PAPERmaking! Vol9 Nr1 2023

PAPER making! The e-magazine for the Fibrous Forest Products Sector

Produced by: The Paper Industry Technical Association

Publishers of: Paper Technology International ®

Volume 9 / Number 1 / 2023

PAPER making! FROM THE PUBLISHERS OF PAPER TECHNOLOGY INTERNATIONAL ® FROM THE PUBLISHERS OF PAPER TE Volume 9, Number 1, 2023 CONTENTS:

FEATURE ARTICLES: 1. Papermaking : Basis weight control loop of papermaking process. 2. Tissue : Chitosan biopolymers in sanitary and hygiene applications. 3. Decarbonisation : Decarbonization of the European pulp and paper industry. 4. Coated Paper : Dispersion coatings for paper-based packaging. 5. Biorefinery : Non-wood forest products potential for bioeconomy. 6. Packaging : The price of sustainability: trading benefits against sustainability. 7. Wood Panel : Panels from eucalyptus sawdust and vegetal polyurethane resin. 8. UK Industry : CPI report on economic value of UK paper-based industries 2022. 9. Email organisation : How to organize your email inbox. 10. Note Taking : The best note-taking methods. 11. Motivation : Employee motivation – 9 simple tips for managers. 12. Office Exercises : Workplace exercises to keep you healthy at the office.

SUPPLIERS NEWS SECTION: News / Products / Services :

Section 1 – PITA Corporate Members: ABB / VALMET Section 2 – PITA Non-Corporate Members ANDRITZ / VOITH Section 3 – Non-PITA Members A.CELLI / HI-LINE

DATA COMPILATION: Events : PITA Courses & International Conferences / Exhibitions Installations : Overview of equipment orders and installations since early November 2022 Research Articles : Recent peer-reviewed articles from the technical paper press Technical Abstracts : Recent peer-reviewed articles from the general scientific press The Paper Industry Technical Association (PITA) is an independent organisation which operates for the general benefit of its members – both individual and corporate – dedicated to promoting and improving the technical and scientific knowledge of those working in the UK pulp and paper industry. Formed in 1960, it serves the Industry, both manufacturers and suppliers, by providing a forum for members to meet and network; it organises visits, conferences and training seminars that cover all aspects of papermaking science. It also publishes the prestigious journal Paper Technology International ® and the PITA Annual Review , both sent free to members, and a range of other technical publications which include conference proceedings and the acclaimed Essential Guide to Aqueous Coating .

Contents

Page 1 of 1

PAPER making! FROM THE PUBLISHERS OF PAPER TECHNOLOGY INTERNATIONAL ® FROM THE PUBLISHERS OF PAPER TE Volume 9, Number 1, 2023

Novel Parameter Identification Method for Basis Weight Control Loop of Papermaking Process YUNZHU SHEN 1 , WEI TANG 1 , & YUNGANG LIU 2 The basis weight control loop of the papermaking process is a non-linear system with time-delay and time- varying. It is impractical to identify a model that can restore the model of real papermaking process. Determining a more accurate identification model is very important for designing the controller of the control system and maintaining the stable operation of the papermaking process. In this study, a strange nonchaotic particle swarm optimization (SNPSO) algorithm is proposed to identify the models of real papermaking processes, and this identification ability is significantly enhanced compared with particle swarm optimization (PSO). First, random particles are initialized by strange nonchaotic sequences to obtain high-quality solutions. Furthermore, the weight of linear attenuation is replaced by strange nonchaotic sequence and the time-varying acceleration coefficients and a mutation rule with strange nonchaotic characteristics are utilized in SNPSO. The above strategies effectively improve the global and local search ability of particles and the ability to escape from local optimization. To illustrate the effectiveness of SNPSO, step response data are used to identify the models of real industrial processes. Compared with classical PSO, PSO with timevarying acceleration coefficients (PSO-TVAC) and modified particle swarm optimization (MPSO), the simulation results demonstrate that SNPSO has stronger identification ability, faster convergence speed, and better robustness. Contact information: 1. School of Electrical and Control Engineering, Shaanxi University of Science and Technology, Xi'an, Shaanxi Province, 710021, China 2. School of Control Science and Engineering, Shandong University, Ji'nan, Shandong Province, 250061, China PMB, Vol.8, No.1, 2023 https://doi.org/10.26599/PBM.2023.9260004 Creative Commons Attribution 4.0 License

The Paper Industry Technical Association (PITA) is an independent organisation which operates for the general benefit of its members – both individual and corporate – dedicated to promoting and improving the technical and scientific knowledge of those working in the UK pulp and paper industry. Formed in 1960, it serves the Industry, both manufacturers and suppliers, by providing a forum for members to meet and network; it organises visits, conferences and training seminars that cover all aspects of papermaking science. It also publishes the prestigious journal Paper Technology International ® and the PITA Annual Review , both sent free to members, and a range of other technical publications which include conference proceedings and the acclaimed Essential Guide to Aqueous Coating .

Article 1 – Papermaking Basis Weight Control

Page 1 of 16

1#. • Basis Weight Control System

/PWFM1BSBNFUFS*EFOUJGJDBUJPO.FUIPE GPS#BTJT8FJHIU$POUSPM-PPQPG 1BQFSNBLJOH1SPDFTT :VO[IV4IFO 8FJ5BOH :VOHBOH-JV 1 . School of Electrical and Control Engineering, Shaanxi University of Science and Technology, Xi'an, Shaanxi Province, 710021 , China 2 . School of Control Science and Engineering, Shandong University, Ji'nan, Shandong Province, 250061 , China "CTUSBDU The basis weight control loop of the papermaking process is a non-linear system with time-delay and time-varying. It is impractical to identify a model that can restore the model of real papermaking process. Determining a more accurate identification model is very important for designing the controller of the control system and maintaining the stable operation of the papermaking process. In this study, a strange nonchaotic particle swarm optimization (SNPSO) algorithm is proposed to identify the models of real papermaking processes, and this identification ability is significantly enhanced compared with particle swarm optimization (PSO). First, random particles are initialized by strange nonchaotic sequences to obtain high-quality solutions. Furthermore, the weight of linear attenuation is replaced by strange nonchaotic sequence and the time-varying acceleration coefficients and a mutation rule with strange nonchaotic characteristics are utilized in SNPSO. The above strategies effectively improve the global and local search ability of particles and the ability to escape from local optimization. To illustrate the effectiveness of SNPSO, step response data are used to identify the models of real industrial processes. Compared with classical PSO, PSO with time- varying acceleration coefficients (PSO-TVAC) and modified particle swarm optimization (MPSO), the simulation results demonstrate that SNPSO has stronger identification ability, faster convergence speed, and better robustness. ,FZXPSET basis weight control system; papermaking; system identification; particle swarm optimization; strange nonchaotic sequence %0* 10.26599/PBM.2023.9260004

Yunzhu Shen, PhD candidate; E-mail: jdshenyunzhu@163.com

*Corresponding author: Wei Tang, professor, PhD tutor; research interest: papermaking automation; E-mail: wtang906@163.com

Received: 14 November 2022; accepted: 24 December 2022.

