1. An open type intelligent control system for material moisture of a fluidized bed based on fuzzy PID (proportion integration differentiation), which is characterized by comprising a fluidized bed parameter setting and monitoring subsystem (1), a complete machine control subsystem (2), a moisture detection and on-line analysis component (3), a slurry control and supply component (4) and a fluidized bed host (5),

parameter setting and monitoring subsystem (1): used for setting system parameters and control strategies;

the whole machine control subsystem (2): the slurry control and supply component (4) is used for carrying out fuzzy PID control according to the set parameters and sending the result to the slurry control and supply component;

moisture detection and on-line analysis assembly (3): the system is used for detecting the material moisture value of the fluidized bed host (5) and transmitting the material moisture value to the whole machine control subsystem (2);

slurry control and supply assembly (4): the device is used for changing the slurry supply speed of the fluidized bed main machine (5) so as to realize accurate control of the moisture of the boiling materials in the fluidized bed main machine (5).

2. The fuzzy PID based fluidized bed material moisture open type intelligent control system according to claim 1, wherein the algorithm of the fuzzy PID control comprises the following steps:

step 1, designing a fuzzy controller: taking a difference E between an actual measured value of the moisture of the material to be taken and a moisture target value and a change EC of the difference as input of a fuzzy controller, and taking three parameters of delta Kp, delta Ti and delta Td as output of the fuzzy controller;

step 2, language variable selection: if desired, 7 linguistic variables are preselected to represent the fuzzy subset of fuzzy controller inputs and outputs, namely: negative big [ NB ], negative middle [ NM ], negative small [ NS ], zero [ ZO ], positive small [ PS ], positive middle [ PM ], positive big [ PB ];

and 3, selecting a membership function: fuzzy subsets at two ends select a trapezoidal membership function, and a fuzzy subset in the middle selects a triangular membership function;

step 4, determining a fuzzy rule table: determining three fuzzy rule tables according to the regulation rule of the PID parameters, programming by adopting a maximum and minimum value reasoning method during fuzzy rule reasoning to obtain a fuzzy relation, and calculating a fuzzy output set;

step 5, a fuzzy solving method: an inflection point gravity center method is adopted as a fuzzy solution method of a fuzzy PID algorithm, a triangular membership function graph is used in the precision process, inverse function calculation is carried out according to a defined membership programming formula, a corresponding fuzzy solution programming formula can be obtained, so that an inflection point is calculated, and a precision value is calculated by using the gravity center method;

step 6, PID parameter calculation: the accurate value is the change value of three parameters of PID, and the sum calculation is carried out on the change value and the initial set value, so that the current PID parameter can be obtained, and the fuzzy PID closed-loop control is realized; namely:

Kp₀+△Kp＝Kp，Ti₀+△Ti＝Ti，Td₀+△Td＝Td。

wherein: kp₀、Ti₀And Td₀The initial PID parameters are delta Kp, delta Ti and delta Td, the change values of the three PID parameters are Kp, delta Ti and delta Td, and the current PID parameters are Kp, delta Ti and delta Td.

3. The fluidized bed material moisture open type intelligent control system based on the fuzzy PID as claimed in claim 2, wherein the material moisture target value adopts a fixed value or a slope change value or an experimental fitting curve.

4. The fuzzy PID based open intelligent control system for fluidized bed material moisture according to claim 3, wherein the ramp value is calculated according to the following formula:

wherein t is the timing after the moisture control is started, PV₀Material moisture at the initial time, PV is target moisture value at t time, PV_endThe final moisture set point, T, to be reached_ATo achieve PV_endThe total time required.

5. The fuzzy PID-based fluidized bed material moisture open type intelligent control system according to any one of claims 1-4, wherein the parameter setting and monitoring subsystem (1) comprises a touch screen, a memory card and a communication module, the touch screen is used for setting parameters and control strategies of fuzzy PID control, the memory card is used for storing the parameters and the control strategies, and the communication module is used for communicating with the complete machine control subsystem (2).

6. The fluidized bed material moisture open type intelligent control system based on the fuzzy PID as claimed in claim 5, characterized in that the moisture detection and on-line analysis component (3) adopts a near infrared analysis system to collect spectrum and analyze material moisture value in real time.

