Predictive control system and method for brown stock washing treatment in a pulp mill

Abstract

Systems and methods are provided for predictive control of brown stock handling at a pulp mill. Various online sensors produce output signals representative of actual values of respective process characteristics, each of which is directly or indirectly affected by adjustments to a corresponding process variable. A controller uses the output signals or related measurement data to dynamically set target values for the process characteristics based on predicted effects of control responses to the corresponding process variables. The controller further produces control signals to actuators associated with the corresponding process variables based on detected differences between the respective actual values and the target values. Exemplary brown stock washing control systems can optimize various types of brown stock washing configurations, including, for example, vacuum drum washers, compaction baffle washers, chemical washers, direct displacement washers, and wash presses. Cloud-based analysis and machine learning can also be implemented to improve control algorithms over time.

Description

Technical Field

[0001] This invention generally relates to control systems and methods for pulp mills. More particularly, this invention relates to supervised and stand-alone (offline) control systems and methods for brown stock washing across a variety of potential configurations. Background Technology

[0002] Conventional pulp washing systems in pulp mills are designed to separate pulp from black liquor. Water is used in countercurrent washing systems to facilitate the washing of black liquor from the pulp. Pulp mill operators face the challenge of balancing improved separation with the costs associated with using water for separation. Process variability from a range of factors, including, for example, temperature, pH, conductivity, kappa, wood species, consistency, entrained air, fiber freeness, soap concentration, residual alkali, etc., makes it difficult to maintain peak washing efficiency. The result is reduced efficiency, leading to higher manufacturing costs or lost revenue from lower production rates.

[0003] The most commonly used technology in the pulp and paper industry is passive (reactive). The spray flow rate in the coarse pulp washing process is adjusted based on the conductivity of the filtrate at the final washing stage (stage). Course control is achieved by manually adjusting the tank dilution rate on the washer through operator adjustment of the tank level (liquid level). Defoamers and filter aids can be proportionally controlled based on the production rate, with manual adjustments by the operator in response to changes in tank level, washer speed, and conductivity.

[0004] An inherent limitation of conventional methods is that they are inherently passive. These techniques respond to shifts in the balance of the washing system and often rely on manual operator intervention. Those skilled in the art will appreciate that changes made online by the operator can actually lead to further process variations.

[0005] The goal is to provide comprehensive process control of the slurry washing process, allowing for proactive (pre-acting) control of the slurry washing system, either substantially in real time or at any given selected time. However, the inherent dynamic nature of the slurry washing process has traditionally made predictive analysis and associated process corrections extremely difficult and unreliable. Summary of the Invention

[0006] According to the various embodiments disclosed herein, the overall process control objective of the coarse pulp washing process described above is achieved by maintaining the optimal efficiency of each washing machine, thereby resulting in more stable coarse pulp washing process control and improved washing efficiency.

[0007] In short, the systems and methods disclosed herein reduce process variability and further optimize key operating parameters such as washer speed, spray flow rate, dilution ratio, and chemical feed (e.g., defoamers and filter aids). An exemplary stand-alone control system for coarse slurry washing incorporates sensing devices (e.g., The entrained air monitor and controller are integrated into a single digital support package, which can be integrated into and / or implemented alongside the pulp mill's distributed control system (DCS). Components of this control system may include the controller, entrained air monitor and other instruments, telemetry technology for remote access to the controller, and a cloud-based analytics engine that provides adaptive algorithms for controlling variables in the pulp washing process and chemical additives. The DCS can be used to acquire inputs from the control algorithms, defoamer feed skid, and filter aid feed skid.

[0008] The coarse pulp washing control system and method can preferably be optimized for various types of coarse pulp washing configurations, including but not limited to systems with vacuum drum washers, compaction baffle washers, chemical washers, direct displacement washers, horizontal belt washers, pressure diffusers, and washing presses.

[0009] In a specific embodiment of a system as disclosed herein for predictive control of rough pulp processing at a pulp mill, multiple online sensors are configured to generate output signals representing actual values ​​of corresponding process characteristics. For example, a first online sensor generates an output signal representing an actual value of a first process characteristic directly affected by adjustments to at least a first process variable, and a second online sensor generates an output signal representing an actual value of a second process characteristic directly affected by adjustments to at least a second process variable. The second process variable is at least indirectly affected by adjustments to the first process variable. At least first and second actuators are configured to adjust the actual values ​​of the first and second process variables, respectively. A controller acquires output signals from the multiple online sensors and determines measurement data corresponding to the actual values ​​of at least the first and second process characteristics. The controller dynamically sets target values ​​for the first and second process characteristics, respectively, based on the predicted effects of corresponding control responses to at least the first and second process variables. The controller then generates control signals to one or more actuators associated with at least the first and second process variables based on the detected difference between the corresponding actual values ​​and the target values.

[0010] In one exemplary aspect of the foregoing implementation, each of the corresponding first and second process variables has a corresponding optimal range or threshold level. Target values ​​are then dynamically set for the first and second process characteristics, respectively, based on the optimal range or threshold level and the predicted impact of the corresponding control responses to at least the first and second process variables.

