Design and Temperature Modeling Simulation of the Full Closed Hot Air Circulation Tobacco Bulk Curing Barn

Abstract

For now, the open humidification method is applied in the tobacco bulk curing barn, which has some disadvantages, such as the loss of the oil content and aroma components of the tobacco leaves and the waste heat loss of the exhaust air flow. In this context, a tobacco bulk curing barn with totally closed hot air circulation is designed to perfect the curing quality of tobacco and avoid the loss of residual heat in the bulk curing barn. Meanwhile, due to the balance and symmetry of input and output of the curing barn temperature, according to the law of conservation of energy, a mathematical model of the temperature control system of the closed hot air circulation tobacco bulk curing barn is established, and the temperature transfer function of the system is obtained. On this basis, 10 algorithms are used to optimize the full closed hot air circulation tobacco bulk curing barn temperature control system PID parameters. The result of the sobol sequence seeker optimization algorithm (SSOA) is better than the other algorithms. So, the PID control strategy based on the SSOA is used to simulate and experiment the temperature control system of tobacco bulk curing barn. The simulation and experimental results show that for the tobacco bulk curing barn temperature control system, the sobol sequence seeker optimization algorithm PID control has better dynamic characteristics compared with fuzzy PID control, and the temperature control system of tobacco bulk curing barn has fast adjustment and small overshoot. Therefore, the new baking barn with proper PID parameters can improve the tobacco’s curing quality and save energy by reducing the residual heat.

1. Introduction

In 1960, the first large-scale bulk baking operation began in North Carolina [1]. During the past decade, the tobacco bulk curing barn have been widely promoted and applied. The traditional tobacco baking barn with natural ventilation has been replaced by the tobacco bulk curing barn with forced ventilation. The bulk tobacco curing barn add the hot air circulation system and the automatic control system, saving the energy costs and freeing the manpower. Due to the application of forced ventilation technology, in recent years, a large number of studies have been made on the structure and different air flow directions of the tobacco curing barn, especially the common tobacco curing rooms with the air ascending type [2,3] and the air descending type [4].

Generally, the tobacco curing can be divided into three stages: yellowing, dry leaves and dry stems [5]. The variation of flue baking process is mainly the result of water evaporation and enzyme reaction [6]. And the distribution of air flow, temperature and relative humidity in the tobacco baking room also affected the quality of tobacco baking.

The tobacco bulk curing barn adopts forced ventilation and hot air circulation, which is characterized by large quantity of loaded tobacco leaves, labor-saving, time-saving and easy automatic control of tobacco leaves baking. However, most of the current tobacco bulk curing barn adopt open humidity drainage, and the moisture is discharged directly outside the baking room. On the one hand, due to the open way of moisture drainage in the baking room, the loss of residual heat from the exhaust airflow is caused, so the energy utilization rate of the heating system in the tobacco bulk curing barn is generally low. On the other hand, due to the open humidity drainage of the baking room, the oil content and aroma of the tobacco leaves in the baking room are reduced. To reduce the energy consumption of the tobacco bulk curing barn and improve the quality of the tobacco leaf baking, some relevant scholars have conducted many studies on the temperature and humidity control of the tobacco bulk curing barn with the open moisture drainage. The artificial neural network [7] was used to study the distribution of temperature and relative humidity in tobacco drying process. Zhang et al. [8] simulated the temperature field in the barn in the drying process of tobacco based on the finite volume method. Wei et al. [9] studied the temperature distribution of tobacco baking room through experiments and numerical simulation. Bao et al. [10] conducted numerical studies on temperature, humidity and velocity fields in the tobacco baking room. To reduce this dependency and achieve high quality the curing process, Bai et al. [11] proposed a new structure of the tobacco bulk baking room based on the heat pump and conducted numerical simulation with CFD to conduct the temperature and humidity of the baking room. Portia et al. [12] proposed a biomass fired flue curing tobacco barn design for small-medium scale farmers in Zimbabwe with a clear objective to control the temperature and humidity. Condori [13] developed a control system based on digital image processing for the tobacco curing process in a bulk-curing barn. He et al. [14] designed an intelligent biomass fuel burner as an alternative to coal-fired heating for tobacco curing. Wu et al. [15] used a deep learning-based method to model of the bulk tobacco flue-curing process. Zhao et al. [16] measured the optimum humidity for air-curing of cigar tobacco leaves during the browning period. Wei et al. [17] explored the practicability of on-line monitoring of moisture, starch, protein, and soluble sugars for tobacco leaves by near-infrared spectroscopy and deep transfer learning. However, the studies of oil content and aroma retention of the tobacco leaves, the totally closed hot air circulation centered on tobacco baking room were rarely considered.