Background

In the fluidized bed one-step granulation process, the moisture content of the material is crucial to the final quality of the product. In the prior production, the moisture of the material is only used as a reference during sampling, and no control is added. In recent years, although many manufacturers introduce online moisture detection equipment to detect moisture in fluidized bed materials, the moisture detection equipment is only used as a judgment basis for operators, and automatic production control is not achieved.

In the fluidized bed granulation, the control of the moisture of the material is more complex compared with other control objects, and on one hand, the hysteresis quality and the inertia are larger; on the other hand, the state of the material is greatly affected by moisture, and the material may collapse due to an excessively high liquid spraying speed. In the control process, the moisture can be expected to better track a target curve, particularly the moisture of the initial material is inconsistent with the deviation of the set moisture value, and the traditional PID adjustment cannot achieve the control effect of small overshoot while achieving the set value quickly and accurately.

For some materials, the moisture of the material does not have a definite relation with the final product quality, the material is always constant at a certain numerical value and is not the best effect, and the moisture value of the material is often required to follow a certain track to obtain a better product, so that the tracking design of realizing a specific moisture track is a big problem.

Disclosure of Invention

The technical problem to be solved by the invention is as follows: the defects of the prior art are overcome, and the open type intelligent control system for the material moisture of the fluidized bed based on the fuzzy PID is provided, so that the real-time collection and monitoring of the material moisture of the fluidized bed can be realized; meanwhile, for the granulating process, an expected moisture target track can be designed, and the moisture can be accurately controlled.

The invention relates to a fluidized bed material moisture open type intelligent control system based on fuzzy PID, which comprises a fluidized bed parameter setting and monitoring subsystem, a complete machine control subsystem, a moisture detection and on-line analysis component, a slurry control and supply component and a fluidized bed host,

the parameter setting and monitoring subsystem: used for setting system parameters and control strategies;

the whole machine control subsystem: the slurry control and supply component is used for carrying out fuzzy PID control according to set parameters and sending results to the slurry control and supply component;

moisture detection and on-line analysis assembly: the system is used for detecting the material moisture value of the fluidized bed host and transmitting the material moisture value to the complete machine control subsystem;

slurry control and supply assembly: the device is used for changing the slurry supply speed of the fluidized bed main machine, so that the moisture of the boiling materials in the fluidized bed main machine is accurately controlled.

The real-time collection and monitoring of the material moisture of the fluidized bed can be realized; meanwhile, for the granulating process, an expected moisture target track can be designed, and the moisture can be accurately controlled. By adopting the open design, the algorithm parameters and the tracking track model can be modified according to the actual process, the control precision is high, and the requirement of a complex industrial field is met.

Preferably, the algorithm of the fuzzy PID control comprises the following steps:

step 1, designing a fuzzy controller: taking a difference E between an actual measured value of the moisture of the material to be taken and a moisture target value and a change EC of the difference as input of a fuzzy controller, and taking three parameters of delta Kp, delta Ti and delta Td as output of the fuzzy controller;

excessive calculation is avoided, and when an actual control algorithm is written, the two-input three-output fuzzy controller is decomposed into three two-input one-output fuzzy controllers, so that the complexity of the algorithm is reduced.

Step 2, language variable selection: if desired, 7 linguistic variables are preselected to represent the fuzzy subset of fuzzy controller inputs and outputs, namely: negative big [ NB ], negative middle [ NM ], negative small [ NS ], zero [ ZO ], positive small [ PS ], positive middle [ PM ], positive big [ PB ];

and 3, selecting a membership function: fuzzy subsets at two ends select a trapezoidal membership function, and a fuzzy subset in the middle selects a triangular membership function;

step 4, determining a fuzzy rule table: determining three fuzzy rule tables according to the regulation rule of the PID parameters, programming by adopting a maximum and minimum value reasoning method during fuzzy rule reasoning to obtain a fuzzy relation, and calculating a fuzzy output set;

step 5, a fuzzy solving method: an inflection point gravity center method is adopted as a fuzzy solution method of a fuzzy PID algorithm, a triangular membership function graph is used in the precision process, inverse function calculation is carried out according to a defined membership programming formula, a corresponding fuzzy solution programming formula can be obtained, so that an inflection point is calculated, and a precision value is calculated by using the gravity center method;

the control output quantity is smoother, and the problem of overlarge calculation quantity is avoided.

