value; the system plays the advantages of the sliding mode variable structure control, such as fast response speed and good robustness, and simultaneously imitates deep control, proposes a no-difference optimization control strategy to reduce the calculation amount in the longitudinal direction to ensure the real-time characteristics of the control, so as to realize the fast control of the desuperheating water injection valve of the waste heat power generation control system under the set precision.

Description

TECHNICAL FIELD

[0001] The application belongs to non-ferrous metal smelting optimization control technology, and particularly relates to a desuperheating water injection valve control system for a waste heat power generation control system. BACKGROUND

[0002] In the smelting process of non-ferrous metal, a large amount of high-temperature flue gas is generated, and the waste heat contained in the flue gas is large. If all or a small amount of the waste heat is utilized, energy is wasted greatly, and the environment is polluted. Therefore, before the flue gas is desulfurized to produce acid, a large amount of waste heat is utilized to generate power, and the generated power is used for smelting production, which has a significant effect on reducing production cost and improving economic benefit of enterprises, and can reduce heat emission of a sulfuric acid system, reduce heat pollution, and achieve the purpose of energy saving and emission reduction.

[0003] In the waste heat power generation control system, there are many factors affecting the temperature of superheated steam, and the factors interact with each other. In addition, the control requirements of superheated steam are different for different factors. Common factors include the inlet temperature of a superheater, steam flow, and desuperheating water quantity. In addition, due to the complex structure of the superheater, the long pipeline, and the large heating area, when a disturbance exists, the system has characteristics such as large time delay and large inertia. Therefore, it is difficult to realize stable control of the temperature of superheated steam. SUMMARY

[0004] The application provides a desuperheating water injection valve control system for a waste heat power generation control system. The waste heat power generation control system adjusts the injection quantity by controlling the opening degree of the desuperheating water injection valve to realize control of the temperature of superheated steam. The response speed and robustness are improved by optimizing the desuperheating water injection valve control system, and rapid and stable control of the waste heat power generation system is realized.

[0005] The application provides a desuperheating water injection valve control system for a waste heat power generation control system, which comprises an SMC control module, a saturation limiting module, a rate limiting module, and a compensator module.

[0006] The desuperheating water injection valve control system inputs a preset temperature expectation value r of superheated steam and a temperature error feedback value e. The difference between the preset temperature expectation value r of superheated steam and the temperature error feedback value e is input into the SMC control module. The input of the SMC control module further comprises the output of the compensator module. The output w of the SMC control module is a theoretical control value. The output w of the SMC control module is sequentially input into the saturation limiting module and the rate limiting module in series, to obtain a desuperheating water injection valve adjusting speed u, which is the output of the desuperheating water injection valve control system.

[0007] The difference ΔP1 between the output w of the SMC control module and the output v of the saturation limiting module, and the difference ΔP2 between the output v of the saturation limiting module and the desuperheating water injection valve adjusting speed u are weighted and summed, and the obtained result is input into the compensator module.

[0008] Specifically, the input ΔP of the compensator module is represented by the following formula:

[0009] ΔP = Q1ΔP1 + Q2ΔP2

[0010] Wherein, Q1 is the action threshold limit compensation coefficient; Q2 is the maximum action change rate limit compensation coefficient.

[0011] Specifically, the control process of the SMC control module is as follows:

[0012] T1 is the threshold of the selection function sat(S) function, T2 is the system steady state threshold, and S is the sliding film surface crossing coefficient; the selection function sat(S) is represented by the following formula:

[0013]

[0014] When |sat(S)|≤T1, the adaptive sliding mode variable structure optimization control is adopted to complete the control of the controlled system, and the cascade differential evolution control keeps tracking state; when T1<|sat(S)|≤T2, the control output is completed by the cascade differential evolution control, and the adaptive sliding mode variable structure optimization control is converted to tracking; when |sat(S)|>T2, the adaptive sliding mode variable structure optimization control is adopted again, and the cascade differential evolution control keeps tracking state.