7PM /P

1#. • Basis Weight Control System

` )NTRODUCTION The basis weight control loop of the papermaking process has the characteristics of large time-delay and external interference; hence, it is difficult to establish an accurate mathematical model [1] . Previous studies [2 3] have demonstrated that its model is usually approximated to a low-order transfer function with time-delay, including first-order-plus-dead-time (FOPDT) or second-order- plus-dead-time (SOPDT) structures. Because these models not only effectively reflect the basic dynamics of the paper industry process, they can also contribute to the design of the controller. With the extensive use of model-based advanced control methods in the papermaking process, the high-precision identification of models is one of the challenges that urgently needs to be solved in the control field. Therefore, it is very important to identify the FOPDT or SOPDT models for designing the controller and maintaining stable operation of the papermaking process. Several effective system identification techniques [4 9] have been proposed. The least square method (LSM) [10] is an effective method to identify models of time-delay. This method is an identification algorithm based on discrete systems; therefore, the continuous transfer function needs to be discretized. The improper selection of sampling intervals and non-integer optimization problem of systems with time-delay can increase the difficulty of identification. A frequency domain identification method [11] is proposed to identify the parameters of a system with time-delay. It requires a small amount of calculation and strong robustness, but ignores the high-frequency dynamics of the system, and is easy to converge locally. Filter-based adaptive identification algorithm [12] is the most commonly employed method to identify time-delay systems and avoid local minima under full excitation; however, full excitation is not suitable for practical industrial processes. Therefore, it is meaningful to determine a system identification method with strong identification ability, good robustness, and suitable for engineering practice. Recently, intelligent optimization technologies have

garnered significant attention in system identification [13 17] . One of the most popular optimization algorithms for system identification is particle swarm optimization (PSO)-based algorithm [18] . PSO algorithm is a population-based optimization algorithm introduced by Kennedy and Eberhart in 1995 [19] . It is used to identify time-delay model parameters and achieved good results [20] . However, classical PSO is easily classified as local optimal value and its local search ability is relatively poor. To improve the performance of PSO algorithm, various versions of the improved PSO algorithm have been proposed. Zou et al [21] proposed an improved PSO algorithm to identify the infinitive impulse response (IIR) system. Using the golden section rate to divide the solution space, different inertia weights and normal distribution improve the global search ability and convergence speed of the algorithm to obtain high-quality solutions. Feng et al [22] proposed a modified PSO algorithm to identify micro piezoelectric actuators. This method improves the ability of global optimization and ensures the accuracy of parameter identification. Strange nonchaos [23 28] is an unstable dynamic behavior with randomness and ergodicity in deterministic nonlinear systems, which is highly sensitive to system parameters. Based on the strange nonchaotic behavior, it can perform the overall search faster than the probability-dependent random ergodic search. In this study, an strange nonchaotic particle swarm optimization (SNPSO) algorithm is proposed to identify the papermaking process. First, random particles are initialized by strange nonchaotic sequences to obtain high-quality solutions. Furthermore, the weights updating with strange nonchaotic features and time-varying acceleration coefficient are used to improve the global search ability and search speed. Finally, a mutation individual with strange nonchaotic characteristics is utilized to further improve the global search ability. Simulation results demonstrate the effectiveness of the algorithm. To highlight the advantages of the proposed algorithm, a comparative study is also implemented, and it is proved that the convergence speed, global search ability, and robustness of SNPSO are superior to classical PSO [19] , PSO with

7PM /P

1#. • Basis Weight Control System

time-varying acceleration coefficients (PSO-TVAC) [29] , and modified particle swarm optimization (MPSO) [22] . ` "ASIS` WEIGHT` CONTROL` IN` PAPERMAKING` PROCESS The evaluation of paper quality includes the basis weight of paper, moisture, ash, color, etc. The basis weight means the weight of paper per unit area expressed in g/m 2 . The smaller the basis weight fluctuation, the more uniform the paper, and the better the quality. The control of basis weight includes cross-direction (CD) basis weight control and machine-direction (MD) basis weight control. The CD basis weight control is completed by the headbox dilution water valve, while the MD basis weight control is completed by the pulp pump controlling the pulp flow. The main reasons for the MD basis weight fluctuation are the changes in paper machine speed, pulp amount, pulp concentration, etc. When the paper machine is in normal production, the paper machine speed is stable; therefore, the change in paper machine speed is generally not considered when considering the MD basis weight control. In addition to the above factors, CD basis weight fluctuation also needs to consider the interaction between dilution water valves. The basis weight control system is illustrated in Fig. 1. The pulp in the pulp tank is transported to the headbox using the pulp pump. The quality control system (QCS) calculates the difference between the value of basis weight measured by the scanner and set value, and the output controls the feeding of the pulp pump to stabilize the pulp amount, so as to achieve the effect of basis weight stability control.

The basis weight control system is a nonlinear, large, time-delay system. Flow, pressure, and other non-linear analog signals are easily subjected to electromagnetic interference during transmission, which is easy to cause detection errors, and then affects the accuracy of the identification model. Time-varying factors such as temperature change, flow jitter, and valve wear make the process object and model time-varying and uncertain. The time delay will greatly increase the control difficulty. With the demand of consumers for high-quality paper and use of advanced control technology, determining a parameter identification method to accurately identify the process model with a small time delay is the key to improving the quality of papermaking process. ` 0RINCIPLE`OF`PARAMETER`IDENTIFICATION`BASED` ON`OPTIMIZATION`ALGORITHM The block diagram of system identification based on PSO/Improved PSO (IPSO) is illustrated in Fig. 2. The purpose is to make the output y ( t ) of the selected time-delay system approach the output ŷ ( t ) of the real unknown system under the same input signal ( u /( t )). Thus, the accurate mathematical model of the real unknown system is determined, which lays a foundation for selecting the appropriate control method and designing the control system. In this study, the SNPSO is adopted to identify the unknown parameters of the transfer function FOPDT and SOPDT. The integral of absolute error (IAE) as the performance index is given by: J = ϯ 0 t e ( t ) d t (1) where e ( t ) = y ( t ) ϖ ŷ ( t ) .