In this paper, a kind of dense baking room with full closed hot air circulation was proposed, to avoid the defect of open humidity drainage and further improve the energy utilization rate and quality of the tobacco leaves in the dense baking room. According to the energy conservation theorem, this kind of full-closed hot air circulation dense oven temperature control system is modeled, and PID (proportion integrals differential) control strategy based on sobol sequence seeker optimization algorithm is adopted to simulate and experiment the temperature control system of dense oven.

The major contributions of this paper are summed up as follows:

• A new kind of dense baking room with full closed hot air circulation was proposed. The new dense baking room can improve the energy utilization rate and quality of the tobacco leaves.
• According to the energy conservation theorem, the temperature control system of the new dense baking room is modeled.
• The PID control strategy based on sobol sequence seeker optimization algorithm is adopted to simulate step response curve of the temperature control system of new oven.
• The PID control strategy based on sobol sequence seeker optimization algorithm is adopted to actual experiment the temperature control system of new oven.

The rest of the article structure is as follows. Part 2 design the full closed hot air circulation tobacco bulk curing barn. Section 3 establishes of dynamic temperature model for the new curing barn. Section 4 shows the simulation results of temperature control system for the new curing barn. Section 5 shows the actual experiment results of temperature control system for the new curing barn. At last, Section 6 gives some conclusions.

2. Design and Working Principle of Full Closed Hot Air Circulation Tobacco Bulk Curing Barn

At present, the tobacco bulk curing barn adopts the open mode of moisture removal. During the moisture removal stage of the tobacco curing, the moisture removal air is directly discharged to the outdoor tobacco curing room. To avoid the defect of the current open moisture drainage and further improve the energy utilization rate and the quality of tobacco leaves in the tobacco bulk curing barn, this paper adopts the fully closed hot air circulation. The construction of the full closed hot-air circulation tobacco bulk curing barn is shown in the Figure 1. Compared with the open type of tobacco bulk curing barn, the fully closed hot air circulation tobacco bulk curing barn eliminates the wet outlet and adds the dehumidification equipment. Since the dew-point temperature of the dehumidified air flow is higher than 40 °C during the dense baking process of tobacco leaves, the dehumidification equipment can adopt an inter-wall heat exchanger to dehumidify the hot and humid air flow in the air return with outdoor air as the cold source.

When tobacco leaves are baking intensively, the hot air heated by the hot air stove is sent to the tobacco leaves baking room from the hot air inlet. In the tobacco baking room, the hot air is fed into the tobacco for heat and humidity exchange, and the hot air increases the humidity of the hot air and lowers the temperature, which then flows through the dehumidification equipment from the return air outlet. After dehumidification in the dehumidification equipment, the hot air with low temperature and high humidity returns to the hot air stove for reheating and is sent to the tobacco baking room, and so on.