[0015] Further, the adaptive sliding mode variable structure optimization control is as follows:

[0016] An adaptive sliding mode variable structure method based on differential evolution algorithm is adopted, and according to the state feedback and error value, adaptive control is realized by optimizing the sliding mode coefficient, so as to realize the optimization of the waste heat power generation control system; wherein the optimization number in the control process is obtained by the method of designing virtual sliding film surface to obtain optimal threshold value;

[0017] Specifically, for the optimization number in the control process, 2×N virtual sliding surfaces are designed with the sliding surface as the symmetric surface, and the optimal threshold value is only updated when the system state crosses the virtual sliding surface; the auxiliary sliding surface is represented by the following formula:

[0018]

[0019] Among them, Z0 is the initial value of the controlled state; T4 is the setting accuracy; N is the number of auxiliary synovial surfaces; T1 is the threshold of the selection function sat(S); the distance between the initial value of the state and the threshold is divided into four equal sections. In the first interval, the optimal threshold is calculated every time the system state changes by 20 times the accuracy. In the second interval, the optimal threshold is calculated every time the system state changes by 10 times the accuracy. In the third interval, the optimal threshold is calculated every time the system state changes by 5 times the accuracy. In the interval closest to the threshold, the optimal threshold is calculated every time the system state experiences a change in accuracy.

[0020] Furthermore, the cascade differential evolution control is specifically as follows:

[0021] For a single optimization process, the traditional differential evolution algorithm is used as the basic module unit. All module units point to the same fitness equation. Each unit is used as an independent layer, and the two layers are connected sequentially.

[0022] The execution of the cascade differential evolution algorithm starts from the first layer. The preset population size of the differential evolution algorithm in the first layer is 20 to 100, and the preset number of iterations is 200 to 1000. The obtained optimization results are weighted and used as the feasible solution space of the differential evolution module in the second layer to realize the zero-difference optimization control of the system.

[0023] Furthermore, the compensator module uses the following function for calculation:

[0024] f(x)=-bλ+a(Q1ΔP1+Q2ΔP2)

[0025] Among them, λ is the compensation factor; a is the control feedforward compensation coefficient; b is the error compensation coefficient.

[0026] The invention discloses a temperature-reducing water spray valve control system for a waste heat power generation control system, which improves the response speed and robustness by optimizing the control method, thereby realizing rapid and stable control of the waste heat power generation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a schematic structural diagram of a temperature-reducing water spray valve control system for a waste heat power generation control system according to the present invention;

[0028] Figure 2 This is a simulation comparison diagram of the synovial variable structure optimization control of the present invention. DETAILED DESCRIPTION

[0029] The present invention provides a temperature reduction water spray valve control system for a waste heat power generation control system, comprising an SMC control module, a saturation limiter module, a rate limiter module and a compensator module;

[0030] The input of the control system of the desuperheating water injection valve is a preset superheated steam temperature expectation value r and a temperature error feedback value e; the difference between the preset superheated steam temperature expectation value r and the temperature error feedback value e is input into an SMC control module, and the input of the SMC control module further includes the output of a compensator module; the output w of the SMC control module is a theoretical control value, and the output w of the SMC control module is sequentially input into a saturation limiting module and a rate limiting module in series to obtain a desuperheating water injection valve adjustment rate u as the output of the control system of the desuperheating water injection valve.

[0031] The difference ΔP1 between the output w of the SMC control module and the output v of the saturation limiting module and the difference ΔP2 between the output v of the saturation limiting module and the desuperheating water injection valve adjustment rate u are weighted and summed to obtain a result input into the compensator module.

[0032] Specifically, the input ΔP of the compensator module is represented by the following formula:

[0033] ΔP = Q1ΔP1 + Q2ΔP2

[0034] Wherein, Q1 is an action threshold limit compensation coefficient; Q2 is a maximum action change rate limit compensation coefficient.

[0035] Specifically, the control process of the SMC control module is specifically:

[0036] T1 is a threshold value of a selection function sat(S), T2 is a system steady state threshold value, and S is a sliding film surface crossing coefficient; the selection function sat(S) is represented by the following formula:

[0037]

[0038] When |sat(S)|≤T1, the system state is far from the sliding film surface, the adaptive sliding mode variable structure optimization control is adopted to complete the control of the controlled system, and the cascade differential evolution control keeps tracking state; when T1<|sat(S)|≤T2, the control output is completed by the cascade differential evolution control, and the adaptive sliding mode variable structure optimization control is switched to tracking, so as to ensure the continuity of the control output and realize disturbance-free switching; when |sat(S)|>T2, the cascade differential evolution control cannot stably implement the control, and the adaptive sliding mode variable structure optimization control is adopted again to make the controlled system quickly return to the system steady state threshold value T2.