'JH Ȟ Structure of basis weight control system

7PM /P

1#. • Basis Weight Control System

k v k jn + c 1 r 1 ( p jn ϖ x jn ) + c 2 r 2 ( p gn ϖ

= w

v k + 1 jn x jn ) x

k + 1 jn (2) where r 1 and r 2 represent random numbers between 0 and 1, c 1 and c 2 are cognitive and social coefficients, w represents the inertia weight that decays linearly with the number of iterations and is expressed as w = w max ϖ ( w max ϖ w min ) iter / iter max , iter and iter max represent the current iteration number and maximum number of iterations, respectively, and w max and w min are 0.9 and 0.4, respectively. Fig. 3 illustrates a working principle of PSO, which helps to better understand its operation mode. = x jn + v k + 1 jn

` 2EVIEW`OF`THREE`03/`STRATEGIES Since the introduction of the PSO method, it has been applied in several applications, such as system identification [19] , odor source localization [30] , defense against synchronization (SYN) flooding attacks [31 32] , and other complex issues [33] . Here, several PSO algorithms are reviewed and become the performance measurement of SNPSO. 4.1 Ȟ PSO algorithm PSO is an intelligent search algorithm that mimics the social behavior of birds. Every particle in PSO is initialized at a random position within a given search space. These particles collect and exchange information with each other centered on their location. Furthermore, the positions of these particles are updated at a certain speed according to the effective information. Assuming that the dimension of the search space is N , for particle j , the position vector X j and velocity vector V j can be written as X j = ( x j 1 , ..., x jn ) and V j = ( v j 1 , ..., v jn ), respectively. The best position pbest j of the j -th particle is the best previous position that gives the best fitness value, and is expressed as pbest j =( p j 1 , ..., p jn ). The best one among all the positions of particles is the global optimal position gbest , it is expressed as gbest = pbest g = ( p g 1 , ..., p gN ). Each particle modifies its position according to a certain speed v jn ( n =1, ..., N ), and the distance forms the individual optimal and global solutions, pbest jn and gbest n , respectively. The new velocity and position updating equations of each particle in the next iteration are given by: Note: Ȟ n ( t ) is noise signal. 'JH Ȟ Block diagram of system identification based on PSO/ IPSO

4.2 Ȟ PSO-TVAC algorithm The time-varying inertia weight can make PSO algorithm converge to the solution with higher accuracy. However, the cognitive coefficient c 1 and social coefficient c 2 are 'JH Ȟ Flowchart of PSO working principle

7PM /P

1#. • Basis Weight Control System

fixed, which is an unfavorable limitation to determine the optimal solution. In Refs. [32 33], the authors introduce two dynamical coefficients to optimize the target. The cognitive coefficient c 1 decreases with the increase in the number of iterations, and the social coefficient c 2 increases with the increase in the number of iterations. This ensures that individuals can fully roam the entire target space in the early stage of search, rather than spend a lot of energy clustering near the local optimal solution. In the later search stage, small cognitive coefficient and large social coefficient encourage the particles to converge to the global optimal state before the end of the search. The dynamical cognitive coefficient c 1 and social coefficient c 2 are expressed as [26] : c 1 = ( c 1 f ϖ c 1 i ) iter / iter max + c 1 i (3) c 2 = ( c 2 f ϖ c 2 i ) iter / iter max + c 2 i (4) where the cognitive coefficient c 1 decreases from c 1 i to c 1 f , and the social coefficient c 2 increases from c 2 i to c 2 f . c 1 i and c 2 f are set to 2.5, each, while c 1 f and c 2 i are set to 0.5, each. In addition, the position updating Eq. (2) is slightly modified as: x k + 1 jn = x jn + C v k + 1 jn (5) where C = 2/ (2 ϖ ϕ ϖ ϕ 2 ϖ 4 ϕ ) ; ϕ is set to 4.1. 4.3 Ȟ MPSO algorithm The MPSO is introduced in Ref. [19]. The dynamical cognitive coefficient c 1 and social coefficient c 2 are expressed as: c 1 = ( c 1 f ϖ c 1 i ) e ϖ 70 Ç È w k ϖ w min w max ϖ w min 6 + c 1 i (6) c 2 = ( c 2 f ϖ c 2 i ) e ϖ 70 Ç È w k ϖ w min w max ϖ w min 6 + c 2 i (7) where c 1 i , c 1 f , c 2 i , and c 2 f are in the range of [0, 4] and c 1 i > c 2 f > c 2 i > c 1 f . c 1 i and c 2 f are set to 1.7, each, while c 1 f and c 2 i are set to 1.3, each. In addition, a mutation rule is applied for each particle. If r 3 >0.9, then n =[ N·r 4 ] and x jn = x min +( x max ϖ x min ) r 5 . r 3 , r 4 , and r 5 are random values in the range of [0, 1]. ` 3.03/`ALGORITHM A common feature of the above improved PSO algorithm is to improve the convergence accuracy and speed by altering the cognitive and social coefficients, c 1 and c 2 ,

respectively. However, the influence of linear attenuation weight on the algorithm is not considered. For the linear attenuation weight, the change of w completely depends on the current and maximum number of iterations, which leads to slow convergence in the later stage of the iteration. Therefore, it is difficult to jump out of the local optimal solution. In this section, strange nonchaos is introduced into the PSO algorithm and the SNPSO algorithm is developed. Strange nonchaotic behavior is an unpredictable dynamical behavior in nonlinear systems. Mapping this behavior to some parameters in the algorithm will improve the performance of the algorithm. To obtain strange nonchaotic sequences, a strange nonchaotic system is described as follows [23] :

º ¼ ½ ½ ½ ½ v 1 y ( n ) + a + ε cos(2π φ ( n ) ), if y ( n ) Ш 0 v 2 y ( n ) + a + ε cos(2π φ ( n ) ) + 1, if y ( n ) > 0 »