The paper designed this kind of full closed hot air circulation tobacco bulk curing barn, the specific structure and components of the bulk curing barn are shown in the Figure 2. The full closed hot air circulation tobacco bulk curing barn is composed of the tobacco leaves baking room 1, the heating chamber 2, the hot air inlet 3, the hot air circulation fan in the barn 4, the hot blast heater 5, the return air inlet 13, the temperature sensor 14, the humidity sensor 15, and the controller 16. In the full closed hot air circulation tobacco bulk curing barn added the wall-mounted air-cooled dehumidifier 6, the dehumidifying air outlet duct 7, the fresh air outlet duct 8, the dehumidifying air blower 9, the fresh air inlet duct 10, the drainpipe 11, the fresh air fan 12, the fresh air fan inlet air blower joint 27, the fresh air fan outlet joint 28, the dehumidifying air fan inlet joint 29, the dehumidifying air fan outlet joint 30, and at the same time cancel the dehumidifying air outlet 31. The wall-mounted air-cooled dehumidifier 6 is composed of the insulation shell 17, the dehumidifying air outlet 18, the interwall heat exchange module 19, the fresh air inlet 20, the support 21, the water seal 22, the drainage outlet 23, the condensate water collector plate 24, the fresh air outlet 25, and the dehumidifying air inlet 26. The wall-mounted air-cooled dehumidifier 6 is an interwall air cooled dehumidifier with automatic control. The structural connection diagram of the interwall air cooled dehumidifier is shown in the Figure 3.

The specific working principle of the fully closed hot air cycle intensive oven: when it is necessary to dehumidify, the hot and humid airflow returned from the hot air return air inlet 13 below the tobacco leaves baking room 1. A part of it enters from the dehumidified air outlet 18 and flows through the interwall air-cooled dehumidifier 6 under the dehumidifying air blower 9. The outdoor fresh air enters from fresh air inlet pipe 10 and flows through the wall-mounted air-cooled dehumidifier 6 under the traction of fresh air fan 12. In the wall heat exchange module 19 of the wall-mounted air-cooled dehumidifier 6, the hot and humid air flow and fresh air from the outside are cooled and condensed. The condensed water is collected by the condensate water collector plate 24 and discharged to the outside by the drainpipe 11; The desiccated hot and humid air flow enters the heating chamber 2, and then enters the tobacco leaves baking room 1 after being heated by the hot blast heater 5, to reduce the waste heat loss of the moisture flow in the conventional intensive curing room and the loss of beneficial components such as aroma components of the tobacco leaves.

In this design, the aroma constituents and other gases are fixed in the intensive curing room. The aroma constituents and other gases are beneficial to perfecting the curing quality of tobacco leaves. The current intensive curing room can avoid the loss of residual heat, save energy and reduce emission.

3. Establishment of Dynamic Temperature Model for Full Closed Hot Air Circulation Tobacco Bulk Curing Barn

According to the specifications of the current standard tobacco bulk curing barn, the structural parameters of the heating chamber should be 2 m (long) × 2.7 m (wide) × 3.5 m (high), and the structural parameters of the tobacco leaves baking room should be 8 m (long) × 2.7 m (wide) × 3.5 m (high).

To simplify the calculation appropriately, the following assumptions and simplifications were made when establishing the temperature mathematical model of tobacco bulk curing barn: (1) Because the temperature difference between vertical and horizontal directions was very small in the tobacco leaves baking room, it was assumed that the temperature distribution in the tobacco leaves baking room was even. (2) The thermal efficiency of hot blast heater in the tobacco bulk curing barn is approximately unchanged. (3) The heat transfer coefficient between the tobacco baking room, the outside world is approximately constant and does not change with the change of the outside climate. Based on this assumption and simplification, a mathematical model of tobacco curing room temperature was established. Whether the model is correct or not will be further proved by specific baking experiments in Section 5 of this paper.

3.1. Dynamic Temperature Model of Heating Chamber in the Tobacco Bulk Curing Barn

Due to the balance and symmetry of input and output of the curing barn temperature, based on the law of energy conservation [18,19], the energy balance equation of the heating chamber is:

$$\Delta Q=\eta Q_{coal}-Q_{dehumidification}-Q_{heat dissipation}$$

(1)

In the Formula (1): ΔQ—the hot blast heater in a unit time internal energy supply heating chamber heat; η—the hot blast heater thermal efficiency; Q_coal—the amount of heat generated by the fuel in the hot blast heater per unit time; Q_{heat dissipation}—the amount of heat lost in a heating chamber per unit of time; Q_{dehumidification}—the amount of heat lost by dehumidifier per unit time.