[0011] In another exemplary aspect, the user interface is generated in association with a user computing device, and the controller provides acquired signals or measurement data corresponding to actual values ​​of at least the first and second process characteristics for display via the user interface. The user interface may further enable the user to specify one or more optimal ranges or threshold levels corresponding to the first and second process variables, respectively.

[0012] In another exemplary aspect, the first sensor may be an entrained air sensor (e.g., provided by the applicant) configured to generate an output signal representing the level of entrained air as a first process characteristic. (unit), and the first process variable corresponds to the defoamer flow rate.

[0013] In another exemplary aspect, the second sensor is a washing machine speed sensor configured to generate an output signal representing the washing machine speed as a second process characteristic, and the second process variable corresponds to the drum dilution rate.

[0014] The second sensor may alternatively be a washing machine speed sensor configured to generate an output signal representing the washing machine speed as a second process characteristic, and the second process variable corresponds to the defoamer flow rate.

[0015] The second sensor may alternatively be a washing machine speed sensor configured to generate an output signal representing the washing machine speed as a second process characteristic, and the second process variable corresponds to the defoamer pump speed and the drum dilution rate.

[0016] The second sensor may alternatively be a flow meter configured to generate an output signal representing the dilution flow rate to the washing machine as a second process characteristic, and the second process variable corresponds to the defoamer pump speed.

[0017] In another exemplary aspect, the plurality of online sensors further includes a third online sensor that generates an output signal representing an actual value of a third process characteristic, which is directly affected by adjustments to at least a third process variable, and which is at least indirectly affected by adjustments to the first and second process variables. The controller further dynamically sets a target value for the third process characteristic based on the predicted effects of corresponding control responses to at least the first, second, and third process variables, and thus generates a control signal to a third actuator based on the difference between the detected actual value and the target value of the third process characteristic.

[0018] In another exemplary aspect: a first sensor generates an output signal representing the level of entrained air as a first process characteristic, and the first process variable corresponds to the defoamer flow rate; a second sensor generates an output signal representing the washing machine speed as a second process characteristic, and the second process variable corresponds to the drum dilution rate; a third sensor is a liquid solids meter configured to generate an output signal representing the level of liquid solids as a third process characteristic, and the third process variable corresponds to the spray flow rate.

[0019] In another exemplary aspect, in which the system further includes multiple cascaded filtrate tanks and corresponding multiple washer streams and spray flows, a liquid-solid meter is provided in association with a first filtrate tank among the multiple cascaded filtrate tanks. Based on the difference between the actual and target values ​​of the detected third process characteristic, the controller generates a control signal to a third actuator for the spray flow rate associated with the first washer stream.

[0020] In another exemplary aspect, a conductivity sensor is provided to measure an actual conductivity value, wherein the actual conductivity value is influenced by a plurality of process characteristics including first, second, and third process characteristics, and by a plurality of process variables including first, second, and third process variables. The controller further dynamically sets target values ​​for the first, second, and third process characteristics based on the predicted influence of the corresponding control response relative to the optimal conductivity value, and further generates control signals to the first, second, and third actuators based on the dynamically set target values.

[0021] In short, such an implementation (further, for example, according to one or more exemplary aspects as described above) enables the use of machine learning environments to create dynamic algorithms for pulp mill operators or third-party managers to address at least pulp cleanliness and manufacturing costs. Unlike simple control algorithms, this system allows for the implementation of multivariate algorithms based on individual processes.

[0022] The benefits of such a comprehensive control system for coarse pulp washing can involve the optimization of washing efficiency, which can further bring a variety of benefits to pulp mill operators, such as: increased pulp yield; increased paper yield; reduced bleaching costs; reduced paper machine chemical costs; reduced defoamer and filter aid usage; reduced energy costs; reduced soda loss (soda replenishment); and / or improved pulp quality.

[0023] Many objects, features and advantages of the embodiments set forth herein will become apparent to those skilled in the art when the following disclosure is read in conjunction with the accompanying drawings. Attached Figure Description

[0024] Figure 1 This is a block diagram illustrating an implementation of the predictive control system as disclosed herein.

[0025] Figure 2 This is a flowchart illustrating an implementation of the predictive control method disclosed herein. Detailed Implementation

[0026] Although various embodiments of the invention and their use are discussed in detail below, it should be understood that the invention provides many applicable inventive concepts that can be embodied in a wide variety of specific contexts. The specific embodiments discussed herein are merely illustrative of particular ways of carrying out and using the invention and do not define the scope of the invention.

[0027] The following detailed description of embodiments of the present disclosure is provided with reference to one or more accompanying drawings. The drawings are provided for the purpose of explaining the present disclosure and are not intended to be limiting. Those skilled in the art will understand that various modifications and variations can be made to the teachings of the present disclosure without departing from its scope. For example, features shown or described as part of one embodiment may be used with another embodiment to produce yet another embodiment.

[0028] This disclosure is intended to cover such modifications and variations that fall within the scope of the appended claims and their equivalents. Other objects, features, and aspects of this disclosure are disclosed in the following detailed description. Those skilled in the art will understand that this discussion is merely a description of exemplary embodiments and is not intended to limit the broader aspects of this disclosure.