[0039] Further, the adaptive sliding mode variable structure optimization control is specifically:

[0040] An adaptive sliding mode variable structure method based on a differential evolution algorithm is adopted, the adaptive control is realized by optimizing the sliding mode coefficient according to the state feedback and the error value, so as to realize the optimization of the waste heat power generation control system; wherein the optimization number in the control process is obtained by using the method of designing a virtual sliding film surface to obtain an optimal threshold value.

[0041] Specifically, for the number of optimization in the control process, 2xN virtual sliding mode surfaces are designed, which are symmetric to the sliding mode surfaces, and the optimal threshold is only updated when the system state crosses the virtual sliding mode surface, and the auxiliary sliding mode surface is expressed by the following formula:

[0042]

[0043] Wherein, Z0 is the initial value of the controlled state; T4 is the set precision; N is the number of auxiliary sliding mode surfaces; T1 is the threshold value of the selection function sat(S) function; the distance between the state initial value and the threshold value is divided into four sections, the optimal threshold value is calculated once every 20 times of precision change of the system state in the first section, the optimal threshold value is calculated once every 10 times of precision change of the system state in the second section, the optimal threshold value is calculated once every 5 times of precision change of the system state in the third section, and the optimal threshold value is calculated once every time the system state in the section closest to the threshold value experiences a precision change.

[0044] Further, the cascade differential evolution control is specifically:

[0045] For a single optimization process, the traditional differential evolution algorithm is used as a basic module unit, all module units are directed to the same fitness equation, and each unit is used as an independent layer, and two layers are sequentially connected.

[0046] The cascade differential evolution algorithm is executed from the first layer, the preset population size of the differential evolution algorithm of the first layer is 20-100, the preset iteration number is 200-1000, the obtained optimization result is weighted, and the differential evolution module of the second layer is used as the feasible solution space, so that the system is controlled without error optimization.

[0047] Further, the compensator module is calculated by using the following function:

[0048] f(x)=-bλ+a(Q1ΔP1+Q2ΔP2)

[0049] Wherein, λ is a compensation factor; a is a control feedforward compensation coefficient; and b is an error compensation coefficient.

[0050] The method of the application is further described below in combination with an embodiment:

[0051] The action speed of the temperature reducing water injection valve adjusting valve of the waste heat power generation control system is usually uniform after accepting the instruction, and the characteristic curve thereof is approximately proportional, and the change rate limit parameter is set to 1, that is, the change rate of the control quantity output is strictly limited to the interval [-1, 1]. Under the limitation condition, the compensation weighting coefficients Q1 and Q2 are set to 1 through multiple simulation debugging; the threshold of the sat(s) function is set to 0.3, the number of individuals of the differential evolution optimization is set to 60, and the iteration number is 600. The simulation curve finally obtained by using the temperature reducing water injection valve control system of the application is shown in the FSMC curve in Figure 2 .

[0052] The waste heat power generation system is simulated by using the conventional control scheme PID control, and the simulation curve is shown in the PID curve in Figure 2 .

[0053] The waste heat power generation system is optimized and controlled by using only the self-adaptive sliding mode variable structure method in the temperature reducing water injection valve control system of the application, and the single optimization is not optimized by introducing the cascade differential evolution strategy, and the simulation curve obtained is shown in the FSMC curve in Figure 2 .

[0054] Figure 2 The three schemes in the waste heat power generation control system all realize stable operation, and all have small overshoot; the FSMC belongs to the difference control, the static error of the simulation result is 0.0018, and the index completely meets the industrial operation demand standard. The traditional PID enters the steady state at about 140s, the FSMC scheme enters the steady state at about 60s, which is better than the conventional control, the rising time and the regulation time of the DSMC scheme are the shortest, and the time consumption is about 50s; and the overshoot of the control scheme introducing the sliding mode variable structure algorithm is less than 1%, which has a significant advantage over the PID control.