The heat generated by the fuel in the hot blast heater per unit time was calculated according to the amount of coal consumed in the dense baking process of tobacco leaves, and the thermal efficiency of the hot blast heater was determined. The heat loss of the heating chamber in a unit time is determined by the density, the heat transfer temperature difference and the heat transfer coefficient of the envelope. The heat loss of the dehumidifier per unit time is determined according to the enthalpy difference before and after the dehumidification of the dehumidified air passing through the dehumidifier.

$${\rho }_1{C}_{\mathrm{p}1}{V}_1\Delta T=\eta {\displaystyle {\int }_0^{\mathsf{\tau }}\frac{U_1}{U_0}}Q\mathrm{d}t-{\displaystyle {\int }_0^{\mathsf{\tau }}{A}_1{h}_1}\Delta T\mathrm{d}t-{\displaystyle {\int }_0^{\mathsf{\tau }}{A}_2{h}_2}\Delta T\mathrm{d}t$$

(2)

The mathematical model Formula (3) of heating chamber temperature is obtained by Laplace transform of the Equation (2).

$$G_1(s)=\frac{\Delta T(s)}{U_1(s)}=\eta \frac{Q}{U_0}\cdot \frac{1}{{\rho }_1{C}_{\mathrm{p}1}{V}_{1}s+A_1{h}_1+A_2{h}_2}$$

(3)

In the Formula (3): ρ₁—the heating chamber indoor air density, 1.029 kg/m³; C_p1—the specific heat capacity at constant pressure of air in the heating chamber, 1.009 × 10³ J/(kg·K); V₁—the volume of heating chamber; ΔT—the difference in temperature of heating chamber between indoor air and outdoor air, K; η—the hot blast stove thermal efficiency, take 75%, U₁—the circulation fan in the 0-τ time type hot blast stove air quantity, m³. U₀—the air quantity of coal burning, 7.1 m³/kg; Q—the calorific value of coal, 29,271.2 × 10³ J/kg; A₁—the heat exchanger area of dehumidification equipment, 0.5 m (long), 0.18 m (wide) and 0.25 m (high); h₁—the heat transfer coefficient between heat exchanger of dehumidification equipment and the outside world, 48.93 w/(m·K); A₂—the inner wall area of heating chamber; h₂—the heat transfer coefficient between the enclosure structure of heating chamber, the outside world is 0.79 w/(m·K).

The mathematical model Formula (4) of dynamic transfer function of the heating chamber temperature was obtained by substituting the parameters into Formula (3).

$$G_1(s)=\frac{3092.02817}{9.8115664s+0.0501706}$$

(4)

3.2. Dynamic Temperature Model of the Tobacco Baking Room in Tobacco Bulk Curing Barn

Similarly, due to the balance and symmetry of input and output of the curing barn temperature, according to the law of energy conservation, the energy balance equation of the tobacco baking room is the Equation (5).

$$\Delta Q_1=Q_{heatingchamber}-Q_{dehumidification}-Q_{1_{heat dissipation}}-Q_{tobacco leaves}$$

(5)

In the Formula (5): ΔQ₁—the heat change of the tobacco baking room in a unit time; Q_{heating chamber}—the heating chamber supplies the heat of the tobacco baking room per unit time; Q_{dehumidification}—the heat lost by the hot air in the smoke chamber of the heating chamber dehumidification equipment in a unit time; Q_{1heat dissipation}—the heat lost in the tobacco baking room per unit time; Q_{tobacco leaves}—the heat consumed per unit time of baking.