[0029] Throughout the specification and claims, unless the context otherwise indicates, the following terms will have at least the meaning explicitly associated herein. The meanings confirmed below are not intended to limit the terms, but are merely illustrative examples. The meanings of “a,” “an,” and “the” may include plural pronouns, and the meaning of “(in)” may include “(in)” and “(on)”. The phrase “in one embodiment” as used herein does not necessarily refer to the same embodiment, although this may be the case.

[0030] First refer to Figure 1 Regarding the coarse pulp handling system and processes in a pulp mill, an implementation of a predictive control system 100, as disclosed herein, is provided. As described in detail below, this system includes a controller that receives process information from various sensors and relays (e.g., air gauges, refractometers, Coriolis mass meters, flow meters, thermocouples, consistency transmitters, etc.). This data is used as input to a dynamic process control algorithm to provide recommended control outputs for multiple process parameters in coarse pulp washing, including, for example, spray flow rate, filter aid flow rate, defoamer flow rate, and dilution factor. Telemetry enables cloud-based real-time analytics, continuous visibility for operators and managers, and remote fine-tuning of control logic to maintain control loop health, anomaly detection, and alarms. Data sent to the cloud can be integrated with machine learning environments to customize individual control loops and continuously enhance those loops.

[0031] Implementations of production stage 110 may include various system components associated with process variable 112 and process characteristic 114. As used herein, each of the various process characteristics can be directly affected by adjustments to at least one of the process variables. For example, those skilled in the art will appreciate that the amount of entrained air (as a process characteristic) can be directly affected by adjustments to the defoamer flow rate (as a controlled process variable).

[0032] Therefore, an implementation that adds a data acquisition phase 120 to system 100 provides real-time measurement of at least the aforementioned process characteristics. One or more online sensors 122 are configured to provide substantially continuous control signals representing the process characteristics. The term "sensor" may include sensors, relays, and equivalent monitoring devices that may be provided to directly measure values ​​of process characteristics, or to measure values ​​appropriately derived from process characteristics that can be measured or calculated by them. Various conventional devices for the purpose of continuously sensing or calculating characteristics such as entrained air, washing machine speed, liquid-solid, and conductivity are well known in the art, and exemplary such sensors are considered to be fully compatible with the scope of systems and methods disclosed herein. As used herein, the term "online" generally refers to the use of devices, sensors, or corresponding elements located near the machine or associated process element and generating output signals corresponding to the desired process characteristics in real time, as opposed to manual or automated sample collection and "offline" analysis by one or more operators through visual observation or in a laboratory setting.

[0033] Each sensor can individually implement the corresponding output signal to be acquired, or in some embodiments, one or more individual sensors can provide corresponding output signals for calculations of multiple variables. Each sensor can be installed and configured individually, or the system can provide a modular housing including multiple sensors or sensing elements. Sensors or sensing elements can be permanently or portablely installed in specific locations during the production phase, or their positions can be dynamically adjusted to acquire data from multiple locations during operation.

[0034] One or more additional online sensors can provide basic continuous measurements of a variety of controlled process variables.

[0035] A user interface 124 is further provided and configured to display process information and / or enable the operator to input additional parameters and / or coefficients. For example, the operator may be able to selectively monitor process characteristics and process variables in real time, and may also select control parameters, such as threshold levels and / or optimal ranges, for one or more controlled process characteristics. Unless otherwise stated, the term "user interface" as used herein may include any input-output module relating to a controller and / or a hosted data server, including but not limited to: a fixed operator panel with key-controlled data entry, a touchscreen, buttons, or a dial pad; web portals, such as individual web pages or those that collectively define a hosted website; and mobile device applications, etc. Thus, an example of a user interface may be one remotely generated on a user computing device 150 and communicatively connected to a remote server 134 and / or a local controller 132.

[0036] As used herein, the term "continuous" does not require a specific degree of continuity, at least with respect to the disclosed measurements, but rather can generally describe a series of online measurements corresponding to: the physical and technological capabilities of the sensor, the physical and technological capabilities of the transmission medium, the physical and technological capabilities of the interface and / or controller configured to receive the sensor output signal, and / or the requirements of the associated control loop. For example, measurements may be performed and provided periodically at a rate lower than the maximum possible rate based on a control configuration that smooths the input value over time or conversely does not benefit from an increased input data frequency, and are still considered "continuous."

[0037] Online measurement data from various sensors 122 and input data from one or more users via a user interface are provided to the processing and control phase 130, the implementation of which is as follows: Figure 1 The diagram shows that controller 132 is included. Controller 132 may be a "local" controller configured to directly receive the aforementioned signals and perform prescribed data processing and control functions, while communicating independently with a remote server 134 (or a cloud-based computing network) via communication network 138. Generally, output signals may be provided from various sensors to DCS 136, which then transmits the output signals or measurement data derived therefrom to the controller. In some cases, the DCS itself may derive measurements of one or more process characteristics from other sensing values ​​and generate a representative output signal to the controller. In other cases, the controller may receive output signals directly from one or more online sensors and bypass the DCS. Those skilled in the art will appreciate that these and other potential configurations are within the scope of this disclosure unless specifically indicated otherwise.