(8)

y ( n + 1) =

φ ( n + 1) = φ ( n ) + w mod (1) (9) There are four unknown parameter values in Eq. (8), which is not conducive for determining strange nonchaotic sequences. Therefore, we first determine the values of parameters v 1 and v 2 via several experiments, reduce the number of unknown variables, and then determine strange nonchaotic sequences using the remaining variable parameters a and č . v 1 and v 2 are constant parameters and fixed at v 1 =0.7 and v 2 =1.2, a and č are bifurcation parameters, n is the number of iterations, w is the frequency and is fixed at w = ( 5 ϖ 1)/2 . Strange nonchaotic behavior is characterized by the phase diagram and largest Lyapunov exponent. The largest Lyapunov exponent ē max is calculated by: λ x = lim N͖ ∞ 1 N ϕ i = 1 n ln| φ f φ x i | (10) The bifurcation parameters a and č are fixed at a = 0.5965 and č =0.1. Fig. 4(a) illustrates the phase diagram of a fractal attractor, and the largest Lyapunov exponent equals ϖ 0.0014 (nonchaotic). Therefore, the fractal attractor is a strange nonchaotic attractor. The strange nonchaotic sequence has features of randomness and non-repetition as illustrated in Fig. 4(b). To obtain

7PM /P

1#. • Basis Weight Control System

selected as their dynamic coefficients. A mutation rule with strange nonchaotic characteristics is applied for each particle. If r 3 >0.9, then n =[ N·r 4 ], x jn = x min +( x max ϖ x min ) ŷ j , and r 3 and r 4 are random values in the range of [0, 1]. ŷ j is the normalized strange nonchaotic sequence. The process of the proposed SNPSO algorithm is illustrated in Fig. 5.

higher quality initialization particles, the strange nonchaotic sequence is employed to generate particles instead of random generation. Because the random distribution is not affected by external forces, the distribution of strange nonchaotic sequences can be altered according to the change of parameters. Therefore, we can determine a group of strange nonchaotic particles uniformly distributed in the target space, which can overcome the uncertainty of the particle mass of the random sequence. It is also adopted to replace the linear attenuation weight to improve the convergence speed of the algorithm. From Fig. 4(b), we observe that the randomness of strange nonchaotic sequences is different from that of linear decay sequences. The strange nonchaotic sequences will quickly determine a valid value of w v , and the non-repetition strange nonchaotic sequences will continue to search for a better value of w near the valid value ( w v ). Eq. (6) and Eq. (7) are 'JH Ȟ (a) phase diagram and (b) time sequence diagram of a strange nonchaotic attractor for a =0.5965 and č =0.1

'JH Ȟ Flow chart of SNPSO algorithm

7PM /P

1#. • Basis Weight Control System

` 3IMULATION`EXPERIMENTS This section aims to study the proposed SNPSO algorithm to address the identification problem of the actual process. Therefore, we choose the FOPDT and SOPDT as the estimation models to approximate the following real process models: P 1 ( s ) = 1 s + 1 e ϖ s (11) P 2 ( s ) = 1 ( s + 1) 2 e ϖ s (12) P 3 ( s ) = 1 ( s + 1) 8 (13) P 4 ( s ) = 9 ( s + 1) ( s 2 + 2 s + 9) (14) The inclusion of systems P 1 and P 2 is to confirm that when modeling FOPDT or SOPDT systems with FOPDT or SOPDT models, all the procedures realise an exact match. The identification performance of four algorithms (classical PSO, PSO-TVAC, MPSO, and the proposed SNPSO) is compared. The parameters of four algorithms are set as follows: the sampling time t =20 s; sampling interval ě t =0.01 s; input is step signal; additive noise is Gaussian white noise with a mean value of 0 and variance of 0.16. The parameters of the four algorithms are presented in Table 1. Matlab R2018a is adopted to perform the above design steps under the environment of Intel(R) Core(TM) i5-4210M CPU@2.60 GHz. In total, 10 independent operations are conducted for each system, and detailed result analysis is described as follows. 5BCMF Ȟ 1BSBNFUFSTPGDMBTTJDBM140 140 - 57"$ .1ź 40 BOE4/140 Methods Population Maxgen

follows:

K F T F s + 1

G F ( s ) = (15) where the parameters K F , T F , and L F are the estimation parameters. The classical PSO, PSO-TAVC, MPSO, and the proposed SNPSO are adopted to identify the selected estimation model. To verify the identification ability of the above algorithms, experiments are conducted at step response with zero noise, step response with Gaussian noise, and denoised step response. When the output ŷ ( t ) is a step response with zero noise ( n ( t )=0), Table 2 presents the identification accuracy, minimum (min), maximum (max), mean, and standard (std) deviation of the optimal loss function using the four methods ten times, respectively. The accuracy of model identification using the SNPSO algorithm is as high as 90%, which is much higher than the other three algorithms. Fig. 6 illustrates the parameters K F , T F , and L F of the estimation model obtained by running the four algorithms ten times, respectively. The classical PSO, PSO-TAVC, MPSO, and proposed SNPSO algorithms are represented by blue, red, green, and yellow star-lines. SNPSO has excellent average convergence and stability. Fig. 7 illustrates the variations for objective function J using four algorithms. The SNPSO method converges faster and more precisely than other PSO-based algorithms. e ϖ L F s To verify the robustness of the proposed algorithm, a Gaussian white noise n ( t ) with variance of 0.16 is added to the output ŷ ( t ) . Fig. 8 illustrates the parameters K F , T F , and L F of the estimation model obtained from ten experiments using four algorithms. The SNPSO ensures that the proportion of the distance | K F ϖ 1|<0.004 is 60%, distance | T F ϖ 1|<0.4 is 60%, and distance | L F ϖ 1|< 0.02 is 80%. The values of three parameters of the estimation model obtained by the other three algorithms 5BCMF Ȟ 4ZTUFNJEFOUJGJDBUJPOPG'015%VTJOHGPVS140 - CBTFEBMHPSJUINT Methods Classical PSO PSO-TVAC MPSO SNPSO Best parameter K F =1, T F =1, L F =1 K F =1, T F =1, L F =1 K F =1, T F =1, L F =1 K F =1, T F =1, L F =1 Accuracy/% 20 20 30 90 Min 0 0 0 0 Max 927 3296 1072 506 Mean 201 432 284 50 Std 300 1023 426 160