The heat changed in the temperature of the tobacco baking room per unit time was calculated according to the heating chamber supplies the heat of the tobacco baking room per unit time and the heating chamber loss the heat of the tobacco baking room per unit time; the heating chamber loss the heat of the tobacco baking room per unit time is determined by the heat lost by the hot air in the smoke chamber of the heating chamber dehumidification equipment in a unit time, the heat lost in the tobacco baking room per unit time and the heat consumed per unit time of baking.

$${\rho }_2{C}_{\mathrm{p}2}{V}_2\frac{\mathrm{d}{T}_{\mathrm{i}}}{\mathrm{d}t}=m_{\mathrm{p}}{C}_{\mathrm{p}2}(T_{\mathrm{p}}-T_{\mathrm{i}})-A_1{h}_1(T_{\mathrm{i}}-T_{\mathrm{o}})-A_3{h}_3(T_{\mathrm{i}}-T_{\mathrm{o}})-m_{\mathrm{t}}{C}_{\mathrm{pt}}(T_{\mathrm{i}}-T_{\mathrm{t}})$$

(6)

The mathematical model Formula (7) of the temperature of the tobacco baking room can be obtained by simplifying the Formula (6) with Laplace transform.

$$G_2(s)=\frac{T_{\mathrm{i}}(s)}{T_{\mathrm{p}}(s)}=\frac{m_{\mathrm{p}}{C}_{\mathrm{p}2}}{{\rho }_2{C}_{\mathrm{p}2}{V}_{2}s+m_{\mathrm{p}}{C}_{\mathrm{p}2}+A_1{h}_1+A_3{h}_3+m_{\mathrm{t}}{C}_{\mathrm{pt}}}$$

(7)

In the Formula (7): ρ₂—the indoor air density of the tobacco baking room, 1.128 kg/m³; C_p2—the specific heat capacity at constant pressure of indoor air in the tobacco baking room, 1.005 × 10³ J/(kg·K); V₂—the volume of the tobacco baking room; T_i—the smoke chamber temperature, K; $\frac{\mathrm{d}{T}_{\mathrm{i}}}{\mathrm{d}t}$—the rate of temperature change in the tobacco baking room; m_p—the air mass flow per unit time, 0.316 kg/s; T_p—the air supply temperature, K; T_O—the outdoor air temperature, 298 K; A₃—the inner wall area of the tobacco baking room; h₃—the heat transfer coefficient between the envelop structure and the external environment of the tobacco baking room, 0.79 w/(m·K). m _t—the quality of the tobacco baking room tobacco leaves, 3600 kg; C_pt—the constant pressure specific heat capacity of the tobacco baking room tobacco leaves, 12.773 J/(kg·K); T_t—the temperature of tobacco leaves in the tobacco baking room, T_t = (T_i−3) K.

By substituting the parameters into the Formula (7), the mathematical model Formula (8) of the dynamic transfer function of the temperature of the tobacco baking room can be obtained.

$$G_2(s)=\frac{0.31758}{85.703184s+46.4191226}$$

(8)

3.3. Dynamic Temperature Model of Full Closed Hot Air Circulation Tobacco Bulk Curing Barn

By cascaded G₁(s), the temperature control model of the heating chamber, and G₂(s), the temperature control model of the tobacco baking room, the dynamic temperature control model G(s) of the full closed hot air circulation tobacco bulk curing barn was obtained as the Formula (9):

$$G(s)=G_1(s)G_2(s)=\frac{981.966306}{804.882485s^2+459.744086s+2.32887523}$$

(9)

4. Simulation of Temperature Control System for Full Closed Hot Air Circulation Tobacco Bulk Curing Barn

4.1. Design of Temperature Control System for Full Closed Hot Air Circulation Tobacco Bulk Curing Barn

The PID control is a widely used strategy in the industrial process control, and novel seeker optimization algorithm has also been used to many assignment [20,21,22,23,24,25,26,27,28], and sobol sequence in mathematics has good uniform distribution characteristics [29,30,31]. Therefore, this paper adopts the PID control strategy based on sobol sequence seeker optimization algorithm to construct the temperature control system of the full closed hot air circulation tobacco bulk curing barn. This control strategy can reduce the number of iterations, avoid falling into local optimization prematurely, and improve the search ability of the algorithm [32,33,34]. The composition of the temperature control system constructed by it is shown in the Figure 4.