[0038] In one embodiment (not shown), a conversion stage may be added for the purpose of converting raw signals from one or more online sensors 122 into signals compatible with the input requirements of the DCS 136 or controller 132. Alternatively or additionally, a conversion stage (or unit) may be provided to convert the raw signals from the DCS to meet the input requirements of the controller. The conversion stage may not only address input requirements but may also further provide data security between the one or more sensors and the DCS or controller as described above, and / or further between the DCS and the controller, and / or between the controller and the user computing device, for example, by selectively enabling access to signals between the respective devices through encryption, decryption, or other means.

[0039] As used herein with respect to data communication between two or more system components or otherwise between communication network interfaces associated with two or more system components, the term "communication network" may refer to any one or any combination of two or more of the following: a telecommunications network (whether wired, wireless, or cellular), a global network (e.g., the Internet), a local network, a network connection, an Internet Service Provider (ISP), and an intermediate communication interface. Any one or more recognized interface standards may be implemented with it, including but not limited to Bluetooth, RF, and Ethernet.

[0040] The controller 132 may be integrated into or otherwise operate in conjunction with the existing distributed control system 136 of the pulp mill assembly. For example, the controller 132 may typically generate control signals to various actuators via the DCS 136, or in some embodiments, the controller 132 may generate control signals directly to some or all of the various actuators associated with the controller's process variables. In one embodiment, the controller 132 may be configured to perform each of the otherwise distinguished local and distributed functions of the DCS 136.

[0041] Exemplary and non-limiting descriptions of online sensors, relays, and measurements associated with data acquisition phase 120 and pulp mill DCS 136 may include flow meters, valve positions, vacuum gauges, motor loads, level indicators, thermometers, pH meters, mass meters, refractometers, and entrained air monitors (e.g., ), tachometer, pressure gauge and interlock signal.

[0042] The controller 132 can be designed to communicate with and from the DCS 136 or other process measurement, monitoring, and control devices via Modbus RTU, TCP / IP, or discrete signals. Furthermore, the controller can wirelessly communicate with any Modbus RTU, TCP / IP, or discrete-capable device using a transmitter and receiver.

[0043] As used herein, terms such as “controller” or “computer” may refer to, be embodied in, or otherwise include within a machine such as a general-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof, which is designed and programmed to perform or cause to perform certain actions, functions, and algorithms described herein. A general-purpose processor may be a microprocessor, but alternatively, a processor may be a microcontroller, a state machine, or a combination thereof. A processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors incorporating a DSP core, or any other such configuration.

[0044] Depending on the implementation, certain actions, events, or functions of any of the algorithms described herein may be performed in a different order, or may be added, combined, or omitted entirely (e.g., not all described actions or events are necessary for the practice of the algorithm). Furthermore, in some implementations, actions or events may be performed concurrently (e.g., through multithreading, interrupt handling, or multiple processors or processor cores, or on other parallel architectures), rather than sequentially.

[0045] The steps of the computer-implemented methods, processes, or algorithms described in conjunction with the embodiments disclosed herein may be directly embodied in controller hardware, in software modules executed by a processor, or a combination of both. The software modules may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM, or any other form of computer-readable medium known in the art. Exemplary computer-readable media may be coupled to a processor, enabling the processor to read information from and write information to the memory / storage medium. Alternatively, the medium may be integrated into the processor. The processor and medium may reside in an ASIC. The ASIC may reside in a user terminal. Alternatively, the processor and medium may reside as discrete components in the user terminal.

[0046] In one implementation, a local controller 132 and / or a remote server 134 from the data processing and control phase 130 may be communicatively connected to a proprietary cloud-based data storage. The data storage may be configured, for example, to acquire, process, and collect / store data for purposes such as: developing correlations over time, improving existing linear regression or other related iterative algorithms, etc. The controller 132 may be configured to include certain correlations, equations, and / or algorithms in the local data storage while continuously or periodically transmitting (transmitting) relevant data to the remote server, and, for example, periodically retrieving any changes to the correlations, equations, and / or algorithms, which may be determined over time via, for example, machine learning, with additional input data.

[0047] In some implementations, the system's remote data capabilities allow machine learning concepts to be applied to the slurry washing system to enhance control over time. Those skilled in the art of slurry washing can understand or otherwise determine the values ​​of many other variables in the control algorithm. The following is a non-exhaustive list of inputs that may be included in system modeling: mat thickness measurement (using tools to measure thickness, rather than assuming thickness from input calculations); mat consistency measurement (using, for example, near-infrared light or equivalent devices to measure mat consistency in real time); stock temperature; filtrate temperature; spray temperature; pH of stock and filtrate; residual alkali in the stock leaving the digester; conductivity in the washer tank and / or filtrate; filtrate tank level; dropleg vacuum; kappa value measurement of stock; fiber species; refiner load; stock consistency; vacuum box level on a horizontal belt washer; vacuum on a horizontal belt washer; headbox pressure on direct displacement and compaction baffle washers; washer drum motor load; repulper motor load; and / or stock pump motor load.