Other parameters c 1 =2 c 2 =2 c 1 i =2.5, c 1 f =0.5, c 2 i =0.5, c 2 f =2.5 c 1 i =1.7, c 1 f =1.3, c 2 i =1.3, c 2 f =1.7 c 1 i =1.7, c 1 f =1.3, c 2 i =1.3, c 2 f =1.7

Inertia weight w max =0.9 w min =0.4 w max =0.9 w min =0.4 w max =0.9 w min =0.4 w max =0.9 w min =0.4

Classical PSO

100

PSO-TVAC

100

MPSO

100

SNPSO

100

6.1 Ȟ Transfer function P 1 The real process is assumed to be the transfer function (Eq. (11)). The estimation model (FOPDT model) is as

7PM /P

1#. • Basis Weight Control System

'JH Ȟ Identification parameters of FOPTD with zero noise obtained by running the four algorithms ten times

exceed this range. Therefore, the SNPSO algorithm has excellent robustness in comparison with the other three algorithms. To improve the robustness of the algorithm, the wavelet denoised method is employed to process the signal with noise. The basic idea of wavelet denoising is to remove the wavelet coefficients corresponding to the noise in each frequency band, retain the wavelet decomposition coefficients of the original signal, and then reconstruct the processed coefficients to obtain the pure signal. Fig. 9 illustrates the parameters of the estimation model for identifying the denoised and noised step responses using SNPSO. When denoised

'JH Ȟ Variations of objective function J for FOPTD using four algorithms

'JH Ȟ Identification parameters of FOPTD with Gaussian noise by running the four algorithms ten times

'JH Identification parameters of denoised FOPTD with Gaussian noise by running the four algorithms ten times

7PM /P

1#. • Basis Weight Control System

algorithm is adopted for noisy step response, the identification accuracy rates of parameters K F , T F , and L F by SNPSO are increased from 40%, 20%, and 40% to 70%, 60%, and 50%, respectively. It can enhance the stability and robustness of SNPSO. 6.2 Ȟ Transfer function P 2 When the real process is assumed to be the transfer function (Eq. (12)), the estimation model (SOPTD model) is as follows: G S ( s ) = K S ( T S 1 s + 1) ( T S 2 s + 1) e ϖ L S s (16) Parameters K S , T S 1 , T S 2 , and L S are the estimation parameters and are characterized by the classical PSO, PSO-TVAC, MPSO, and proposed SNPSO. For zero noise step response ŷ ( t ) , Table 3 presents five performance indexes of the four methods. The accuracy and mean of identification (mean) using SNPSO are far better than the other three algorithms. The average convergence and stability of SNPSO are further verified

5BCMF 4ZTUFNJEFOUJGJDBUJPOPG4015%VTJOHGPVS140 - CBTFEBMHPSJUINT Methods Accuracy/% Min Max Mean Std

Best parameter K S =1, T S 1 =1, T S 2 =1, L S =1

6392

1786

2312

Classical PSO

K S =1, T S 1 =1, T S 2 =1, L S =1 K S =1, T S 1 =1, T S 2 =1, L S =1 K S =1, T S 1 =1, T S 2 =1, L S =1

6714

2112

2470

PSO-TVAC

6352

1242

2048

MPSO

6309

630

1995

SNPSO

by comparing the parameters K S , T S 1 , T S 2 , and L S of the estimation model obtained by running the four algorithms ten times (as illustrated in Fig. 10). Fig. 11 illustrates

'JH Identification parameters of SOPTD with zero noise obtained by running the four algorithms ten times

7PM /P

1#. • Basis Weight Control System

the output ŷ ( t ) to verify the robustness of the algorithm. Fig. 12 illustrates the SOPTD system parameters K S , T S 1 , T S 2 , and L S obtained from ten experiments using four algorithms. The parameter value of the estimated system obtained by SNPSO algorithm is closer to the parameter value of the real process. Fig. 13 illustrates that the wavelet denoising method improves the stability

and robustness of SNPSO. 6.3 Ȟ Transfer function P 3

When the real process is assumed to be the transfer function (Eq. (13)) the estimation models are as follows: G F ( s ) = K F T F s + 1 e ϖ L F s (17) and G S ( s ) = K S ( T S 1 s + 1) ( T S 2 s + 1) e ϖ L S s (18) Table 4 and Table 5 present the FOPDT and SOPDT model parameters for processes P 3 , respectively. To compare the identification capabilities of the four

the variations of objective function J for four algorithms during the optimization. The convergence speed and convergence precision of four PSO-based algorithms is clearly illustrated in this figure. SNPSO method converges faster and more precisely than other PSO-based algorithms. The noise signal n ( t ) with variance of 0.16 is added to 'JH Ȟ Variations of objective function J for SOPTD using four algorithms

'JH Identification parameters of SOPTD with Gaussian noise by running the four algorithms ten times

7PM /P

1#. • Basis Weight Control System

'JH Identification parameters of denoised SOPTD with Gaussian noise by running the four algorithms ten times 5BCMF '01%5SFTVMUTGPS 1 VTJOHGPVS140 - CBTFEBMHPź SJUINT

5BCMF 401%5SFTVMUTGPS 1 VTJOHGPVS140 - CBTFEBMHPź SJUINT Methods Min Max Mean Std

Best parameter K F =1.016, T F =2.989, L F =5.473 K F =1.016, T F =2.991, L F =5.473 K F =1.016, T F =2.987, L F =5.476 K F =1.016, T F =2.990, L F =5.473

Best parameter K S =1.008, T S 1 =1.954, T S 2 =2.069,

Min

Max

Mean

Std

Methods

173.5

1323.7

366.3

350.8

Classical PSO

315.1

1251.7

610.6

371.5

Classical PSO

L S =4.221 K S =1.008, T S 1 =2.059, T S 2 =1.962, L S =4.234 K S =1.009, T S 1 =1.919, T S 2 =2.185, L S =4.149 K S =1.008, T S 1 =2.011, T S 2 =2.014, L S =4.226

315.1

1251.7

496.9

319.7

PSO-TVAC

173.3

2093.7

630.7

627.8

PSO-TVAC

315.1

500.3

338.1

57.8

MPSO

175.8

1257.1

323.4

317.5

MPSO

315.1

470.3

316.3

51.7

SNPSO

173.1

958.4

290.5

290.4

SNPSO

algorithms, the tables also include a measure of performance: the integral of absolute error (IAE), where the error in each case is the difference between the "real" process and model output. The statistical results demonstrate that SNPSO can determine the best

approximation parameters, and the average convergence performance of SNPSO is better than other methods. Fig. 14 and Fig. 15 illustrate the best FOPDT and