4.2. PID Control of Tobacco Bulk Curing Barn Temperature Based on Sobol Sequence Seeker Optimization Algorithm

4.2.1. Sobol Sequence Seeker Optimization Algorithm Optimize the Tobacco Bulk Curing Barn Temperature PID Process

The sobol sequence seeker optimization algorithm (SSOA) was used to optimize the PID parameters of the full closed hot air circulation tobacco bulk curing barn temperature system. The specific process is shown in the Figure 5.

4.2.2. Parameters Setting

Dai [20] did some studies for the parameters set of seeker optimization algorithm (SOA). The parameters of SSOA are based on the actual experience of the appropriate value. The

Table 1. The parameters set of Algorithms.

Algorithm	Parameters and Value
PSO [35]	c0: 0.9~0.4, c1: 1.4962, c2: 1.4962.
SA-GA [36]	ps: 0.6, pc: 0.7, pm: 0.05, t0: 100, Δt: 0.98.
DE [37]	pm: 0.6, Pc: 0.09.
DA [38]	s: 0~0.2, a: 0~0.2, c: 0~0.2, f: 0~2, e: 0~0.1, w: 0.9~0.4.
BSO [39]	nc: 10, ps: 0.8, psc: 0.2, pc: 0.4.
GSA [40]	g0: 100, alfa: 20, rfinal_per: 2.
SSA [41]	c1 = 0~2, c2 = 0~1, c3 = 0~1.
MVO [42]	WEP_Max: 1, WEP_Min: 0.2, TDR = 0~1, r1 = 0~1, r2 = 0~1, r3 = 0~1.
SOA [43]	η_max 0.95, η_min: 0.0111, w_max: 0.8, w_min: 0.2.
SSOA	η_max 0.95, η_min: 0.0111, w_max: 0.8, w_min: 0.2.

is the parameters in our test.

For the temperature control system of the full closed hot air circulation tobacco bulk curing barn, the population number is 20, the generation is 20, and the range of K_p, K_i and K_d [0,300], refer to

Table 2. Comparison of performance optimal fitness of algorithms independent run 30 times PID control in tobacco bulk curing barn temperature (The numbers in bold are the best).

Simulation Result	Algorithms
	PSO	SA-GA	DE	DA	BSO	GSA	SSA	MVO	SOA	SSOA
Mean ITAE value	0.35251	0.5053	0.5071	0.2967	0.6804	0.6494	0.2699	0.3953	0.28875	0.099885
Standard deviation	0.20955	0.1059	0.0918	0.2134	0.1516	0.1232	0.2122	0.1963	0.18233	0.00012
Maximum ITAE value	0.54032	0.6176	0.6181	0.5411	1.0432	0.9683	0.5448	0.5343	0.53616	0.1002
Minimum ITAE value	0.1000003	0.1555	0.2582	0.1000	0.3982	0.3806	0.0996	0.1000	0.09991	0.0996
Run times (s)	524.7	10631.0	1036.2	669.6	553.3	618.3	481.4	676.5	1683. 6	1702.5

for other parameters of various algorithms.

4.2.3. Algorithm Performance Comparison

In the control system, ITAE (integral of time absolute error) value is often used to evaluate the system. The ITAE value is smaller, it indicates that the control system is the optimal system. The Formula (10) is the solution formula of ITAE [44].

$$ITAE={\displaystyle {\int }_0^{\infty }t}\left|e(t)\right|\mathrm{d}t={\displaystyle {\int }_0^{\infty }t}\left|r(t)-y(t)\right|\mathrm{d}t$$

(10)

In the Formula (10): t—the time, s; e(t)—the system error; r(t)—the system input; y(t)—the system output.