[0048] Now for reference Figure 2 An exemplary method 200 for real-time predictive control of coarse pulp processing, which is substantially based on an embodiment of the system 100 described above, can now be described.

[0049] In a particular implementation, at the start of the process (step 201), the online sensor 122 continuously generates an output signal corresponding to the actual value of the corresponding process characteristic 114 (step 202). The output signal may be provided in its raw form for conversion, correction, and other uses to measure the actual value of the process characteristic, or the measurement result may be directly output from the sensor to the local controller 132.

[0050] The following steps can be performed via a local controller or a remote server. In one particular exemplary embodiment, the local controller performs all the data processing and control operations required for the normal operation of the process, but information may be passed to a remote server for cloud-based analytics and data processing. In one embodiment, the controller may be configured to access, or provide access to, a remote server for system updates, such as updating software or algorithm programming and configuration. However, the scope of the process disclosed herein is not structurally limited to this configuration unless otherwise specifically indicated.

[0051] In step 203 of this embodiment, the controller 132 presents the measured actual value (or the value provided directly from the corresponding online sensor 122) to the user interface 124 for display and monitoring purposes. Various exemplary screens of the user interface can be designed to confirm functions such as operation, configuration, analysis, and system alarms. The controller can also provide a layer of security by password-protecting these functions or by providing restricted access to users based on user classification. In step 204, the user interface allows the user to input information about the optimal range or threshold level of one or more controllable process variables.

[0052] In step 205 of this embodiment, controller 132 predicts the impact of subsequent control responses to process characteristics on corresponding process variables. In other words, the controller considers not only how to optimize the process characteristics themselves through control responses, but also how these control responses will affect certain process characteristics. The controller-based algorithm effectively models these interactions and correlations to perform optimization for each process variable and affected process characteristic, thereby minimizing variability and the possibility of negative feedback loops.

[0053] Based on the predictions made in the previous step, the controller continues in step 206 by dynamically setting target values ​​for the process characteristics. These target values ​​can then be implemented directly or otherwise transmitted to the plant's DCS or equivalent for local implementation. For example, in step 208, the controller can generate control signals to the DCS and / or any one or more system actuators based on the difference between the detected measured values ​​and the dynamically set target values ​​to adjust the corresponding process characteristics.

[0054] In one implementation, the controller may further transmit data (in its raw form or, for example, as data collected over time) to a remote server continuously or periodically in step 209 for analysis.

[0055] The foregoing embodiments of system 100 and method 200 can be further described by illustrative reference to a vacuum drum washer configuration as known in the art. It should be noted that the inventive aspects of the system and method are applicable to many alternative slurry washer embodiments, and the vacuum drum washer is referenced for illustrative purposes only.

[0056] Typically, in this configuration, the dirty pulp passes through the process from beginning to end and becomes cleaner because the dirty liquid is separated from the pulp using cleaning water provided at the "clean" end and moves backward to the "dirty" end of the process.

[0057] Water is applied to the vacuum drum washer via a washing spray. Pulp and dilution water enter the inlet tank and overflow into a barrel rotating within the perforated drum. Water is discharged through a screen on the drum (i.e., the vacuum-forming zone), falls from the drop leg, and creates a vacuum. This vacuum helps draw more pulp and liquid onto the drum. Liquid is discharged through the drop leg, but pulp remains on the drum surface. As the drum rotates, the pulp layer formed on the drum emerges from the barrel (i.e., the extraction zone) and is then struck by the washing spray, which places cleaner water onto the pulp layer (i.e., the replacement zone). Thus, in the pad layer, the dirty liquid is replaced by cleaner water from the spray. This liquid also flows down the drop leg through the drum. A discharge scraper is applied to scrape off the pulp layer (cleaned pulp), which enters the repulper for transfer to the next stage of washing. The cleaned washer surface is now submerged below the barrel level, and the process is repeated.

[0058] Defoamers are typically added to the pulp along with the paper before washing or to the cleaning spray. Defoamers help remove air, which allows liquid to drain more easily through the pulp layer. Air bubbles in the pulp layer block the liquid's passages and hinder drainage.

[0059] Conventional defoamer control typically involves manually setting the flow rate (e.g., by adjusting pump speed) or automatic adjustment based on operator-selected changes in production rate. However, such techniques are inherently passive, as operators can only respond to observed changes in the process and, even then, only by making imprecise adjustments. One problem with such passive control is that excessive use of defoamer drives up costs and can lead to silicone carryover problems (quality issues) in the final pulp. Furthermore, while defoamers eliminate the negative impact of entrained air on the discharge, they do not fundamentally address other variables affecting the discharge, including temperature, consistency, conductivity, alkali content, fiber freeness, and soap content.

[0060] Implementations of the predictive control systems and methods disclosed herein can be applied in this exemplary configuration to introduce a more comprehensive approach to this problem as one of many controllable process characteristics and process variables with relevant influences and correlations. An entrained air monitor (e.g., as previously noted) can be used. (Unit) to provide a continuous and real-time signal representing entrained air, wherein the control system can receive inputs of production rate and entrained air value and further determine, for example, the amount of defoamer added to the system by adjusting pump speed.