7PM /P

1#. • Basis Weight Control System

'JH Ȟ Step response of the best FOPDT approaching P 3 using classical PSO, PSO-TVAC, MPSO, and SNPSO

Table 6 and Table 7 present the FOPDT and SOPDT model parameters for processes P 4 , respectively. To compare the identification capabilities of the four algorithms, the tables also include a measure of 'JH Ȟ Nyquist curve: primitive high-order function P 3 (black line), and approximation model using PSO (blue), PSO-TVAC, MPSO, and SNPSO 5BCMF Ȟ '01%5SFTVMUTGPS 1 VTJOHGPVS140 - CBTFEBMHPź SJUINT Methods Min Max Mean Std

SOPDT step responses approximating P 3 using classical PSO, PSO-TVAC, MPSO, and SNPSO. Black line represents the primitive high-order function P 3 . The identification method based on optimization algorithm can make the identification model approach the primary model well. The approximation performance of the SOPDT model is better than that of the FOPDT model. Nyquist curve further proves that the SOPDT model is more approximate to the primary model than the FOPDT model as illustrated in Fig. 16. 6.4 Ȟ Transfer function P 4 When the real process is assumed to be the transfer function (Eq. (14)), the estimation models are as follows: G F ( s ) = K F T F s + 1 e ϖ L F s (19) G S ( s ) = K S ( T S 1 s + 1) ( T S 2 s + 1) e ϖ L S s (20) 'JH Ȟ Step response of the best SOPDT approaching P 3 using classical PSO, PSO-TVAC, MPSO, and SNPSO

Best parameter K F =1.000, T F =0.919, L F =0.294 K F =1.000, T F =0.921, L F =0.318 K F =1.000, T F =0.873, L F =0.360 K F =1.000, T F =0.974, L F =0.243

22.3

2331.2

290.9

718.5

Classical PSO

22.2

4887.8

1119.0

1774.5

PSO-TVAC

23.0

236.7

53.0

66.1

MPSO

22.9

104.1

43.7

29.5

SNPSO

7PM /P

1#. • Basis Weight Control System

5BCMF Ȟ 401%5SFTVMUTGPS 1 VTJOHGPVS140 - CBTFEBMHPź SJUINT Methods Min Max Mean Std

Best parameter K S =1.000, T S 1 =0.996, T S 2 =0.221,

23.6

2908.1

354.2

902.6

Classical PSO

L S =0.000 K S =1.000, T S 1 =0.000, T S 2 =1.036, L S =0.000 K S =1.000, T S 1 =0.935, T S 2 =0.022, L S =0.294 K S =1.000, T S 1 =0.000, T S 2 =0.910, L S =0.354

26.5

2908.1

622.0

1091.5

PSO-TVAC

'JH Step response of the best SOPDT approaching P 4 using classical PSO, PSO-TVAC, MPSO, and SNPSO

22.3

2128.4

558.4

751.6

MPSO

22.3

1648.2

324.8

512.1

SNPSO

performance: the integral of absolute error (IAE), where the error in each case is the difference between the "real" process and model output. The statistical results demonstrate that the average convergence performance of SNPSO is better than other methods. Fig. 17 and Fig. 18 illustrate the best FOPDT and SOPDT step responses approximating P 4 using classical PSO, PSO-TVAC, MPSO, and SNPSO. Black line represents the primitive high-order function P 4 . The identification method based on optimization algorithm can make the identification model approach the primary model well. The approximation performance of the FOPDT model is better than that of the SOPDT model. Nyquist curve further proves that the FOPDT model is more

approximate to the primary model than the SOPDT model as illustrated in Fig. 19. ` #ONCLUSIONS Here, we introduced the application of the strange 'JH Nyquist curve: primitive high-order function P 4 (black line), and approximation model using PSO, PSO-TVAC, MPSO, and SNPSO

'JH Ȟ Step response of the best FOPDT approaching P 4 using classical PSO, PSO-TVAC, MPSO, and SNPSO

7PM /P

1#. • Basis Weight Control System

nonchaotic particle swarm optimization (SNPSO) algorithm in developing a novel method for identifying the optimal parameters of systems with time delay. Based on the randomness and ergodicity of strange nonchaotic dynamics, the strange nonchaotic sequence was adopted to replace the initialized random particles and linear attenuation weight to improve the global search capability. Furthermore, a mutation rule with strange nonchaotic characteristics was employed to improve global search capability further. To illustrate the effectiveness of SNPSO, it was compared with the other three algorithms (classical particle swarm optimization (PSO), PSO with time-varying acceleration coefficients (PSO-TVAC), and modified particle swarm optimization (MPSO)). Simulation results demonstrated that SNPSO is more suitable for parameter identification of systems with time delay. When the first-order-plus-dead-time (FOPDT) or second-order-plus-dead-time (SOPDT) models were employed to model the FOPDT or SOPDT systems, SNPSO achieved the highest accuracy of accurate matching. To identify high-order systems, SNPSO had the best average and best minimum convergence performances. !CKNOWLEDGMENTS The authors are grateful for the financial support received from the National Natural Science Foundation of China (Grant No. 62073206) and Technical Innovation Guidance Project of Shaanxi Province (Grant No. 2020CGHJ-007). 2EFERENCES [1] Ammar M E, Dumont G. Identification of paper machines cross-directional models in closed-loop. In the 2013 5th International Conference on Modelling, Identification and Control (ICMIC) ICMIC, Cairo, Egypt: 2013, 3-9. [2] Sun G, Nie H, Su Y, Nie W. Research on two-degree-of- freedom IMC-PID control for time-delay systems. Application Research of Computers , 2014, 31(8), 2357-2360. [3] Shan W, Tang W, Wang M, Liu B. Two-degree-of-freedom smith predictor based on fractional order PID controller in paper basis weight. Packaging Engineering , 2017, 38(11), 143-147. [4] Xu Y, Chen D. Partially-linear least-squares regularized regression for system identification. IEEE Transactions on Automatic Control , 2009, 54(11), 2637-2641.