Some algorithms are used to optimize the full closed hot air circulation tobacco bulk curing barn temperature control system PID parameters. These algorithms are the particle swarm optimization (PSO), the combined genetic algorithm/simulated annealing algorithm (SA-GA), the differential evolution algorithm (DE), the dragonfly algorithm (DA), the brain storm optimization algorithm (BSO), the gravitational search algorithm (GSA), the salp swarm algorithm (SSA), the multi-verse optimizer (MVO), the SOA and the SSOA. These programs of algorithm are independently run 30 times. After these programs are independently run of 30 times, the minimum fitness ITAE value, the maximum fitness ITAE value, standard deviation, average fitness ITAE value, and program run times are saved, as shown in the Table 2.

The optimal fitness between the PSO, SA-GA, DE, DA, BSO, GSP, SSA, MVO, SOA and the SSOA algorithm run for the temperature control system of the full closed hot air circulation tobacco bulk curing barn are shown in the Table 2. The results of the SSOA are better than the other algorithms. The result of minimum ITAE value is 0.0996, the mean ITAE value is 0.099885, the standard deviation is 0.00012, the maximum ITAE value is 0.1002. The minimum ITAE value, the standard deviation, the mean ITAE value and the maximum ITAE value of the SSOA are obviously better than the other algorithms. As seen from Table 2, the PSO algorithm has the more minor program running time, followed by the BSO algorithm, which has less program running time. At the bottom of the list is the SSOA algorithm, which takes the most running time. However, the program running time of SSOA algorithm and SOA algorithm are not much different. The time complexity of the improved SSOA algorithm is basically the same as that of the SOA algorithm.

Meanwhile, the Figure 6 is the ANOVA tests of the global minimum. From the ANOVA, the SSOA is the most robust in these algorithms. The SSOA is a valid and viable method for the majorization PID control problem.

4.2.4. Simulation Results and Analysis of PID Control for Tobacco Bulk Curing Barn Temperature

Since the fitness ITAE values of the PSO, SA-GA, DE, DA, BSO, GSP, SSA, MVO, SOA and the SSOA algorithm not significantly different and the step response curves are similar, this section only compares the step response graphs of the PID fuzzy control and the SSOA optimization PID control.

To the dynamic model of the full closed hot air circulation tobacco bulk curing barn temperature control, the PID control with fuzzy PID control and the seeker optimization algorithm based on sobol sequence are applied to the PID control step response simulation control, the sampling time is 5 s, the PID control parameters of the two algorithms are respectively independent 30 times of the programs, the fitness value is the smallest one is chosen as the optimal value, the fuzzy PID control parameters K_p = 70, K_i = 0.5, K_d = 4, ITAE = 0.43145, and the sobol sequence seeker optimization algorithm PID control parameters K_p = 299.424, K_i = 0.00541, K_d = 53.891, ITAE = 0.099625. The simulation results of the step response curves for the PID control system are shown in the Figure 7.

The PID control of seeker optimization algorithm based on the sobol sequence has obvious improvement over the fuzzy PID control in overshooting and adjusting time. In terms of overshoot, the overshoot controlled by the PID sobol sequence seeker optimization algorithm is 0.6%, while the overshoot controlled by fuzzy PID is 44%. In terms of adjusting the time to reach stability, the PID control sobol sequence seeker optimization algorithm takes 0.8 s, and the fuzzy PID control takes up to 3 s. The control system based on the sobol sequence seeker optimization algorithm can greatly perfect the transient property of the temperature control system, and its control effect is better than that of the fuzzy PID control.

5. Comparative Experiment on PID Control of Temperature of Full Closed Hot Air Circulation Tobacco Bulk Curing Barn

In a tobacco cooperative in Yunnan tobacco district, two full closed hot air circulation tobacco bulk curing barns were rebuilt. Two kinds of tobacco bulk curing barns controllers were installed in two tobacco bulk curing barns equipped with tobacco leaves. The tobacco bulk curing barn controller 1 adopts fuzzy PID control, and its initial parameters K_p = 70 K_i = 0.5, K_d = 4. The tobacco bulk curing barn controller 2 adopts PID control based on the sobol sequence seeker optimization algorithm, and its initial parameters K_p = 299.424, K_i = 0.00541, K_d = 53.891.