[0061] Depending on the location of the entrained air sensor and the defoamer addition point, the control can be applied as either feedback or feedforward control. Feedback control is implemented if the defoamer is added before the point where air can be released and the sensor is after that point, where the system dynamically provides a setpoint for entrained air and then adjusts the defoamer to maintain that setpoint. Feedforward control is implemented if the sensor is located between the defoamer addition point and the point where air can be released, but before air can be released, because the air has nowhere to go before the sensor reads it. In this case, the system can predict the amount of defoamer to add based on the amount of entrained air and the production rate in the system. The system can adjust the initial assumptions over time to provide the best possible control of the defoamer based on these two predictive factors (production rate variation and entrained air level).

[0062] One advantage of this technology is that the amount of defoamer needed can be controlled by adjusting the key variable that the system can control (i.e., how much entrained air is in the system). This is proactive because an increase in entrained air causes the drum level to rise or the washer to accelerate. Before Increase the amount of defoamer and conversely, decrease it when it is no longer needed. This preferably optimizes the amount of defoamer for any target level of entrained air and further eliminates the variability of entrained air from the process. Reduced process variability allows the entire system to operate more smoothly because other loops can be better fine-tuned, and the system can identify and eliminate other sources of variability. In other words, the washing machine speed is now based on other factors such as consistency fluctuations, conductivity variations, etc.

[0063] If this technology is implemented solely to maintain typical entrained air levels, defoamer usage can be substantially reduced. However, by actively reducing entrained air levels, the system can be able to run approximately the same amount of defoamer and use additional drainage to optimize the process, providing significant value to pulp mill operators. For example, by being proactive, the system can maintain bucket levels and washer speeds at more consistent settings, allowing for process optimization rather than constantly chasing the effects of entrained air. This can lead to savings in a variety of areas, including, for example, one or more of the following: increased productivity, cleaner pulp, lower bleaching costs in bleaching plants, lower chemical costs in unbleached plants, lower energy costs, less water used to wash pulp (which must evaporate at a certain cost), and reduced defoamer usage.

[0064] Therefore, the prediction system, as disclosed in the exemplary embodiments herein, measures entrained air as a first process characteristic (which is directly affected by changes in the defoamer flow rate as a first process variable), and further predicts the corresponding changes in drum level and washing machine speed, and adjusts the defoamer before these variables respond, thereby preventing them from changing.

[0065] The washing machine speed is typically adjusted automatically or manually by the operator to control the drum level and thus prevent overflow. Using an automatic control example, the washing machine speed increases as the drum level rises and decreases as the drum level falls. A setpoint can be entered for the drum level, and simple PID control can be implemented for the washing machine speed.

[0066] When the washing machine is running "too fast," the operator knows the tub could overflow at any moment, and when the washing machine is running "slow," the speed has sufficient capacity to respond to a sudden rise in the tub level. The tub level can rise due to, for example, changes in production rate or a lack of drainage. Therefore, operators may typically prefer to keep the washing machine running slowly, allowing sufficient margin for error before the tub overflows.

[0067] There are many techniques for keeping washing machines running slowly, including increasing defoamer, lowering the drum dilution valve position, reducing the spray flow rate on the washing machine, and / or slowing down the production rate. However, these techniques focus solely on controlling the drum to prevent overflow and do not properly consider the optimal washing efficiency for the washing machine, thus failing to properly consider the system as a whole. For example, drum dilution improves washing efficiency without using more water (which needs to be evaporated), while reducing drum dilution reduces the washing efficiency of the washing machine. After defoamer has been increased to slow the washing machine, it should preferably be reduced again once the drum level has dropped, but operators often choose to retain the increased defoamer rate out of caution. Reducing the spray flow rate ultimately leads to reduced pulp cleanliness. Reducing the production rate obviously reduces the system's output.

[0068] In one implementation, entrained air can be further implemented as part of the control scheme. As the entrained air rises, the tub level will increase, at which point the washing machine speed will increase, tub dilution will decrease, and then the defoamer will increase. The control algorithm is configured to predict and anticipate these events, proactively moving the defoamer by observing changes in entrained air, accelerating the response of the entire control scheme, and ultimately providing better controlled washing machine speed and washing efficiency.

[0069] Another exemplary control scheme relating to weak black liquor solids and spray flow control can now be described based on the predictive control system and method disclosed herein.

[0070] In a typical example, the operator measures the conductivity of the washed pulp exiting the final stage washer and determines whether to increase or decrease the spray water flow rate setpoint. In another typical example, the controlled method differs in that an online conductivity probe is present on the filtrate falling into the final stage filtrate tank, and a conductivity setpoint is established to control the spray flow rate.

[0071] Regardless of the method used in the plant, the main challenge here is balancing the final conductivity (the cleanliness of the pulp) with the amount of weak black liquor solids (how much water) entering the recycling zone. More water means cleaner pulp, but more water needs to be evaporated during collection. Less water means less water evaporation, but results in dirtier pulp.

[0072] These conventional methods have several drawbacks, the first being the undesirable lag time in control response. The filtrate tank below the washing machine is large, and it can take several hours for the tank to reach equilibrium as the pulp transitions from the blow tank to the inlet system. Therefore, for applications including final-stage conductivity measurements, a significant amount of filtrate needs to be delivered to move the process. Regardless of which conventional control scheme is used, they often increase variability by tracking process fluctuations.