Step 6, PID parameter calculation: the accurate value is the change value of three parameters of PID, and the sum calculation is carried out on the change value and the initial set value, so that the current PID parameter can be obtained, and the fuzzy PID closed-loop control is realized; namely:

Kp₀+△Kp＝Kp，Ti₀+△Ti＝Ti，Td₀+△Td＝Td。

wherein: kp₀、Ti₀And Td₀The initial PID parameters are delta Kp, delta Ti and delta Td, the change values of the three PID parameters are Kp, delta Ti and delta Td, and the current PID parameters are Kp, delta Ti and delta Td.

Preferably, the material moisture target value adopts a fixed value or a slope change value or an experimental fitting curve.

According to actual needs, the target value of the moisture can adopt a fixed value, a slope change value or a better moisture curve obtained according to experiments, and the three parts can be used independently or can be combined.

Preferably, the ramp value is calculated according to the following formula:

wherein t is the timing after the moisture control is started, PV₀Material moisture at the initial time, PV is target moisture value at t time, PV_endThe final moisture set point, T, to be reached_ATo achieve PV_endThe total time required.

Preferably, the parameter setting and monitoring subsystem comprises a touch screen, a memory card and a communication module, the touch screen is used for setting parameters and control strategies of fuzzy PID control, the memory card is used for storing the parameters and the control strategies, and the communication module is used for communicating with the whole machine control subsystem.

Preferably, the moisture detection and on-line analysis component adopts a near infrared analysis system to collect the spectrum in real time and analyze the moisture value of the material.

The collection effect is good.

Compared with the prior art, the invention has the following beneficial effects:

1. the fluidized bed material moisture open type intelligent control system based on the fuzzy PID can detect the moisture value in real time and accurately control the moisture as the controlled quantity. The moisture control system is a system with large inertia, and due to the characteristics of the material, the closed-loop control cannot have large overshoot. Therefore, a fuzzy PID control algorithm is introduced, and the problem is well solved.

2. The conventional control mode only controls the controlled quantity as a fixed value, which is also the characteristic of most control occasions. However, the moisture of the material is directly increased to a higher level in a short time by the controlled object, which causes the material to collapse on one hand, and on the other hand, the preparation effect of certain materials is not well influenced, and the moisture rises along a certain curve in actual control. The invention sets a plurality of control strategies, can select a fixed value and a slope curve as required, and can also track the first-order, second-order and third-order system curves according to the fitted transfer function. And all parameters can be set on the operation screen, and an open design mode is adopted.

3. And (3) combining related technologies such as advanced control algorithms and transfer functions with the PLC. In the programming process of the fuzzy PID control algorithm and the control strategy, a large number of matrix calculation processes are introduced. The traditional method is to combine a controller with an industrial personal computer and control by using a LabVIEW, MATLAB or a special fuzzy controller. The invention adopts an optimized structure for programming, and most of the work is in the aspect of software design. On one hand, the cost is greatly reduced, on the other hand, a parameter setting interface is opened, a third-party product is not integrated, and the stability is greatly improved.

4. A great deal of work is done on data storage, real-time data acquisition and curve display. When the system runs, on one hand, the current target value curve and the current actual value curve can be observed and known in real time on the operation screen, and meanwhile, the generated data is stored, so that the traceability of the production data is ensured.

Drawings

FIG. 1 is a schematic block diagram of an open intelligent control system for material moisture of a fluidized bed based on fuzzy PID according to the invention;

FIG. 2 is a block diagram of the fuzzy PID control of the present invention;

FIG. 3 is a block diagram of a moisture target control strategy according to the present invention;

FIG. 4 is a flow chart of the operation of the fuzzy PID-based open intelligent control system for the material moisture of the fluidized bed;

FIG. 5 is a schematic diagram of the membership level of the present invention.