[5] Chen T, Ljung L. Implementation of algorithms for tuning parameters in regularized least squares problems in system identification. Automatica , 2013, 49(7), 2213-2220. [6] Poinot T, Trigeassou J C. Identification of fractional systems using an output-error technique. Nonlinear Dynamics , 2004, 38(1/4), 133-154. [7] Silva W. Identification of nonlinear aeroelastic systems based on the volterra theory: progress and opportunities. Nonlinear Dynamics , 2005, 39(1/2), 25-62. [8] Gibson S, Wills A, Ninness B. Maximum-likelihood parameter estimation of bilinear systems. IEEE Transactions on Automatic Control , 2005, 50(10), 1581-1596. [9] Soderstrom T, Hong M, Schoukens J, Pintelon R. Accuracy analysis of time domain maximum likelihood method and sample maximum likelihood method for errors-in-variables and output error identification. Automatica , 2010, 46(4), 721-727. [10] Gu Y, Ding R. A least squares identification algorithm for a state space model with multi-state delays. Applied Mathematics Letters , 2013, 26(7), 748-753. [11] Tao L, Gao F. A frequency domain step response identification method for continuous-time processes with time delay. Journal of Process Control , 2010, 20(7), 800-809. [12] Na J, Ren X, Xia Y. Adaptive parameter identification of linear SISO systems with unknown time-delay. Systems & Control Letters , 2014, 66, 43-50. [13] Ahandani M A, Abbasfam J, Kharrati H. Parameter identification of permanent magnet synchronous motors using quasi-opposition-based particle swarm optimization and hybrid chaotic particle swarm optimization algorithms. Applied Intelligence , 2022, 52, 13082-13096. [14] Zhu M, Axel H, Wen Y. Identification-based controller design using cloud model for course-keeping of ships in waves. Engineering Applications of Artificial Intelligence , 2018, 75, 22-35. [15] Lagos-Eulogio P, Seck-Tuoh-Mora J C, Hernandez-Romero N, Medina-Marin J. A new design method for adaptive IIR system identification using hybrid CPSO and DE. Nonlinear Dynamics , 2017, 88(4), 2371-2389. [16] Prawin J, Rao A, Lakshmi K. Nonlinear parametric identification strategy combining reverse path and hybrid dynamic quantum particle swarm optimization. Nonlinear Dynamics , 2015, 84(4), 797-815. [17] Peng Y, He S, Sun K. Parameter identification for discrete memristive chaotic map using adaptive differential evolution algorithm. Nonlinear Dynamics , 2021, 107(1), 1263-1275. [18] Wang J, Xu Y, She C, Xu P, Bagal H A. Optimal parameter identification of SOFC model using modified gray wolf optimization algorithm. Energy , 2022, DOI: 10.1016/j. energy.2021.122800. [19] Kennedy J, Eberhart R. Particle swarm optimization. IEEE

7PM /P

1#. • Basis Weight Control System

Conference on Neural Networks , 1995, 4, 1942-1948. [20] Cox C, Tindle J, Burn K. A comparison of software-based approaches to identifying FOPDT and SOPDT model parameters from process step response data. Applied Mathematical Modelling , 2016, 40(1), 100-114. [21] Zou D, Deb S, Wang G. Solving IIR system identification by a variant of particle swarm optimization. Neural Computing and Applications , 2018, 30(3), 685-698. [22] Feng Y, Li Y, Liang M. System Identification of micro piezoelectric actuators via rate-dependent prandtl-ishlinskii hysteresis model based on a modified PSO algorithm. IEEE Transactions on Nanotechnology , 2020, DOI: 10.1109/ TNANO.2020.3034965. [23] Shen Y, Zhang Y. Mechanisms of strange nonchaotic attractors in a nonsmooth system with border-collision bifurcations. Nonlinear Dynamics , 2019, 96(2), 1405-1428. [24] Danca M F, Feckan M, Kuznetsov N, Chen G. Rich dynamics and anticontrol of extinction in a prey-predator system. Nonlinear Dynamics , 2019, 98(2), 1421-1445. [25] Yuan Y, Miao P, Xie J. Coexistence of strange nonchaotic attractors and a special mixed attractor caused by a new intermittency in a periodically driven vibro-impact system. Nonlinear Dynamics , 2017, 87(2), 1-21. [26] Shen Y, Zhang Y. Strange nonchaotic attractors in a quasiperiodically-forced piecewise smooth system with farey tree. Fractals , 2019, DOI: 10.1142/S0218348X19501184.

[27] Li G, Yue Y, Grebogi C, Li D, Xie J. Strange nonchaotic attractors and multistability in a two-degree-of-freedom quasiperiodically forced vibro-impact system. Fractals , 2021, DOI: 10.1142/S0218348X21501036. [28] Zhang Y, Shen Y. A new route to strange nonchaotic attractors in an interval map. International Journal of Bifurcation and Chaos , 2020, DOI: 10.1142/S0218127420500637. [29] Mohammadi-Ivatloo B, Moradi-Dalvand M, Rabiee A. Combined heat and power economic dispatch problem solution using particle swarm optimization with time varying acceleration coefficients. Electric Power Systems Research , 2013, 95, 9-18. [30] Lu Q, Han Q, Liu S. A finite-time particle swarm optimization algorithm for odor source localization. Information Sciences , 2014, 277(1), 111-140. [31] Ahmed Z, Mahbub M, Jahan S. Defense against SYN flood attack using LPTR-PSO: a three phased scheduling approach. International Journal of Advanced Computer Science and Applications , 2017, 8(9), 433-441. [32] Jamali S, Shaker V. Defense against SYN flooding attacks: a particle swarm optimization approach. Computers & Electrical Engineering , 2014, 40(6), 2013-2025. [33] Gaxiola F, Melin P, Valdez F, Castro J R, Castillo O. Optimization of type-2 fuzzy weights in backpropagation learning for neural networks using GAs and PSO. Applied Soft Computing , 2016, 38, 860-871. PBM

7PM /P

PAPER making! FROM THE PUBLISHERS OF PAPER TECHNOLOGY INTERNATIONAL ® FROM THE PUBLISHERS OF PAPER TE Volume 9, Number 1, 2023