[0073] Another drawback involves soap solubility. Several factors influence the solubility profile, the most significant being solids (concentration), temperature, and residual alkali. Once soap separates, it cannot return to the solution. The soap entrains air in bubbles that impede drainage, and these bubbles are unaffected by defoamers. Therefore, soap separation typically leads to poor drainage and high defoamer dosages, as operators are unable to successfully add defoamer in an attempt to repair the drainage, resulting in dirty pulp and / or low black liquor solids.

[0074] By controlling the conductivity at the end of the process, the plant is vulnerable to soap separation in the early stages of washing, as the solids level will fluctuate wildly based on spray flow variations made in response to variability introduced in the final stages and the pulp. Conductivity control cannot manage soap separation. Therefore, the safest option is to run with low solids, thus minimizing the frequency with which they cross the solubility line.

[0075] Another drawback arises when responding to conductivity measurements at the end of the process, as some of the change in the final conductivity measurement is based on dilution. When the operator adds more spray water to clean the pulp, the conductivity immediately decreases, not because of better washing, but because the solids content at the measurement point has been diluted. This gives the false sense of “control” over conductivity when faced with sample dilution. Ultimately, this leads to the misconception that conductivity is controllable, when in reality the amount of material carried forward during the process remains variable.

[0076] According to an embodiment of the predictive control system disclosed herein, the liquid solids leaving the first-stage filtrate tank are measured as a third process characteristic, and the spray flow rate at the corresponding first-stage washer is controllably adjusted as a third process variable. The aforementioned liquid solids measurement can be provided, for example, via a solids meter (e.g., a Coriolis mass flow meter, or a refractometer) on the first-stage filtrate.

[0077] In the exemplary embodiments disclosed herein, the system obtains comprehensive data points from the entire process to model washing efficiency. In other words, the controller may receive inputs from (or corresponding to) entrained air sensors, washing machine speed, defoamer dosage, conductivity, mass meter, production rate, drum dilution flow rate, filtrate tank level, residual alkali, and / or temperature, and predictively model the final conductivity against the parameters introduced in the first stage of the washing machine.

[0078] Implementing this predictive modeling allows for continuous adjustment of the solid target in weak black liquor to further maintain control over conductivity.

[0079] As an example, an initial target of 15.5% solids might be provided for a given plant to prevent it from exceeding the soap solubility limit. This figure can be adjusted seasonally (e.g., due to temperature variations from winter to summer), but is generally run as high as possible without allowing the soap solubility limit to be crossed.

[0080] In the predictive system disclosed herein, additional data from the plant can be used to continuously predict soap solubility limits. This allows the system to maintain appropriate solids levels based on temperature variations and residual alkali entering the washing machine. In this case, there is no need to manually change the setpoint several times a year, as it is now automatically adjusted based on temperature changes. This prevents accidental deviations typically caused in conventional applications due to temperature fluctuations or variations in the cooking process (residual alkali).

[0081] Furthermore, all the aforementioned variables can be used to force a reduction in the solids target when needed to help maintain conductivity in more stringent controls. For plants that are more concerned with the impact of conductivity variability on their process (final pulp cleanliness) than with maintaining liquid solids for recycling, the system can dynamically reduce the solids setpoint and better maintain conductivity at the far end to adapt to changes in washing efficiency within the system. Even if, for illustration, an upper limit is assumed on solids based on, for example, soap solubility, the system can adjust the solids setpoint to maintain the best possible final conductivity.

[0082] In short, depending on the individual stages operating with consistent efficiency, essentially optimal spray flow / solids control can be provided by systems and methods as disclosed herein. In exemplary and non-limiting embodiments, all three of the foregoing schemes can work together, where consistent execution across the individual washers allows for a better correlation of how solids variations will affect the final conductivity of the washer line. In other words, solids control, along with other controls, can perform much better. Effective process modeling and prediction become possible by fine-tuning the individual loops, and particularly in light of the measurement and control of entrained air, compared to what is possible in conventional plants due to excessive variability.

[0083] The conditional language used herein, such as in particular “may,” “possibly,” “possibly,” and “for example,” unless otherwise specifically stated or otherwise understood in the context in which they are used, is generally intended to convey that certain implementations include certain features, elements, and / or states that are not included in other implementations. Therefore, such conditional language is not generally intended to imply that features, elements, and / or states are necessary in any way for one or more implementations, or that one or more implementations necessarily include logic for determining whether to include or perform such features, elements, and / or states in any particular implementation, with or without author input or prompting.

[0084] The foregoing detailed description has been provided for purposes of illustration and description. Therefore, although specific embodiments of the new and useful invention have been described, these references are not intended to be construed as limiting the scope of the invention, except as set forth in the following claims.