Wherein: 1. a parameter setting and monitoring subsystem; 2. a complete machine control subsystem; 3. a moisture detection and online analysis component; 4. a slurry control and supply assembly; 5. a fluidized bed main machine.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Example 1

As shown in figures 1-4, the fluidized bed material moisture open type intelligent control system based on the fuzzy PID of the invention comprises a fluidized bed parameter setting and monitoring subsystem 1, a complete machine control subsystem 2, a moisture detection and on-line analysis component 3, a slurry control and supply component 4 and a fluidized bed host 5,

parameter setting and monitoring subsystem 1: used for setting system parameters and control strategies;

the whole machine control subsystem 2: the slurry control and supply component 4 is used for carrying out fuzzy PID control according to set parameters and sending the result to the slurry control and supply component;

moisture detection and on-line analysis assembly 3: the system is used for detecting the material moisture value of the fluidized bed host 5 and transmitting the material moisture value to the whole machine control subsystem 2;

slurry control and supply assembly 4: the device is used for changing the slurry supply speed of the fluidized bed main machine 5, so that the moisture of the boiling materials in the fluidized bed main machine 5 can be accurately controlled.

The parameter setting and monitoring subsystem 1 adopts an HMI (human machine interface) for setting fuzzy control open parameters, a moisture target track and corresponding error values; the whole machine control subsystem 2 adopts PLC, and a fuzzy control program, a PID closed-loop program and a control strategy standard program block are integrated in the PLC; the moisture detection and on-line analysis component 3 adopts a near infrared analysis system to collect a spectrum in real time and analyze a moisture value of the material; the slurry control and supply assembly 4 consists of a peristaltic pump pipeline, wherein the peristaltic pump can be controlled by a frequency converter; the fluidized bed host machine 5 is used for storing materials, the whole machine control subsystem 2 carries out fuzzy PID control according to set parameters, the result is transmitted to the slurry control and supply assembly 4, and the moisture of the boiling materials in the fluidized bed is accurately controlled by changing the rotating speed of the peristaltic pump.

The parameter setting and monitoring subsystem 1 comprises a touch screen, a memory card and a communication module, wherein the touch screen is used for setting parameters and control strategies of fuzzy PID control, the memory card is used for storing the parameters and the control strategies, and the communication module is used for communicating with the whole machine control subsystem 2.

The real-time collection and monitoring of the material moisture of the fluidized bed can be realized; meanwhile, for the granulating process, an expected moisture target track can be designed, and the moisture can be accurately controlled. By adopting the open design, the algorithm parameters and the tracking track model can be modified according to the actual process, the control precision is high, and the requirement of a complex industrial field is met.

Firstly, algorithm design of the fuzzy PID control:

1. designing a fuzzy controller: taking a difference E between an actual measured value of the moisture of the material to be taken and a moisture target value and a change EC of the difference as input of a fuzzy controller, and taking three parameters of delta Kp, delta Ti and delta Td as output of the fuzzy controller;

excessive calculation is avoided, and when an actual control algorithm is written, the two-input three-output fuzzy controller is decomposed into three two-input one-output fuzzy controllers, so that the complexity of the algorithm is reduced.

Fuzzy PID algorithm relates to fuzzy control of three parameters, and a rule table and knowledge base information are stored in an array, including matrix operation; the control strategy part relates to the contents of automatic control fields such as transfer functions and the like, and the calculation process is more complicated. The two parts directly use the SCL language, the combination of a complex algorithm and industrial control can be realized, the intervention of third-party software such as LabVIEW, MATLAB and the like is avoided, and a control system is simpler.

2. Language variable selection: if desired, 7 linguistic variables are preselected to represent the fuzzy subset of fuzzy controller inputs and outputs, namely: negative big [ NB ], negative middle [ NM ], negative small [ NS ], zero [ ZO ], positive small [ PS ], positive middle [ PM ], positive big [ PB ];

3. selection of membership function: fuzzy subsets at two ends select a trapezoidal membership function, and a fuzzy subset in the middle selects a triangular membership function;

fuzzy subsets corresponding to both ends are for example: NB and PB, selecting a trapezoidal membership function. A triangular membership function is selected corresponding to the intermediate fuzzy subset. The membership degree diagram is shown in FIG. 5:

the programming formula of the left end trapezoidal membership function is as follows:

the programming formula of the intermediate triangle membership function is as follows:

the right end trapezoidal membership function programming formula is as follows:。

4. determining a fuzzy rule table: determining three fuzzy rule tables according to the regulation rule of the PID parameters, programming by adopting a maximum and minimum value reasoning method during fuzzy rule reasoning to obtain a fuzzy relation, and calculating a fuzzy output set;

the output of the fuzzy controller directly influences the proportional coefficient, the integral time and the differential time in the PID, so that three fuzzy rules are determined according to the regulation rule of PID parameters, namely a table 1:

TABLE 1

And programming by adopting a maximum and minimum value reasoning method during fuzzy rule reasoning to obtain a fuzzy relation and calculating a fuzzy output set.