Claims

1. A temperature-reducing water spray valve control system for a waste heat power generation control system, characterized in that: Including SMC control module, saturation limit module, rate limit module and compensator module; The inputs to the attemperation spray valve control system are the preset desired superheated steam temperature r and the temperature error feedback value e. The difference between the preset desired superheated steam temperature r and the temperature error feedback value e is input into the SMC control module, which also includes the output of the compensator module. The output w of the SMC control module is the theoretical control variable. The output w of the SMC control module is sequentially passed through the saturation limiter module and the rate limiter module in series to obtain the attemperation spray valve adjustment rate u, which serves as the output of the attemperation spray valve control system. The difference ΔP1 between the output w of the SMC control module and the output v of the saturation limiter module, and the difference ΔP2 between the output v of the saturation limiter module and the adjustment rate u of the desuperheating spray valve are weighted and summed, and the result is input into the compensator module; The control process of the SMC control module is as follows: T1 is the threshold of the selection function sat(S), T2 is the system steady-state threshold, and S is the sliding surface crossing coefficient. The selection function sat(S) is expressed using the following formula: When |sat(S)| ≤T1, the controlled system is controlled by adaptive sliding mode variable structure optimization control, and the cascade differential evolution control maintains the tracking state; when T1<|sat(S)|≤T2, the control output is completed by cascade differential evolution control, and the adaptive sliding mode variable structure optimization control switches to tracking; when |sat(S)|>T2, the system again adopts adaptive sliding mode variable structure optimization control, and the cascade differential evolution control maintains the tracking state.

2. The attemperation water spray valve control system for waste heat power generation control system according to claim 1, characterized in that: The input ΔP of the compensator module is expressed using the following formula: Among them, Q1 is the action threshold limit compensation coefficient; Q2 is the maximum action change rate limit compensation coefficient.

3. The attemperation water spray valve control system for waste heat power generation control system according to claim 1, characterized in that: The adaptive sliding mode variable structure optimization control is specifically as follows: An adaptive sliding mode variable structure method based on differential evolution algorithm is adopted to realize adaptive control by optimizing the sliding mode coefficient according to state feedback and error value. The number of optimization times in the control process is obtained by designing a virtual sliding surface to obtain the optimal threshold.

4. The attemperation water spray valve control system for waste heat power generation control system according to claim 3, characterized in that: For the number of optimization searches during the control process of the adaptive sliding mode variable structure optimization control, 2×N virtual sliding surfaces with the sliding surface as the symmetric surface are designed. The optimal threshold is updated only when the system state crosses the virtual sliding surface. The virtual sliding surface is expressed by the following formula: in, is the initial value of the controlled state; T4 is the setting accuracy; N is the number of virtual sliding surfaces; To select the threshold of the function sat(S); divide the distance between the initial state value and the threshold into four equal segments. In the first interval, the optimal threshold is calculated every time the system state changes by 20 times the accuracy. In the second interval, the optimal threshold is calculated every time the system state changes by 10 times the accuracy. In the third interval, the optimal threshold is calculated every time the system state changes by 5 times the accuracy. In the interval closest to the threshold, the optimal threshold is calculated every time the system state experiences a change in accuracy.

5. The attemperation water spray valve control system for waste heat power generation control system according to claim 1, characterized in that: The cascade differential evolution control is specifically as follows: For a single optimization process, the traditional differential evolution algorithm is used as the basic module unit. All module units point to the same fitness equation. Each unit is used as an independent layer, and the two layers are connected sequentially. The execution of the cascade differential evolution algorithm starts from the first layer. The preset population size of the differential evolution algorithm in the first layer is 20-100, and the preset number of iterations is 200-1000. The obtained optimization results are weighted and used as the feasible solution space of the differential evolution module in the second layer to realize the zero-difference optimization control of the system.

6. The attemperation water spray valve control system for waste heat power generation control system according to claim 2, characterized in that: The compensator module uses the following function for calculation: in, is the compensation factor; a is the control feedforward compensation coefficient; b is the error compensation coefficient.

Images