The temperature control curve of the actual dry ball in the tobacco baking room using tobacco bulk curing barn controller 1 is shown in the Figure 8. The maximum error of the dry ball temperature is 1.1 °C, and the average error is about 0.5 °C, indicating that the over harmonic and oscillation are obvious and the error is large.

The temperature control curve of the actual dry ball in the tobacco baking room using tobacco bulk curing barn controller 2 is shown in the Figure 9. The maximum error of the dry ball temperature of the full closed hot air circulation tobacco bulk curing barn controlled by the PID based on the sobol sequence seeker optimization algorithm is 0.5 °C, and the average error is about 0.2 °C, Compared with the full closed hot air circulation tobacco bulk curing barn controlled by the fuzzy PID, the temperature over harmonic and error of the tobacco baking room is obviously reduced, and the control accuracy is obviously improved.

6. Conclusions

In this paper, a new tobacco bulk curing barn with totally closed hot air circulation is designed, by adding dehumidification equipment instead of open dehumidification. According to the four phases to analyze the new baking barn. These four phases include the structure design of the new baking room, the establishment of the mathematical model of temperature control, the simulation and actual experiment of the temperature control system of the baking room.

In the first phase, a new tobacco bulk curing barn is designed. The curing barn adopts a totally closed hot air circulation mode by adding new dehumidification equipment. The concrete structure of the newly designed oven and the condensation dehumidifier are described in detail. The new curing barn can avoid the defect of open humidity drainage, avoid the loss of residual heat, further improve the energy utilization rate, save energy and quality of the tobacco leaves in the dense baking room.

In the second phase, according to the energy conservation theorem, by cascaded the temperature control model of the heating chamber and the temperature control model of the tobacco baking room, the dynamic temperature control model of the full closed hot air circulation tobacco bulk curing barn was obtained. The mathematical model of temperature control system for the full closed hot air circulation tobacco bulk curing barn is a second order model. The coefficient of this mathematical model is very large and the PID parameters of temperature control system are difficult to determine, which makes it a fragile second-order control system model.

In the third phase, based on the sobol sequence seeker optimization algorithm PID control strategy, the temperature control system of the full closed hot air circulation tobacco bulk curing barn is simulated. The SSOA algorithm is proposed by using sobol sequence to improve SOA algorithm. The PSO, SA-GA, DE, DA, BSO, GSP, SSA, MVO, SOA and the SSOA algorithm are used to optimize the full closed hot air circulation tobacco bulk curing barn temperature control system PID parameters. Although the SSOA algorithm has the most program running time, the optimal fitness ITAE value of the SSOA are better than the other algorithms. The ANOVA of the SSOA is the most robust in these algorithms. The PID control with fuzzy PID control and the seeker optimization algorithm based on sobol sequence are applied to the PID control step response simulation control. According to the simulation results of the step response curves for the PID control system, the control system based on the sobol sequence seeker optimization algorithm can greatly perfect the transient property of the temperature control system, and its control effect is better than that of the fuzzy PID control.

In the last phase, based on the sobol sequence seeker optimization algorithm PID control strategy, the temperature control system of the full closed hot air circulation tobacco bulk curing barn is simulated. Compared with the fuzzy PID control, the temperature control system of the full closed hot air circulation tobacco bulk curing barn based on the sobol sequence seeker optimization algorithm PID control strategy has obvious improvement in overregulation and adjustment time. The maximum error of the dry ball temperature for the full closed hot air circulation tobacco bulk curing barn controlled by PID based on the sobol sequence seeker optimization algorithm is 0.5 °C, and the average error is about 0.2 °C. For the temperature control system of the full closed hot air circulation tobacco bulk curing barn with high inertia and time variability, the PID control based on the sobol sequence seeker optimization algorithm has better stability and dynamic characteristics.