5. The method for deblurring comprises the following steps: an inflection point gravity center method is adopted as a fuzzy solution method of a fuzzy PID algorithm, a triangular membership function graph is used in the precision process, inverse function calculation is carried out according to a defined membership programming formula, a corresponding fuzzy solution programming formula can be obtained, so that an inflection point is calculated, and a precision value is calculated by using the gravity center method;

the commonly used precision process methods are: a maximum membership method, a center of gravity method, and a weighted average method. In order to make the control output quantity smoother and avoid the problem of overlarge calculated quantity, the system adopts an inflection point gravity center method as a fuzzy solution method of a fuzzy PID algorithm. In the precision process, a triangular membership function graph is used, inverse function calculation is carried out according to a defined membership programming formula, a corresponding fuzzy solution programming formula can be obtained, so that an inflection point is calculated, and a gravity center method is used for calculating a precision value.

6. And (3) PID parameter calculation: the accurate value is the change value of three parameters of PID, and the sum calculation is carried out on the change value and the initial set value, so that the current PID parameter can be obtained, and the fuzzy PID closed-loop control is realized; namely:

Kp₀+△Kp＝Kp，Ti₀+△Ti＝Ti，Td₀+△Td＝Td。

wherein: kp₀、Ti₀And Td₀The initial PID parameters are delta Kp, delta Ti and delta Td, the change values of the three PID parameters are Kp, delta Ti and delta Td, and the current PID parameters are Kp, delta Ti and delta Td.

And taking the difference E between the actual measured value of the moisture of the material to be taken and the set value of the moisture and the change EC of the difference as the input of the fuzzy controller, and taking three parameters of delta Kp, delta Ti and delta Td as the output of the fuzzy controller. Selecting a trapezoidal membership function corresponding to fuzzy subsets at two ends in the fuzzification process of the fuzzy controller; selecting a triangular membership function corresponding to the middle fuzzy subset; in the inference mechanism of the fuzzy rule, a maximum minimum value method is adopted; and (4) solving a required numerical value through an inflection point gravity center method according to the calculated fuzzy result, and adding the numerical value and a preset value of a screen to obtain a parameter which is currently input into PID closed-loop control, so as to complete closed-loop regulation of the moisture.

Second, design of control strategy scheme

The material moisture target value adopts a fixed value or a slope change value or an experimental fitting curve.

According to actual needs, the target value of the moisture can adopt a fixed value, a slope change value or a better moisture curve obtained according to experiments, and the three parts can be used independently or can be combined.

1. Fixed value moisture target value

The constant value moisture value is simple, and a corresponding fixed value is set on the HMI human-machine interface.

2. Ramping the moisture target value

When the actual moisture value is required to increase along a straight line with a certain slope, the formula is:

wherein t is the timing after the moisture control is started, PV₀Material moisture at the initial time, PV is target moisture value at t time, PV_endThe final moisture set point, T, to be reached_ATo achieve PV_endThe total time required.

3. Experiment fitting curve water target value

Corresponding to other curves, for example, certain materials, a pharmaceutical factory research and development personnel can obtain a moisture target curve through a large amount of experiments, a moisture transfer function can be obtained through a curve fitting mode, and an operator only needs to change related model parameters on an HMI (human machine interface). The PLC program of the system is internally programmed, and the idea is to program conventional first-order, second-order and third-order transfer functions and reserve parameter interfaces thereof.

First order system transfer function:

second order system transfer function:

third order system transfer function:

in the above formula, K, Tp₁，Tp₂，Tp₃May be modified on the touch screen according to the fitted curve equation.

The water set value can be singly selected to be a fixed value, a slope or a special fitting curve according to actual control requirements, and various curves can be combined and arranged according to time and sequence. The special fitting curve can adopt first-order, second-order and third-order function curves, and the program of each function curve is programmed into the controller.

The control system consists of an S71500 PLC, an external auxiliary circuit and a control circuit, wherein the whole machine control system is compiled in a CPU, the control strategy part and the calculation part of the fuzzy PID all use SCL language, and optimized block access is used for design so as to improve the running efficiency of the program.