Claims

1. A computer-implemented method for predictive control of brown stock handling at a pulp mill, the method comprising: continuously measuring actual values of a first process characteristic via a first online sensor, the first process characteristic being directly influenced by adjustments to at least a first process variable; continuously measuring actual values of a second process characteristic via a second online sensor, the second process characteristic being directly influenced by adjustments to at least a second process variable, wherein the second process variable is at least indirectly influenced by adjustments to the first process variable; dynamically setting target values for the first and second process characteristics based on predicted influences of corresponding control responses to at least the first and second process variables, respectively; and generating control signals to one or more actuators associated with at least the first and second process variables based on detected differences between the corresponding actual values and target values, wherein the first sensor is an air-entrainment sensor configured to generate output signals representative of an air-entrainment level as the first process characteristic, and the first process variable corresponds to a defoamer flow rate; and the second sensor is a washer speed sensor configured to generate output signals representative of a washer speed as the second process characteristic, and the second process variable corresponds to a dilution rate of a vat, or the second sensor is a flow meter configured to generate output signals representative of a dilution flow rate to a washer as the second process characteristic, and the second process variable corresponds to a defoamer pump speed.

2. The computer-implemented method of claim 1, further characterized by: each of the corresponding first and second process variables having a corresponding optimal range or threshold level; and further dynamically setting target values for the first and second process characteristics based on the optimal range or threshold level and predicted influences of corresponding control responses to at least the first and second process variables, respectively.

3. The computer-implemented method of one of claims 1 or 2, further characterized by providing acquisition signals or measurement data corresponding to actual values of at least the first and second process characteristics for display via a user interface.

4. The computer-implemented method of claim 3, further comprising enabling a user to specify one or more of the optimal ranges or threshold levels corresponding to the first and second process variables, respectively, via the user interface and a communication network connected thereto.

5. The computer-implemented method of any one of claims 1 to 2, further characterized by: the plurality of online sensors further comprising a third online sensor configured to generate output signals representative of actual values of a third process characteristic, the third process characteristic being directly influenced by adjustments to at least a third process variable, the third process variable being at least indirectly influenced by adjustments to the first and second process variables, dynamically setting target values for the third process characteristic based on predicted influences of corresponding control responses to at least the first, second, and third process variables, and generating control signals to a third actuator based on detected differences between actual values of the third process characteristic and target values.

6. The computer-implemented method of claim 5, further characterized by: the third sensor is a flow meter configured to generate output signals representative of a flow rate of a dilution stream as the third process characteristic, and the third process variable corresponds to a dilution pump speed.

7. The computer-implemented method of claim 5, further characterized by: the third sensor is a flow meter configured to generate output signals representative of a flow rate of a defoamer stream as the third process characteristic, and the third process variable corresponds to a defoamer pump speed.

8. The computer-implemented method of claim 5, further characterized by: the third sensor is a flow meter configured to generate output signals representative of a flow rate of a white water stream as the third process characteristic, and the third process variable corresponds to a white water pump speed.

9. The computer-implemented method of claim 5, further characterized by: the third sensor is a flow meter configured to generate output signals representative of a flow rate of a brown stock stream as the third process characteristic, and the third process variable corresponds to a brown stock pump speed.

6. The computer-implemented method according to claim 5, further characterized by: the first sensor being an entrainment air sensor configured to generate an output signal representative of an entrainment air level as the first process characteristic, the first process variable corresponding to a defoamer flow rate, the second sensor being a washer speed sensor configured to generate an output signal representative of a washer speed as the second process characteristic, the second process variable corresponding to a dilution rate of the vat, the third sensor being a liquor solids meter configured to generate an output signal representative of a liquor solids level as the third process characteristic, and the third process variable corresponding to a spray flow rate.

7. The computer-implemented method according to claim 5, further characterized by: the pulp mill further comprising a plurality of cascading filtrate tanks and a corresponding plurality of washers and spray flows, the liquor solids meter being provided in association with a first filtrate tank of the plurality of cascading filtrate tanks, and generating, based on a detected difference between the actual value and the target value of the third process characteristic, a control signal to the third actuator associated with the spray flow rate of the first washer.

8. The computer-implemented method according to claim 7, further characterized by: the pulp mill further comprising a conductivity sensor for measuring an actual conductivity value, wherein the actual conductivity value is influenced by each of a plurality of process characteristics including the first, second and third process characteristics, and a plurality of process variables including the first, second and third process variables.

9. The computer-implemented method according to claim 8, further characterized by: dynamically setting target values for the first, second and third process characteristics further in view of the predicted influence of the corresponding control responses relative to the optimal conductivity value; and generating control signals to the first, second and third actuators further in view of the dynamically set target values.

10. A system for predictive control of brown stock treatment at a pulp mill, the system comprising: a plurality of online sensors configured to generate output signals representative of actual values of respective process characteristics, the plurality of online sensors comprising: a first online sensor configured to generate an output signal representative of an actual value of a first process characteristic, the first process characteristic being directly influenced by adjustments to at least a first process variable, and a second online sensor configured to generate an output signal representative of an actual value of a second process characteristic, the second process characteristic being directly influenced by adjustments to at least a second process variable, wherein the second process variable is at least indirectly influenced by adjustments to the first process variable; at least first and second actuators configured to adjust actual values of the first and second process variables, respectively; and a controller configured to instruct execution of the method according to any one of claims 1 to 9.

Images