Review Article: Chitosan Superabsorbent Biopolymers in Sanitary and Hygiene Applications PEENAL ARVIND MISTRY 1 MEERA NAMBIDAS KONAR 1 SRINIVASAN LATHA 2 UTKARSH CHADHA 3 PREETAM BHARDWAJ 4,5 & TOLERA KUMA ETICHA 6 The consumption of diapers and sanitary products has constantly been rising. Several problems are associated with using chemical-based sanitary products, which are difficult to degrade easily and cause nappy rash and bacterial infections in babies. Therefore, there is an increasing shift towards natural-based sanitary products because of their biodegradability, non-toxicity, and biocompatibility. Several studies are being carried out in which researchers have incorporated natural polymers, such as cellulose, starch, alginate, and xantham gum for producing superabsorbent materials. Chitosan (CS) is one such natural polymer that exhibits anti-microbial activity because of the functional groups present in its structure. Moreover, it is also easily available, biodegradable, and non- toxic. This review mainly focuses on CS’s properties and several approaches to synthesizing natural polymer-based superabsorbent products, such as sanitary pads and diapers. It also briefly discusses the diversified applications of CS as a biopolymer in the cosmetic, medical, food, and textile industries. In addition, this study implies using CS as a superabsorbent biopolymer in the manufacturing and producing sanitary products for women and children. Due to the excellent water retention capacity, swelling ability, and anti-microbial activity exhibited by CS can be considered a potential candidate for producing superabsorbent biopolymers. Contact information: 1 School of Bio Sciences and Technology, Vellore Institute of Technology, Vellore, Tamil Nadu 632014, India 2 Departement of Chemistry, School of Advanced Sciences, Vellore Institute of Technology, Vellore, Tamil Nadu 632014, India 3 Department of Materials Science and Engineering, Faculty of Applied Sciences and Engineering, School of Graduate Studies, University of Toronto, Toronto, ON, M5S 2Z9, Canada 4 Research and Development Cell, Battrixx, Kabra Extrusion Technik Limited, Chakan Industrial Area, Phase II, Village Bhamboli, Chakan, Tal-Khed, Pune 410501, India 5 School of Electronics Engineering, Vellore Institute of Technology, Vellore, Tamil Nadu 632014, India 6 Department of Biology, College of Natural and Computational Sciences, Ambo University, Ambo, Ethiopia Hindawi - International Journal of Polymer Science, Volume 2023, Article ID 4717905, 14 pages https://doi.org/10.1155/2023/4717905 Creative Commons Attribution International License 4.0

Article 2 – Tissue

Page 1 of 15

PAPER making! FROM THE PUBLISHERS OF PAPER TECHNOLOGY INTERNATIONAL ® FROM THE PUBLISHERS OF PAPER TE Volume 9, Number 1, 2023

Decarbonization Prospects for the European Pulp and Paper Industry: Different Development Pathways and Needed Actions SATU LIPIÄINEN, EEVA-LOTTA APAJALAHTI & ESA VAKKILAINEN. The pulp and paper industry (PPI) has several opportunities to contribute to meeting prevailing climate targets. It can cut its own CO2 emissions, which currently account for 2% of global industrial fossil CO2 emissions, and it has an opportunity to produce renewable energy, fuels, and materials for other sectors. The purpose of this study is to improve understanding of the decarbonization prospects of the PPI. The study provides insights on the magnitude of needed annual renewal rates for several possible net-zero target years of industrial fossil CO2 emissions in the PPI and discusses decarbonization opportunities, namely, energy and material efficiency improvement, fuel switching, electrification, renewable energy production, carbon capture, and new products. The effects of climate policies on the decarbonization opportunities are critically evaluated to provide an overview of the current and future business environment of the European PPI. The focus is on Europe, but other regions are analyzed briefly to widen the view. The analysis shows that there are no major technical barriers to the fossil-free operation of the PPI, but the sector renovates slowly, and many new opportunities are not implemented on a large scale due to immature technology, poor economic feasibility, or unclear political environment. Contact information: LUT School of Energy Systems, LUT University, 53850 Lappeenranta, Finland Energies 2023, 16, 746. https://doi.org/10.3390/en16020746 Creative Commons Attribution 4.0 International License

Article 3 – Decarbonisation

Page 1 of 19

Page 1 Page 2 Page 3 Page 4 Page 5 Page 6 Page 7 Page 8 Page 9 Page 10 Page 11 Page 12 Page 13 Page 14 Page 15 Page 16 Page 17 Page 18 Page 19 Page 20 Page 21 Page 22 Page 23 Page 24 Page 25 Page 26 Page 27 Page 28 Page 29 Page 30 Page 31 Page 32 Page 33 Page 34 Page 35 Page 36 Page 37 Page 38 Page 39 Page 40 Page 41 Page 42 Page 43 Page 44 Page 45 Page 46 Page 47 Page 48 Page 49 Page 50 Page 51 Page 52 Page 53 Page 54 Page 55 Page 56 Page 57 Page 58 Page 59 Page 60 Page 61 Page 62 Page 63 Page 64 Page 65 Page 66 Page 67 Page 68 Page 69 Page 70 Page 71 Page 72 Page 73 Page 74 Page 75 Page 76 Page 77 Page 78 Page 79 Page 80 Page 81 Page 82 Page 83 Page 84 Page 85 Page 86 Page 87 Page 88 Page 89 Page 90 Page 91 Page 92 Page 93 Page 94 Page 95 Page 96 Page 97 Page 98 Page 99 Page 100 Page 101 Page 102 Page 103 Page 104 Page 105 Page 106 Page 107 Page 108 Page 109 Page 110 Page 111 Page 112 Page 113 Page 114 Page 115 Page 116 Page 117 Page 118 Page 119 Page 120 Page 121 Page 122 Page 123 Page 124 Page 125 Page 126 Page 127 Page 128 Page 129 Page 130 Page 131 Page 132 Page 133 Page 134 Page 135 Page 136 Page 137 Page 138 Page 139 Page 140 Page 141 Page 142 Page 143 Page 144 Page 145 Page 146 Page 147 Page 148 Page 149 Page 150 Page 151 Page 152 Page 153 Page 154 Page 155 Page 156 Page 157 Page 158 Page 159 Page 160 Page 161 Page 162 Page 163 Page 164 Page 165 Page 166 Page 167 Page 168 Page 169 Page 170 Page 171 Page 172 Page 173 Page 174 Page 175 Page 176 Page 177 Page 178 Page 179 Page 180 Page 181 Page 182 Page 183 Page 184 Page 185 Page 186 Page 187 Page 188 Page 189 Page 190 Page 191 Page 192 Page 193 Page 194 Page 195 Page 196 Page 197 Page 198 Page 199 Page 200

Made with FlippingBook - professional solution for displaying marketing and sales documents online