Third, the moisture detection and on-line analysis component 3 and the slurry control and supply component 4

1. The system adopts near infrared probe to collect material spectrum, and uses the spectrum analysis software mounted on personal computer to make analysis and prediction of current material moisture value. And transmitting the data to an HMI (human machine interface), displaying the current moisture curve in real time, and simultaneously, continuously acquiring a moisture value in the touch screen by the PLC to participate in closed-loop calculation.

2. The accurate output value calculated by the fuzzy PID controller sends a message instruction to the G120 frequency converter in a PROFINET communication mode, and then the rotating speed of the peristaltic pump is controlled, so that the control of the moisture is realized. Because the fluidized bed material is greatly influenced by instantaneous moisture, the HMI human-computer interface can limit the pump speed during actual operation so as to avoid the phenomenon of bed collapse caused by overhigh pump speed.

Four, parameter visualization open design

1. Visualization of fuzzy algorithm parameters

The system is preset with an initial fuzzy controller algorithm, and can be directly modified on an HMI (human machine interface) when the system is actually required to be modified. For example, the range of the fuzzy subset, the fuzzy rule and the like can be set on the operation interface, so that the poor control effect caused by the mismatching of the model or the rule is avoided. Meanwhile, the model parameters can be corrected in real time in the debugging stage according to the actual moisture curve, so that a large amount of time is saved.

2. Visualization of control strategy parameters

The fixed value control system is a system with a great proportion of the current industrial control occasions, but corresponding to some occasions, if a controlled quantity is required to track a preset section of curve, the traditional method cannot be realized. The control strategies of a fixed value, a slope and a first-order, a second-order and a third-order system are integrated in the program in advance, and a user can set related parameters on an HMI (human machine interface) according to needs, so that the combination and the predefinition of the target track can be realized.

3. Full automation design

The system is designed by Siemens Bomb diagram software and integrates a full-automatic program. The automatic production operation can be realized by adopting a formula to set process parameters in advance and matching with an open fuzzy PID controller algorithm and the preset function of a target track.

Fifth, the operation process

1. And setting parameters. Before the system runs, initial parameters of the fuzzy PID controller are set in advance, wherein the initial parameters comprise a proportional coefficient, an integral time and a differential time, and some data in a knowledge base and a rule base in a fuzzy algorithm. Meanwhile, corresponding to the target track, a suitable value is selected, such as fixed value tracking or slope tracking, or some curve is fitted.

2. And (5) operating the system. When the system is in operation, air is filtered and heated by the air inlet processing unit to blow up materials in the fluidized bed main machine 5, so that the materials are kept in a boiling state. The hot air is discharged from the exhaust system after passing through the material. During the water closed-loop control, the water content detection and analysis component continuously detects the water content of the current material, transmits the value to the control system, and adjusts the rotating speed of the peristaltic pump after the fuzzy PID algorithm so as to adjust the current water content value to be consistent with the target set value.

3. And monitoring and collecting data. The system is provided with data acquisition and overrun alarm functions, a data acquisition period can be set, and key process parameters can be stored in a storage medium at the rear part of the screen according to a preset period. Meanwhile, the screen has a history curve, and the control effect of the parameters is displayed in an intuitive mode.

The invention relates to a fluidized bed material moisture open type intelligent control system based on fuzzy PID, which comprises the following working processes:

after the equipment is started, setting each initial PID parameter through an HMI (human machine interface), and selecting a required moisture target track according to the requirement; the granulation recipe is edited and called and the automatic production is started. The material in the fluidized bed host 5 is in a flowing state under the action of hot air, after entering an injection stage, the moisture detection and online analysis component 3 continuously detects the moisture of the material and transmits the moisture to a PLC control program, the inside of the program continuously compares a set value with a current value of a target track and the change rate of the difference to obtain delta Kp, delta Ti and delta Td through fuzzy calculation, and the delta Kp, the delta Ti and the delta Td are summed with an initial PID parameter to refresh the current PID parameter so as to participate in PID control. In the production process, particularly in the early debugging stage, the parameters in the fuzzy algorithm and the target track are changed without restarting the production, and the parameters can be adjusted in real time in the running process.